diff options
| author | Boris Kolpackov <boris@codesynthesis.com> | 2014-04-27 12:32:17 +0200 |
|---|---|---|
| committer | Boris Kolpackov <boris@codesynthesis.com> | 2014-04-27 12:32:17 +0200 |
| commit | 2c57a439676481590179718091f7404eb2e03ad7 (patch) | |
| tree | a008f641f62fc995c9f93a5b37ee5f329bbaa525 | |
| parent | 78e1a4d4a0a6022c13f251fbe23a20b2262830b2 (diff) | |
Initial source code import
50 files changed, 19203 insertions, 0 deletions
@@ -0,0 +1,31 @@ +Summary: Everything is licensed under the MIT License (text below). + +Code found in the xml/details/expat/ directory is distributed under +the MIT License (see the xml/details/expat/LICENSE file for copyright +information). + +Code found in the xml/details/genx/ directory is distributed under +the MIT License (see the xml/details/genx/LICENSE file for copyright +information). + +The rest is Copyright (c) 2013-2014 Code Synthesis Tools CC and is +distributed under the MIT License: + +Permission is hereby granted, free of charge, to any person obtaining +a copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to +permit persons to whom the Software is furnished to do so, subject to +the following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. diff --git a/build/bootstrap.make b/build/bootstrap.make new file mode 100644 index 0000000..c09afd8 --- /dev/null +++ b/build/bootstrap.make @@ -0,0 +1,70 @@ +# file : build/bootstrap.make +# copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +# license : MIT; see accompanying LICENSE file + +project_name := libstudxml + +# First try to include the bundled bootstrap.make if it exist. If that +# fails, let make search for the external bootstrap.make. +# +build := build-0.3 + +-include $(dir $(lastword $(MAKEFILE_LIST)))../../$(build)/bootstrap.make + +ifeq ($(patsubst %build/bootstrap.make,,$(lastword $(MAKEFILE_LIST))),) +include $(build)/bootstrap.make +endif + +# Aliases +# +.PHONY: $(out_base)/ \ + $(out_base)/.test \ + $(out_base)/.dist \ + $(out_base)/.clean + +ifdef %interactive% + +.PHONY: test dist clean + +test: $(out_base)/.test +dist: $(out_base)/.dist +clean: $(out_base)/.clean + +ifneq ($(filter $(.DEFAULT_GOAL),test dist clean),) +.DEFAULT_GOAL := +endif + +endif + +# Make sure the distribution prefix is set if the goal is dist. +# +ifneq ($(filter $(MAKECMDGOALS),dist),) +ifeq ($(dist_prefix),) +$(error dist_prefix is not set) +endif +endif + +# If we don't have dependency auto-generation then we need to manually +# make sure that generated files are generated before C++ file are +# compiler. To do this we make the object files ($2) depend in order- +# only on generated files ($3). +# +ifeq ($(cxx_id),generic) + +define include-dep +$(if $2,$(eval $2: | $3)) +endef + +else + +define include-dep +$(call -include,$1) +endef + +endif + +# Don't include dependency info for certain targets. +# +ifneq ($(filter $(MAKECMDGOALS),clean disfigure dist),) +include-dep = +endif diff --git a/build/export/libstudxml/stub.make b/build/export/libstudxml/stub.make new file mode 100644 index 0000000..0ddde78 --- /dev/null +++ b/build/export/libstudxml/stub.make @@ -0,0 +1,9 @@ +# file : build/export/libstudxml/stub.make +# copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +# license : MIT; see accompanying LICENSE file + +$(call include-once,$(src_root)/xml/makefile,$(out_root)) + +$(call export,\ + l: $(out_root)/xml/studxml.l,\ + cpp-options: $(out_root)/xml/studxml.l.cpp-options) diff --git a/build/import/libstudxml/LICENSE b/build/import/libstudxml/LICENSE new file mode 120000 index 0000000..5853aae --- /dev/null +++ b/build/import/libstudxml/LICENSE @@ -0,0 +1 @@ +../../../LICENSE
\ No newline at end of file diff --git a/build/import/libstudxml/configuration-rules.make b/build/import/libstudxml/configuration-rules.make new file mode 100644 index 0000000..423df97 --- /dev/null +++ b/build/import/libstudxml/configuration-rules.make @@ -0,0 +1,14 @@ +# file : build/import/libstudxml/configuration-rules.make +# copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +# license : MIT; see accompanying LICENSE file + +$(dcf_root)/import/libstudxml/configuration-dynamic.make: | $(dcf_root)/import/libstudxml/. + $(call message,,$(scf_root)/import/libstudxml/configure $@) + +ifndef %foreign% + +$(dcf_root)/.disfigure:: + $(call message,rm $(dcf_root)/import/libstudxml/configuration-dynamic.make,\ +rm -f $(dcf_root)/import/libstudxml/configuration-dynamic.make) + +endif diff --git a/build/import/libstudxml/configure b/build/import/libstudxml/configure new file mode 100755 index 0000000..15ddb89 --- /dev/null +++ b/build/import/libstudxml/configure @@ -0,0 +1,52 @@ +#! /usr/bin/env bash + +# file : build/import/libstudxml/configure +# copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +# license : MIT; see accompanying LICENSE file + +# $1 - out file +# +# bld_root - build root +# project_name - project name +# + +source $bld_root/dialog.bash + +$echo +$echo "Configuring external dependency on 'libstudxml' for '$project_name'." +$echo + +$echo +$echo "Would you like to configure dependency on the installed " +$echo "version of 'libstudxml' as opposed to the development build?" +$echo + +installed=`read_y_n y` + +path= + +if [ "$installed" = "n" ]; then + +$echo +$echo "Please enter the src_root for 'libstudxml'." +$echo + +src_root=`read_path --directory --exist` + +$echo +$echo "Please enter the out_root for 'libstudxml'." +$echo + +out_root=`read_path --directory $src_root` + +fi + +echo libstudxml_installed := $installed >$1 + +if [ "$installed" = "n" ]; then + +echo src_root := $src_root >>$1 +echo scf_root := \$\(src_root\)/build >>$1 +echo out_root := $out_root >>$1 + +fi diff --git a/build/import/libstudxml/stub.make b/build/import/libstudxml/stub.make new file mode 100644 index 0000000..2f73a97 --- /dev/null +++ b/build/import/libstudxml/stub.make @@ -0,0 +1,29 @@ +# file : build/import/libstudxml/stub.make +# copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +# license : MIT; see accompanying LICENSE file + +$(call include-once,$(scf_root)/import/libstudxml/configuration-rules.make,$(dcf_root)) + +libstudxml_installed := + +$(call -include,$(dcf_root)/import/libstudxml/configuration-dynamic.make) + +ifdef libstudxml_installed + +ifeq ($(libstudxml_installed),y) + +$(call export,l: -lstudxml,cpp-options: ) + +else + +# Include export stub. +# +$(call include,$(scf_root)/export/libstudxml/stub.make) + +endif + +else + +.NOTPARALLEL: + +endif diff --git a/xml/details/config-vc.h b/xml/details/config-vc.h new file mode 100644 index 0000000..229e52e --- /dev/null +++ b/xml/details/config-vc.h @@ -0,0 +1,13 @@ +/* file : xml/details/config-vc.h + * copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC + * license : MIT; see accompanying LICENSE file + */ + +/* Configuration file for Windows/VC++. */ + +#ifndef XML_DETAILS_CONFIG_VC_H +#define XML_DETAILS_CONFIG_VC_H + +#define LIBSTUDXML_BYTEORDER 1234 + +#endif /* XML_DETAILS_CONFIG_VC_H */ diff --git a/xml/details/config.h.in b/xml/details/config.h.in new file mode 100644 index 0000000..7df02a7 --- /dev/null +++ b/xml/details/config.h.in @@ -0,0 +1,15 @@ +/* file : xml/details/config.h.in + * copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC + * license : MIT; see accompanying LICENSE file + */ + +/* This file is automatically processed by configure. */ + +#ifndef XML_DETAILS_CONFIG_H +#define XML_DETAILS_CONFIG_H + +#undef LIBSTUDXML_STATIC_LIB +#undef LIBSTUDXML_EXTERNAL_EXPAT +#undef LIBSTUDXML_BYTEORDER + +#endif /* XML_DETAILS_CONFIG_H */ diff --git a/xml/details/config.hxx b/xml/details/config.hxx new file mode 100644 index 0000000..d466d89 --- /dev/null +++ b/xml/details/config.hxx @@ -0,0 +1,14 @@ +// file : xml/details/config.hxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#ifndef XML_DETAILS_CONFIG_HXX +#define XML_DETAILS_CONFIG_HXX + +#ifdef _MSC_VER +# include <xml/details/config-vc.h> +#else +# include <xml/details/config.h> +#endif + +#endif // XML_DETAILS_CONFIG_HXX diff --git a/xml/details/expat/LICENSE b/xml/details/expat/LICENSE new file mode 100644 index 0000000..dcb4506 --- /dev/null +++ b/xml/details/expat/LICENSE @@ -0,0 +1,22 @@ +Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd + and Clark Cooper +Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006 Expat maintainers. + +Permission is hereby granted, free of charge, to any person obtaining +a copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to +permit persons to whom the Software is furnished to do so, subject to +the following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. diff --git a/xml/details/expat/README b/xml/details/expat/README new file mode 100644 index 0000000..3b1ced2 --- /dev/null +++ b/xml/details/expat/README @@ -0,0 +1,2 @@ +This directory contains the Expat XML parser library for internal +use by libstudxml. diff --git a/xml/details/expat/ascii.h b/xml/details/expat/ascii.h new file mode 100644 index 0000000..d10530b --- /dev/null +++ b/xml/details/expat/ascii.h @@ -0,0 +1,92 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#define ASCII_A 0x41 +#define ASCII_B 0x42 +#define ASCII_C 0x43 +#define ASCII_D 0x44 +#define ASCII_E 0x45 +#define ASCII_F 0x46 +#define ASCII_G 0x47 +#define ASCII_H 0x48 +#define ASCII_I 0x49 +#define ASCII_J 0x4A +#define ASCII_K 0x4B +#define ASCII_L 0x4C +#define ASCII_M 0x4D +#define ASCII_N 0x4E +#define ASCII_O 0x4F +#define ASCII_P 0x50 +#define ASCII_Q 0x51 +#define ASCII_R 0x52 +#define ASCII_S 0x53 +#define ASCII_T 0x54 +#define ASCII_U 0x55 +#define ASCII_V 0x56 +#define ASCII_W 0x57 +#define ASCII_X 0x58 +#define ASCII_Y 0x59 +#define ASCII_Z 0x5A + +#define ASCII_a 0x61 +#define ASCII_b 0x62 +#define ASCII_c 0x63 +#define ASCII_d 0x64 +#define ASCII_e 0x65 +#define ASCII_f 0x66 +#define ASCII_g 0x67 +#define ASCII_h 0x68 +#define ASCII_i 0x69 +#define ASCII_j 0x6A +#define ASCII_k 0x6B +#define ASCII_l 0x6C +#define ASCII_m 0x6D +#define ASCII_n 0x6E +#define ASCII_o 0x6F +#define ASCII_p 0x70 +#define ASCII_q 0x71 +#define ASCII_r 0x72 +#define ASCII_s 0x73 +#define ASCII_t 0x74 +#define ASCII_u 0x75 +#define ASCII_v 0x76 +#define ASCII_w 0x77 +#define ASCII_x 0x78 +#define ASCII_y 0x79 +#define ASCII_z 0x7A + +#define ASCII_0 0x30 +#define ASCII_1 0x31 +#define ASCII_2 0x32 +#define ASCII_3 0x33 +#define ASCII_4 0x34 +#define ASCII_5 0x35 +#define ASCII_6 0x36 +#define ASCII_7 0x37 +#define ASCII_8 0x38 +#define ASCII_9 0x39 + +#define ASCII_TAB 0x09 +#define ASCII_SPACE 0x20 +#define ASCII_EXCL 0x21 +#define ASCII_QUOT 0x22 +#define ASCII_AMP 0x26 +#define ASCII_APOS 0x27 +#define ASCII_MINUS 0x2D +#define ASCII_PERIOD 0x2E +#define ASCII_COLON 0x3A +#define ASCII_SEMI 0x3B +#define ASCII_LT 0x3C +#define ASCII_EQUALS 0x3D +#define ASCII_GT 0x3E +#define ASCII_LSQB 0x5B +#define ASCII_RSQB 0x5D +#define ASCII_UNDERSCORE 0x5F +#define ASCII_LPAREN 0x28 +#define ASCII_RPAREN 0x29 +#define ASCII_FF 0x0C +#define ASCII_SLASH 0x2F +#define ASCII_HASH 0x23 +#define ASCII_PIPE 0x7C +#define ASCII_COMMA 0x2C diff --git a/xml/details/expat/asciitab.h b/xml/details/expat/asciitab.h new file mode 100644 index 0000000..79a15c2 --- /dev/null +++ b/xml/details/expat/asciitab.h @@ -0,0 +1,36 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +/* 0x00 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x04 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x08 */ BT_NONXML, BT_S, BT_LF, BT_NONXML, +/* 0x0C */ BT_NONXML, BT_CR, BT_NONXML, BT_NONXML, +/* 0x10 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x14 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x18 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x1C */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x20 */ BT_S, BT_EXCL, BT_QUOT, BT_NUM, +/* 0x24 */ BT_OTHER, BT_PERCNT, BT_AMP, BT_APOS, +/* 0x28 */ BT_LPAR, BT_RPAR, BT_AST, BT_PLUS, +/* 0x2C */ BT_COMMA, BT_MINUS, BT_NAME, BT_SOL, +/* 0x30 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT, +/* 0x34 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT, +/* 0x38 */ BT_DIGIT, BT_DIGIT, BT_COLON, BT_SEMI, +/* 0x3C */ BT_LT, BT_EQUALS, BT_GT, BT_QUEST, +/* 0x40 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX, +/* 0x44 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT, +/* 0x48 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x4C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x50 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x54 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x58 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_LSQB, +/* 0x5C */ BT_OTHER, BT_RSQB, BT_OTHER, BT_NMSTRT, +/* 0x60 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX, +/* 0x64 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT, +/* 0x68 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x6C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x70 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x74 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x78 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER, +/* 0x7C */ BT_VERBAR, BT_OTHER, BT_OTHER, BT_OTHER, diff --git a/xml/details/expat/config.h b/xml/details/expat/config.h new file mode 100644 index 0000000..b301012 --- /dev/null +++ b/xml/details/expat/config.h @@ -0,0 +1,35 @@ +#ifndef EXPAT_CONFIG_H +#define EXPAT_CONFIG_H + +#ifdef _MSC_VER +# include <xml/details/config-vc.h> +#else +# include <xml/details/config.h> +#endif + +#define BYTEORDER LIBSTUDXML_BYTEORDER + +#define XML_NS 1 +#define XML_DTD 1 +#define XML_CONTEXT_BYTES 1024 + +#define UNUSED(x) (void)x; + +#ifdef _WIN32 +/* Windows + * + */ +#define WIN32_LEAN_AND_MEAN +#include <windows.h> +#undef WIN32_LEAN_AND_MEAN + +#define HAVE_MEMMOVE 1 + +#else +/* POSIX + * + */ +#define HAVE_MEMMOVE 1 +#endif + +#endif /* EXPAT_CONFIG_H */ diff --git a/xml/details/expat/expat.h b/xml/details/expat/expat.h new file mode 100644 index 0000000..77baa3a --- /dev/null +++ b/xml/details/expat/expat.h @@ -0,0 +1,1047 @@ +/* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#ifndef Expat_INCLUDED +#define Expat_INCLUDED 1 + +#ifdef __VMS +/* 0 1 2 3 0 1 2 3 + 1234567890123456789012345678901 1234567890123456789012345678901 */ +#define XML_SetProcessingInstructionHandler XML_SetProcessingInstrHandler +#define XML_SetUnparsedEntityDeclHandler XML_SetUnparsedEntDeclHandler +#define XML_SetStartNamespaceDeclHandler XML_SetStartNamespcDeclHandler +#define XML_SetExternalEntityRefHandlerArg XML_SetExternalEntRefHandlerArg +#endif + +#include <stdlib.h> +#include <xml/details/expat/expat_external.h> + +#ifdef __cplusplus +extern "C" { +#endif + +struct XML_ParserStruct; +typedef struct XML_ParserStruct *XML_Parser; + +/* Should this be defined using stdbool.h when C99 is available? */ +typedef unsigned char XML_Bool; +#define XML_TRUE ((XML_Bool) 1) +#define XML_FALSE ((XML_Bool) 0) + +/* The XML_Status enum gives the possible return values for several + API functions. The preprocessor #defines are included so this + stanza can be added to code that still needs to support older + versions of Expat 1.95.x: + + #ifndef XML_STATUS_OK + #define XML_STATUS_OK 1 + #define XML_STATUS_ERROR 0 + #endif + + Otherwise, the #define hackery is quite ugly and would have been + dropped. +*/ +enum XML_Status { + XML_STATUS_ERROR = 0, +#define XML_STATUS_ERROR XML_STATUS_ERROR + XML_STATUS_OK = 1, +#define XML_STATUS_OK XML_STATUS_OK + XML_STATUS_SUSPENDED = 2 +#define XML_STATUS_SUSPENDED XML_STATUS_SUSPENDED +}; + +enum XML_Error { + XML_ERROR_NONE, + XML_ERROR_NO_MEMORY, + XML_ERROR_SYNTAX, + XML_ERROR_NO_ELEMENTS, + XML_ERROR_INVALID_TOKEN, + XML_ERROR_UNCLOSED_TOKEN, + XML_ERROR_PARTIAL_CHAR, + XML_ERROR_TAG_MISMATCH, + XML_ERROR_DUPLICATE_ATTRIBUTE, + XML_ERROR_JUNK_AFTER_DOC_ELEMENT, + XML_ERROR_PARAM_ENTITY_REF, + XML_ERROR_UNDEFINED_ENTITY, + XML_ERROR_RECURSIVE_ENTITY_REF, + XML_ERROR_ASYNC_ENTITY, + XML_ERROR_BAD_CHAR_REF, + XML_ERROR_BINARY_ENTITY_REF, + XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF, + XML_ERROR_MISPLACED_XML_PI, + XML_ERROR_UNKNOWN_ENCODING, + XML_ERROR_INCORRECT_ENCODING, + XML_ERROR_UNCLOSED_CDATA_SECTION, + XML_ERROR_EXTERNAL_ENTITY_HANDLING, + XML_ERROR_NOT_STANDALONE, + XML_ERROR_UNEXPECTED_STATE, + XML_ERROR_ENTITY_DECLARED_IN_PE, + XML_ERROR_FEATURE_REQUIRES_XML_DTD, + XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING, + /* Added in 1.95.7. */ + XML_ERROR_UNBOUND_PREFIX, + /* Added in 1.95.8. */ + XML_ERROR_UNDECLARING_PREFIX, + XML_ERROR_INCOMPLETE_PE, + XML_ERROR_XML_DECL, + XML_ERROR_TEXT_DECL, + XML_ERROR_PUBLICID, + XML_ERROR_SUSPENDED, + XML_ERROR_NOT_SUSPENDED, + XML_ERROR_ABORTED, + XML_ERROR_FINISHED, + XML_ERROR_SUSPEND_PE, + /* Added in 2.0. */ + XML_ERROR_RESERVED_PREFIX_XML, + XML_ERROR_RESERVED_PREFIX_XMLNS, + XML_ERROR_RESERVED_NAMESPACE_URI +}; + +enum XML_Content_Type { + XML_CTYPE_EMPTY = 1, + XML_CTYPE_ANY, + XML_CTYPE_MIXED, + XML_CTYPE_NAME, + XML_CTYPE_CHOICE, + XML_CTYPE_SEQ +}; + +enum XML_Content_Quant { + XML_CQUANT_NONE, + XML_CQUANT_OPT, + XML_CQUANT_REP, + XML_CQUANT_PLUS +}; + +/* If type == XML_CTYPE_EMPTY or XML_CTYPE_ANY, then quant will be + XML_CQUANT_NONE, and the other fields will be zero or NULL. + If type == XML_CTYPE_MIXED, then quant will be NONE or REP and + numchildren will contain number of elements that may be mixed in + and children point to an array of XML_Content cells that will be + all of XML_CTYPE_NAME type with no quantification. + + If type == XML_CTYPE_NAME, then the name points to the name, and + the numchildren field will be zero and children will be NULL. The + quant fields indicates any quantifiers placed on the name. + + CHOICE and SEQ will have name NULL, the number of children in + numchildren and children will point, recursively, to an array + of XML_Content cells. + + The EMPTY, ANY, and MIXED types will only occur at top level. +*/ + +typedef struct XML_cp XML_Content; + +struct XML_cp { + enum XML_Content_Type type; + enum XML_Content_Quant quant; + XML_Char * name; + unsigned int numchildren; + XML_Content * children; +}; + + +/* This is called for an element declaration. See above for + description of the model argument. It's the caller's responsibility + to free model when finished with it. +*/ +typedef void (XMLCALL *XML_ElementDeclHandler) (void *userData, + const XML_Char *name, + XML_Content *model); + +XMLPARSEAPI(void) +XML_SetElementDeclHandler(XML_Parser parser, + XML_ElementDeclHandler eldecl); + +/* The Attlist declaration handler is called for *each* attribute. So + a single Attlist declaration with multiple attributes declared will + generate multiple calls to this handler. The "default" parameter + may be NULL in the case of the "#IMPLIED" or "#REQUIRED" + keyword. The "isrequired" parameter will be true and the default + value will be NULL in the case of "#REQUIRED". If "isrequired" is + true and default is non-NULL, then this is a "#FIXED" default. +*/ +typedef void (XMLCALL *XML_AttlistDeclHandler) ( + void *userData, + const XML_Char *elname, + const XML_Char *attname, + const XML_Char *att_type, + const XML_Char *dflt, + int isrequired); + +XMLPARSEAPI(void) +XML_SetAttlistDeclHandler(XML_Parser parser, + XML_AttlistDeclHandler attdecl); + +/* The XML declaration handler is called for *both* XML declarations + and text declarations. The way to distinguish is that the version + parameter will be NULL for text declarations. The encoding + parameter may be NULL for XML declarations. The standalone + parameter will be -1, 0, or 1 indicating respectively that there + was no standalone parameter in the declaration, that it was given + as no, or that it was given as yes. +*/ +typedef void (XMLCALL *XML_XmlDeclHandler) (void *userData, + const XML_Char *version, + const XML_Char *encoding, + int standalone); + +XMLPARSEAPI(void) +XML_SetXmlDeclHandler(XML_Parser parser, + XML_XmlDeclHandler xmldecl); + + +typedef struct { + void *(*malloc_fcn)(size_t size); + void *(*realloc_fcn)(void *ptr, size_t size); + void (*free_fcn)(void *ptr); +} XML_Memory_Handling_Suite; + +/* Constructs a new parser; encoding is the encoding specified by the + external protocol or NULL if there is none specified. +*/ +XMLPARSEAPI(XML_Parser) +XML_ParserCreate(const XML_Char *encoding); + +/* Constructs a new parser and namespace processor. Element type + names and attribute names that belong to a namespace will be + expanded; unprefixed attribute names are never expanded; unprefixed + element type names are expanded only if there is a default + namespace. The expanded name is the concatenation of the namespace + URI, the namespace separator character, and the local part of the + name. If the namespace separator is '\0' then the namespace URI + and the local part will be concatenated without any separator. + It is a programming error to use the separator '\0' with namespace + triplets (see XML_SetReturnNSTriplet). +*/ +XMLPARSEAPI(XML_Parser) +XML_ParserCreateNS(const XML_Char *encoding, XML_Char namespaceSeparator); + + +/* Constructs a new parser using the memory management suite referred to + by memsuite. If memsuite is NULL, then use the standard library memory + suite. If namespaceSeparator is non-NULL it creates a parser with + namespace processing as described above. The character pointed at + will serve as the namespace separator. + + All further memory operations used for the created parser will come from + the given suite. +*/ +XMLPARSEAPI(XML_Parser) +XML_ParserCreate_MM(const XML_Char *encoding, + const XML_Memory_Handling_Suite *memsuite, + const XML_Char *namespaceSeparator); + +/* Prepare a parser object to be re-used. This is particularly + valuable when memory allocation overhead is disproportionatly high, + such as when a large number of small documnents need to be parsed. + All handlers are cleared from the parser, except for the + unknownEncodingHandler. The parser's external state is re-initialized + except for the values of ns and ns_triplets. + + Added in Expat 1.95.3. +*/ +XMLPARSEAPI(XML_Bool) +XML_ParserReset(XML_Parser parser, const XML_Char *encoding); + +/* atts is array of name/value pairs, terminated by 0; + names and values are 0 terminated. +*/ +typedef void (XMLCALL *XML_StartElementHandler) (void *userData, + const XML_Char *name, + const XML_Char **atts); + +typedef void (XMLCALL *XML_EndElementHandler) (void *userData, + const XML_Char *name); + + +/* s is not 0 terminated. */ +typedef void (XMLCALL *XML_CharacterDataHandler) (void *userData, + const XML_Char *s, + int len); + +/* target and data are 0 terminated */ +typedef void (XMLCALL *XML_ProcessingInstructionHandler) ( + void *userData, + const XML_Char *target, + const XML_Char *data); + +/* data is 0 terminated */ +typedef void (XMLCALL *XML_CommentHandler) (void *userData, + const XML_Char *data); + +typedef void (XMLCALL *XML_StartCdataSectionHandler) (void *userData); +typedef void (XMLCALL *XML_EndCdataSectionHandler) (void *userData); + +/* This is called for any characters in the XML document for which + there is no applicable handler. This includes both characters that + are part of markup which is of a kind that is not reported + (comments, markup declarations), or characters that are part of a + construct which could be reported but for which no handler has been + supplied. The characters are passed exactly as they were in the XML + document except that they will be encoded in UTF-8 or UTF-16. + Line boundaries are not normalized. Note that a byte order mark + character is not passed to the default handler. There are no + guarantees about how characters are divided between calls to the + default handler: for example, a comment might be split between + multiple calls. +*/ +typedef void (XMLCALL *XML_DefaultHandler) (void *userData, + const XML_Char *s, + int len); + +/* This is called for the start of the DOCTYPE declaration, before + any DTD or internal subset is parsed. +*/ +typedef void (XMLCALL *XML_StartDoctypeDeclHandler) ( + void *userData, + const XML_Char *doctypeName, + const XML_Char *sysid, + const XML_Char *pubid, + int has_internal_subset); + +/* This is called for the start of the DOCTYPE declaration when the + closing > is encountered, but after processing any external + subset. +*/ +typedef void (XMLCALL *XML_EndDoctypeDeclHandler)(void *userData); + +/* This is called for entity declarations. The is_parameter_entity + argument will be non-zero if the entity is a parameter entity, zero + otherwise. + + For internal entities (<!ENTITY foo "bar">), value will + be non-NULL and systemId, publicID, and notationName will be NULL. + The value string is NOT nul-terminated; the length is provided in + the value_length argument. Since it is legal to have zero-length + values, do not use this argument to test for internal entities. + + For external entities, value will be NULL and systemId will be + non-NULL. The publicId argument will be NULL unless a public + identifier was provided. The notationName argument will have a + non-NULL value only for unparsed entity declarations. + + Note that is_parameter_entity can't be changed to XML_Bool, since + that would break binary compatibility. +*/ +typedef void (XMLCALL *XML_EntityDeclHandler) ( + void *userData, + const XML_Char *entityName, + int is_parameter_entity, + const XML_Char *value, + int value_length, + const XML_Char *base, + const XML_Char *systemId, + const XML_Char *publicId, + const XML_Char *notationName); + +XMLPARSEAPI(void) +XML_SetEntityDeclHandler(XML_Parser parser, + XML_EntityDeclHandler handler); + +/* OBSOLETE -- OBSOLETE -- OBSOLETE + This handler has been superceded by the EntityDeclHandler above. + It is provided here for backward compatibility. + + This is called for a declaration of an unparsed (NDATA) entity. + The base argument is whatever was set by XML_SetBase. The + entityName, systemId and notationName arguments will never be + NULL. The other arguments may be. +*/ +typedef void (XMLCALL *XML_UnparsedEntityDeclHandler) ( + void *userData, + const XML_Char *entityName, + const XML_Char *base, + const XML_Char *systemId, + const XML_Char *publicId, + const XML_Char *notationName); + +/* This is called for a declaration of notation. The base argument is + whatever was set by XML_SetBase. The notationName will never be + NULL. The other arguments can be. +*/ +typedef void (XMLCALL *XML_NotationDeclHandler) ( + void *userData, + const XML_Char *notationName, + const XML_Char *base, + const XML_Char *systemId, + const XML_Char *publicId); + +/* When namespace processing is enabled, these are called once for + each namespace declaration. The call to the start and end element + handlers occur between the calls to the start and end namespace + declaration handlers. For an xmlns attribute, prefix will be + NULL. For an xmlns="" attribute, uri will be NULL. +*/ +typedef void (XMLCALL *XML_StartNamespaceDeclHandler) ( + void *userData, + const XML_Char *prefix, + const XML_Char *uri); + +typedef void (XMLCALL *XML_EndNamespaceDeclHandler) ( + void *userData, + const XML_Char *prefix); + +/* This is called if the document is not standalone, that is, it has an + external subset or a reference to a parameter entity, but does not + have standalone="yes". If this handler returns XML_STATUS_ERROR, + then processing will not continue, and the parser will return a + XML_ERROR_NOT_STANDALONE error. + If parameter entity parsing is enabled, then in addition to the + conditions above this handler will only be called if the referenced + entity was actually read. +*/ +typedef int (XMLCALL *XML_NotStandaloneHandler) (void *userData); + +/* This is called for a reference to an external parsed general + entity. The referenced entity is not automatically parsed. The + application can parse it immediately or later using + XML_ExternalEntityParserCreate. + + The parser argument is the parser parsing the entity containing the + reference; it can be passed as the parser argument to + XML_ExternalEntityParserCreate. The systemId argument is the + system identifier as specified in the entity declaration; it will + not be NULL. + + The base argument is the system identifier that should be used as + the base for resolving systemId if systemId was relative; this is + set by XML_SetBase; it may be NULL. + + The publicId argument is the public identifier as specified in the + entity declaration, or NULL if none was specified; the whitespace + in the public identifier will have been normalized as required by + the XML spec. + + The context argument specifies the parsing context in the format + expected by the context argument to XML_ExternalEntityParserCreate; + context is valid only until the handler returns, so if the + referenced entity is to be parsed later, it must be copied. + context is NULL only when the entity is a parameter entity. + + The handler should return XML_STATUS_ERROR if processing should not + continue because of a fatal error in the handling of the external + entity. In this case the calling parser will return an + XML_ERROR_EXTERNAL_ENTITY_HANDLING error. + + Note that unlike other handlers the first argument is the parser, + not userData. +*/ +typedef int (XMLCALL *XML_ExternalEntityRefHandler) ( + XML_Parser parser, + const XML_Char *context, + const XML_Char *base, + const XML_Char *systemId, + const XML_Char *publicId); + +/* This is called in two situations: + 1) An entity reference is encountered for which no declaration + has been read *and* this is not an error. + 2) An internal entity reference is read, but not expanded, because + XML_SetDefaultHandler has been called. + Note: skipped parameter entities in declarations and skipped general + entities in attribute values cannot be reported, because + the event would be out of sync with the reporting of the + declarations or attribute values +*/ +typedef void (XMLCALL *XML_SkippedEntityHandler) ( + void *userData, + const XML_Char *entityName, + int is_parameter_entity); + +/* This structure is filled in by the XML_UnknownEncodingHandler to + provide information to the parser about encodings that are unknown + to the parser. + + The map[b] member gives information about byte sequences whose + first byte is b. + + If map[b] is c where c is >= 0, then b by itself encodes the + Unicode scalar value c. + + If map[b] is -1, then the byte sequence is malformed. + + If map[b] is -n, where n >= 2, then b is the first byte of an + n-byte sequence that encodes a single Unicode scalar value. + + The data member will be passed as the first argument to the convert + function. + + The convert function is used to convert multibyte sequences; s will + point to a n-byte sequence where map[(unsigned char)*s] == -n. The + convert function must return the Unicode scalar value represented + by this byte sequence or -1 if the byte sequence is malformed. + + The convert function may be NULL if the encoding is a single-byte + encoding, that is if map[b] >= -1 for all bytes b. + + When the parser is finished with the encoding, then if release is + not NULL, it will call release passing it the data member; once + release has been called, the convert function will not be called + again. + + Expat places certain restrictions on the encodings that are supported + using this mechanism. + + 1. Every ASCII character that can appear in a well-formed XML document, + other than the characters + + $@\^`{}~ + + must be represented by a single byte, and that byte must be the + same byte that represents that character in ASCII. + + 2. No character may require more than 4 bytes to encode. + + 3. All characters encoded must have Unicode scalar values <= + 0xFFFF, (i.e., characters that would be encoded by surrogates in + UTF-16 are not allowed). Note that this restriction doesn't + apply to the built-in support for UTF-8 and UTF-16. + + 4. No Unicode character may be encoded by more than one distinct + sequence of bytes. +*/ +typedef struct { + int map[256]; + void *data; + int (XMLCALL *convert)(void *data, const char *s); + void (XMLCALL *release)(void *data); +} XML_Encoding; + +/* This is called for an encoding that is unknown to the parser. + + The encodingHandlerData argument is that which was passed as the + second argument to XML_SetUnknownEncodingHandler. + + The name argument gives the name of the encoding as specified in + the encoding declaration. + + If the callback can provide information about the encoding, it must + fill in the XML_Encoding structure, and return XML_STATUS_OK. + Otherwise it must return XML_STATUS_ERROR. + + If info does not describe a suitable encoding, then the parser will + return an XML_UNKNOWN_ENCODING error. +*/ +typedef int (XMLCALL *XML_UnknownEncodingHandler) ( + void *encodingHandlerData, + const XML_Char *name, + XML_Encoding *info); + +XMLPARSEAPI(void) +XML_SetElementHandler(XML_Parser parser, + XML_StartElementHandler start, + XML_EndElementHandler end); + +XMLPARSEAPI(void) +XML_SetStartElementHandler(XML_Parser parser, + XML_StartElementHandler handler); + +XMLPARSEAPI(void) +XML_SetEndElementHandler(XML_Parser parser, + XML_EndElementHandler handler); + +XMLPARSEAPI(void) +XML_SetCharacterDataHandler(XML_Parser parser, + XML_CharacterDataHandler handler); + +XMLPARSEAPI(void) +XML_SetProcessingInstructionHandler(XML_Parser parser, + XML_ProcessingInstructionHandler handler); +XMLPARSEAPI(void) +XML_SetCommentHandler(XML_Parser parser, + XML_CommentHandler handler); + +XMLPARSEAPI(void) +XML_SetCdataSectionHandler(XML_Parser parser, + XML_StartCdataSectionHandler start, + XML_EndCdataSectionHandler end); + +XMLPARSEAPI(void) +XML_SetStartCdataSectionHandler(XML_Parser parser, + XML_StartCdataSectionHandler start); + +XMLPARSEAPI(void) +XML_SetEndCdataSectionHandler(XML_Parser parser, + XML_EndCdataSectionHandler end); + +/* This sets the default handler and also inhibits expansion of + internal entities. These entity references will be passed to the + default handler, or to the skipped entity handler, if one is set. +*/ +XMLPARSEAPI(void) +XML_SetDefaultHandler(XML_Parser parser, + XML_DefaultHandler handler); + +/* This sets the default handler but does not inhibit expansion of + internal entities. The entity reference will not be passed to the + default handler. +*/ +XMLPARSEAPI(void) +XML_SetDefaultHandlerExpand(XML_Parser parser, + XML_DefaultHandler handler); + +XMLPARSEAPI(void) +XML_SetDoctypeDeclHandler(XML_Parser parser, + XML_StartDoctypeDeclHandler start, + XML_EndDoctypeDeclHandler end); + +XMLPARSEAPI(void) +XML_SetStartDoctypeDeclHandler(XML_Parser parser, + XML_StartDoctypeDeclHandler start); + +XMLPARSEAPI(void) +XML_SetEndDoctypeDeclHandler(XML_Parser parser, + XML_EndDoctypeDeclHandler end); + +XMLPARSEAPI(void) +XML_SetUnparsedEntityDeclHandler(XML_Parser parser, + XML_UnparsedEntityDeclHandler handler); + +XMLPARSEAPI(void) +XML_SetNotationDeclHandler(XML_Parser parser, + XML_NotationDeclHandler handler); + +XMLPARSEAPI(void) +XML_SetNamespaceDeclHandler(XML_Parser parser, + XML_StartNamespaceDeclHandler start, + XML_EndNamespaceDeclHandler end); + +XMLPARSEAPI(void) +XML_SetStartNamespaceDeclHandler(XML_Parser parser, + XML_StartNamespaceDeclHandler start); + +XMLPARSEAPI(void) +XML_SetEndNamespaceDeclHandler(XML_Parser parser, + XML_EndNamespaceDeclHandler end); + +XMLPARSEAPI(void) +XML_SetNotStandaloneHandler(XML_Parser parser, + XML_NotStandaloneHandler handler); + +XMLPARSEAPI(void) +XML_SetExternalEntityRefHandler(XML_Parser parser, + XML_ExternalEntityRefHandler handler); + +/* If a non-NULL value for arg is specified here, then it will be + passed as the first argument to the external entity ref handler + instead of the parser object. +*/ +XMLPARSEAPI(void) +XML_SetExternalEntityRefHandlerArg(XML_Parser parser, + void *arg); + +XMLPARSEAPI(void) +XML_SetSkippedEntityHandler(XML_Parser parser, + XML_SkippedEntityHandler handler); + +XMLPARSEAPI(void) +XML_SetUnknownEncodingHandler(XML_Parser parser, + XML_UnknownEncodingHandler handler, + void *encodingHandlerData); + +/* This can be called within a handler for a start element, end + element, processing instruction or character data. It causes the + corresponding markup to be passed to the default handler. +*/ +XMLPARSEAPI(void) +XML_DefaultCurrent(XML_Parser parser); + +/* If do_nst is non-zero, and namespace processing is in effect, and + a name has a prefix (i.e. an explicit namespace qualifier) then + that name is returned as a triplet in a single string separated by + the separator character specified when the parser was created: URI + + sep + local_name + sep + prefix. + + If do_nst is zero, then namespace information is returned in the + default manner (URI + sep + local_name) whether or not the name + has a prefix. + + Note: Calling XML_SetReturnNSTriplet after XML_Parse or + XML_ParseBuffer has no effect. +*/ + +XMLPARSEAPI(void) +XML_SetReturnNSTriplet(XML_Parser parser, int do_nst); + +/* This value is passed as the userData argument to callbacks. */ +XMLPARSEAPI(void) +XML_SetUserData(XML_Parser parser, void *userData); + +/* Returns the last value set by XML_SetUserData or NULL. */ +#define XML_GetUserData(parser) (*(void **)(parser)) + +/* This is equivalent to supplying an encoding argument to + XML_ParserCreate. On success XML_SetEncoding returns non-zero, + zero otherwise. + Note: Calling XML_SetEncoding after XML_Parse or XML_ParseBuffer + has no effect and returns XML_STATUS_ERROR. +*/ +XMLPARSEAPI(enum XML_Status) +XML_SetEncoding(XML_Parser parser, const XML_Char *encoding); + +/* If this function is called, then the parser will be passed as the + first argument to callbacks instead of userData. The userData will + still be accessible using XML_GetUserData. +*/ +XMLPARSEAPI(void) +XML_UseParserAsHandlerArg(XML_Parser parser); + +/* If useDTD == XML_TRUE is passed to this function, then the parser + will assume that there is an external subset, even if none is + specified in the document. In such a case the parser will call the + externalEntityRefHandler with a value of NULL for the systemId + argument (the publicId and context arguments will be NULL as well). + Note: For the purpose of checking WFC: Entity Declared, passing + useDTD == XML_TRUE will make the parser behave as if the document + had a DTD with an external subset. + Note: If this function is called, then this must be done before + the first call to XML_Parse or XML_ParseBuffer, since it will + have no effect after that. Returns + XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING. + Note: If the document does not have a DOCTYPE declaration at all, + then startDoctypeDeclHandler and endDoctypeDeclHandler will not + be called, despite an external subset being parsed. + Note: If XML_DTD is not defined when Expat is compiled, returns + XML_ERROR_FEATURE_REQUIRES_XML_DTD. +*/ +XMLPARSEAPI(enum XML_Error) +XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD); + + +/* Sets the base to be used for resolving relative URIs in system + identifiers in declarations. Resolving relative identifiers is + left to the application: this value will be passed through as the + base argument to the XML_ExternalEntityRefHandler, + XML_NotationDeclHandler and XML_UnparsedEntityDeclHandler. The base + argument will be copied. Returns XML_STATUS_ERROR if out of memory, + XML_STATUS_OK otherwise. +*/ +XMLPARSEAPI(enum XML_Status) +XML_SetBase(XML_Parser parser, const XML_Char *base); + +XMLPARSEAPI(const XML_Char *) +XML_GetBase(XML_Parser parser); + +/* Returns the number of the attribute/value pairs passed in last call + to the XML_StartElementHandler that were specified in the start-tag + rather than defaulted. Each attribute/value pair counts as 2; thus + this correspondds to an index into the atts array passed to the + XML_StartElementHandler. +*/ +XMLPARSEAPI(int) +XML_GetSpecifiedAttributeCount(XML_Parser parser); + +/* Returns the index of the ID attribute passed in the last call to + XML_StartElementHandler, or -1 if there is no ID attribute. Each + attribute/value pair counts as 2; thus this correspondds to an + index into the atts array passed to the XML_StartElementHandler. +*/ +XMLPARSEAPI(int) +XML_GetIdAttributeIndex(XML_Parser parser); + +#ifdef XML_ATTR_INFO +/* Source file byte offsets for the start and end of attribute names and values. + The value indices are exclusive of surrounding quotes; thus in a UTF-8 source + file an attribute value of "blah" will yield: + info->valueEnd - info->valueStart = 4 bytes. +*/ +typedef struct { + XML_Index nameStart; /* Offset to beginning of the attribute name. */ + XML_Index nameEnd; /* Offset after the attribute name's last byte. */ + XML_Index valueStart; /* Offset to beginning of the attribute value. */ + XML_Index valueEnd; /* Offset after the attribute value's last byte. */ +} XML_AttrInfo; + +/* Returns an array of XML_AttrInfo structures for the attribute/value pairs + passed in last call to the XML_StartElementHandler that were specified + in the start-tag rather than defaulted. Each attribute/value pair counts + as 1; thus the number of entries in the array is + XML_GetSpecifiedAttributeCount(parser) / 2. +*/ +XMLPARSEAPI(const XML_AttrInfo *) +XML_GetAttributeInfo(XML_Parser parser); +#endif + +/* Parses some input. Returns XML_STATUS_ERROR if a fatal error is + detected. The last call to XML_Parse must have isFinal true; len + may be zero for this call (or any other). + + Though the return values for these functions has always been + described as a Boolean value, the implementation, at least for the + 1.95.x series, has always returned exactly one of the XML_Status + values. +*/ +XMLPARSEAPI(enum XML_Status) +XML_Parse(XML_Parser parser, const char *s, int len, int isFinal); + +XMLPARSEAPI(void *) +XML_GetBuffer(XML_Parser parser, int len); + +XMLPARSEAPI(enum XML_Status) +XML_ParseBuffer(XML_Parser parser, int len, int isFinal); + +/* Stops parsing, causing XML_Parse() or XML_ParseBuffer() to return. + Must be called from within a call-back handler, except when aborting + (resumable = 0) an already suspended parser. Some call-backs may + still follow because they would otherwise get lost. Examples: + - endElementHandler() for empty elements when stopped in + startElementHandler(), + - endNameSpaceDeclHandler() when stopped in endElementHandler(), + and possibly others. + + Can be called from most handlers, including DTD related call-backs, + except when parsing an external parameter entity and resumable != 0. + Returns XML_STATUS_OK when successful, XML_STATUS_ERROR otherwise. + Possible error codes: + - XML_ERROR_SUSPENDED: when suspending an already suspended parser. + - XML_ERROR_FINISHED: when the parser has already finished. + - XML_ERROR_SUSPEND_PE: when suspending while parsing an external PE. + + When resumable != 0 (true) then parsing is suspended, that is, + XML_Parse() and XML_ParseBuffer() return XML_STATUS_SUSPENDED. + Otherwise, parsing is aborted, that is, XML_Parse() and XML_ParseBuffer() + return XML_STATUS_ERROR with error code XML_ERROR_ABORTED. + + *Note*: + This will be applied to the current parser instance only, that is, if + there is a parent parser then it will continue parsing when the + externalEntityRefHandler() returns. It is up to the implementation of + the externalEntityRefHandler() to call XML_StopParser() on the parent + parser (recursively), if one wants to stop parsing altogether. + + When suspended, parsing can be resumed by calling XML_ResumeParser(). +*/ +XMLPARSEAPI(enum XML_Status) +XML_StopParser(XML_Parser parser, XML_Bool resumable); + +/* Resumes parsing after it has been suspended with XML_StopParser(). + Must not be called from within a handler call-back. Returns same + status codes as XML_Parse() or XML_ParseBuffer(). + Additional error code XML_ERROR_NOT_SUSPENDED possible. + + *Note*: + This must be called on the most deeply nested child parser instance + first, and on its parent parser only after the child parser has finished, + to be applied recursively until the document entity's parser is restarted. + That is, the parent parser will not resume by itself and it is up to the + application to call XML_ResumeParser() on it at the appropriate moment. +*/ +XMLPARSEAPI(enum XML_Status) +XML_ResumeParser(XML_Parser parser); + +enum XML_Parsing { + XML_INITIALIZED, + XML_PARSING, + XML_FINISHED, + XML_SUSPENDED +}; + +typedef struct { + enum XML_Parsing parsing; + XML_Bool finalBuffer; +} XML_ParsingStatus; + +/* Returns status of parser with respect to being initialized, parsing, + finished, or suspended and processing the final buffer. + XXX XML_Parse() and XML_ParseBuffer() should return XML_ParsingStatus, + XXX with XML_FINISHED_OK or XML_FINISHED_ERROR replacing XML_FINISHED +*/ +XMLPARSEAPI(void) +XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status); + +/* Creates an XML_Parser object that can parse an external general + entity; context is a '\0'-terminated string specifying the parse + context; encoding is a '\0'-terminated string giving the name of + the externally specified encoding, or NULL if there is no + externally specified encoding. The context string consists of a + sequence of tokens separated by formfeeds (\f); a token consisting + of a name specifies that the general entity of the name is open; a + token of the form prefix=uri specifies the namespace for a + particular prefix; a token of the form =uri specifies the default + namespace. This can be called at any point after the first call to + an ExternalEntityRefHandler so longer as the parser has not yet + been freed. The new parser is completely independent and may + safely be used in a separate thread. The handlers and userData are + initialized from the parser argument. Returns NULL if out of memory. + Otherwise returns a new XML_Parser object. +*/ +XMLPARSEAPI(XML_Parser) +XML_ExternalEntityParserCreate(XML_Parser parser, + const XML_Char *context, + const XML_Char *encoding); + +enum XML_ParamEntityParsing { + XML_PARAM_ENTITY_PARSING_NEVER, + XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE, + XML_PARAM_ENTITY_PARSING_ALWAYS +}; + +/* Controls parsing of parameter entities (including the external DTD + subset). If parsing of parameter entities is enabled, then + references to external parameter entities (including the external + DTD subset) will be passed to the handler set with + XML_SetExternalEntityRefHandler. The context passed will be 0. + + Unlike external general entities, external parameter entities can + only be parsed synchronously. If the external parameter entity is + to be parsed, it must be parsed during the call to the external + entity ref handler: the complete sequence of + XML_ExternalEntityParserCreate, XML_Parse/XML_ParseBuffer and + XML_ParserFree calls must be made during this call. After + XML_ExternalEntityParserCreate has been called to create the parser + for the external parameter entity (context must be 0 for this + call), it is illegal to make any calls on the old parser until + XML_ParserFree has been called on the newly created parser. + If the library has been compiled without support for parameter + entity parsing (ie without XML_DTD being defined), then + XML_SetParamEntityParsing will return 0 if parsing of parameter + entities is requested; otherwise it will return non-zero. + Note: If XML_SetParamEntityParsing is called after XML_Parse or + XML_ParseBuffer, then it has no effect and will always return 0. +*/ +XMLPARSEAPI(int) +XML_SetParamEntityParsing(XML_Parser parser, + enum XML_ParamEntityParsing parsing); + +/* Sets the hash salt to use for internal hash calculations. + Helps in preventing DoS attacks based on predicting hash + function behavior. This must be called before parsing is started. + Returns 1 if successful, 0 when called after parsing has started. +*/ +XMLPARSEAPI(int) +XML_SetHashSalt(XML_Parser parser, + unsigned long hash_salt); + +/* If XML_Parse or XML_ParseBuffer have returned XML_STATUS_ERROR, then + XML_GetErrorCode returns information about the error. +*/ +XMLPARSEAPI(enum XML_Error) +XML_GetErrorCode(XML_Parser parser); + +/* These functions return information about the current parse + location. They may be called from any callback called to report + some parse event; in this case the location is the location of the + first of the sequence of characters that generated the event. When + called from callbacks generated by declarations in the document + prologue, the location identified isn't as neatly defined, but will + be within the relevant markup. When called outside of the callback + functions, the position indicated will be just past the last parse + event (regardless of whether there was an associated callback). + + They may also be called after returning from a call to XML_Parse + or XML_ParseBuffer. If the return value is XML_STATUS_ERROR then + the location is the location of the character at which the error + was detected; otherwise the location is the location of the last + parse event, as described above. +*/ +XMLPARSEAPI(XML_Size) XML_GetCurrentLineNumber(XML_Parser parser); +XMLPARSEAPI(XML_Size) XML_GetCurrentColumnNumber(XML_Parser parser); +XMLPARSEAPI(XML_Index) XML_GetCurrentByteIndex(XML_Parser parser); + +/* Return the number of bytes in the current event. + Returns 0 if the event is in an internal entity. +*/ +XMLPARSEAPI(int) +XML_GetCurrentByteCount(XML_Parser parser); + +/* If XML_CONTEXT_BYTES is defined, returns the input buffer, sets + the integer pointed to by offset to the offset within this buffer + of the current parse position, and sets the integer pointed to by size + to the size of this buffer (the number of input bytes). Otherwise + returns a NULL pointer. Also returns a NULL pointer if a parse isn't + active. + + NOTE: The character pointer returned should not be used outside + the handler that makes the call. +*/ +XMLPARSEAPI(const char *) +XML_GetInputContext(XML_Parser parser, + int *offset, + int *size); + +/* For backwards compatibility with previous versions. */ +#define XML_GetErrorLineNumber XML_GetCurrentLineNumber +#define XML_GetErrorColumnNumber XML_GetCurrentColumnNumber +#define XML_GetErrorByteIndex XML_GetCurrentByteIndex + +/* Frees the content model passed to the element declaration handler */ +XMLPARSEAPI(void) +XML_FreeContentModel(XML_Parser parser, XML_Content *model); + +/* Exposing the memory handling functions used in Expat */ +XMLPARSEAPI(void *) +XML_MemMalloc(XML_Parser parser, size_t size); + +XMLPARSEAPI(void *) +XML_MemRealloc(XML_Parser parser, void *ptr, size_t size); + +XMLPARSEAPI(void) +XML_MemFree(XML_Parser parser, void *ptr); + +/* Frees memory used by the parser. */ +XMLPARSEAPI(void) +XML_ParserFree(XML_Parser parser); + +/* Returns a string describing the error. */ +XMLPARSEAPI(const XML_LChar *) +XML_ErrorString(enum XML_Error code); + +/* Return a string containing the version number of this expat */ +XMLPARSEAPI(const XML_LChar *) +XML_ExpatVersion(void); + +typedef struct { + int major; + int minor; + int micro; +} XML_Expat_Version; + +/* Return an XML_Expat_Version structure containing numeric version + number information for this version of expat. +*/ +XMLPARSEAPI(XML_Expat_Version) +XML_ExpatVersionInfo(void); + +/* Added in Expat 1.95.5. */ +enum XML_FeatureEnum { + XML_FEATURE_END = 0, + XML_FEATURE_UNICODE, + XML_FEATURE_UNICODE_WCHAR_T, + XML_FEATURE_DTD, + XML_FEATURE_CONTEXT_BYTES, + XML_FEATURE_MIN_SIZE, + XML_FEATURE_SIZEOF_XML_CHAR, + XML_FEATURE_SIZEOF_XML_LCHAR, + XML_FEATURE_NS, + XML_FEATURE_LARGE_SIZE, + XML_FEATURE_ATTR_INFO + /* Additional features must be added to the end of this enum. */ +}; + +typedef struct { + enum XML_FeatureEnum feature; + const XML_LChar *name; + long int value; +} XML_Feature; + +XMLPARSEAPI(const XML_Feature *) +XML_GetFeatureList(void); + + +/* Expat follows the GNU/Linux convention of odd number minor version for + beta/development releases and even number minor version for stable + releases. Micro is bumped with each release, and set to 0 with each + change to major or minor version. +*/ +#define XML_MAJOR_VERSION 2 +#define XML_MINOR_VERSION 1 +#define XML_MICRO_VERSION 0 + +#ifdef __cplusplus +} +#endif + +#endif /* not Expat_INCLUDED */ diff --git a/xml/details/expat/expat_external.h b/xml/details/expat/expat_external.h new file mode 100644 index 0000000..926e32c --- /dev/null +++ b/xml/details/expat/expat_external.h @@ -0,0 +1,119 @@ +/* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#ifndef Expat_External_INCLUDED +#define Expat_External_INCLUDED 1 + +/* Link to Expat API statically */ + +#define XML_STATIC 1 + +/* External API definitions */ + +#if defined(_MSC_EXTENSIONS) && !defined(__BEOS__) && !defined(__CYGWIN__) +#define XML_USE_MSC_EXTENSIONS 1 +#endif + +/* Expat tries very hard to make the API boundary very specifically + defined. There are two macros defined to control this boundary; + each of these can be defined before including this header to + achieve some different behavior, but doing so it not recommended or + tested frequently. + + XMLCALL - The calling convention to use for all calls across the + "library boundary." This will default to cdecl, and + try really hard to tell the compiler that's what we + want. + + XMLIMPORT - Whatever magic is needed to note that a function is + to be imported from a dynamically loaded library + (.dll, .so, or .sl, depending on your platform). + + The XMLCALL macro was added in Expat 1.95.7. The only one which is + expected to be directly useful in client code is XMLCALL. + + Note that on at least some Unix versions, the Expat library must be + compiled with the cdecl calling convention as the default since + system headers may assume the cdecl convention. +*/ +#ifndef XMLCALL +#if defined(_MSC_VER) +#define XMLCALL __cdecl +#elif defined(__GNUC__) && defined(__i386) && !defined(__INTEL_COMPILER) +#define XMLCALL __attribute__((cdecl)) +#else +/* For any platform which uses this definition and supports more than + one calling convention, we need to extend this definition to + declare the convention used on that platform, if it's possible to + do so. + + If this is the case for your platform, please file a bug report + with information on how to identify your platform via the C + pre-processor and how to specify the same calling convention as the + platform's malloc() implementation. +*/ +#define XMLCALL +#endif +#endif /* not defined XMLCALL */ + + +#if !defined(XML_STATIC) && !defined(XMLIMPORT) +#ifndef XML_BUILDING_EXPAT +/* using Expat from an application */ + +#ifdef XML_USE_MSC_EXTENSIONS +#define XMLIMPORT __declspec(dllimport) +#endif + +#endif +#endif /* not defined XML_STATIC */ + + +/* If we didn't define it above, define it away: */ +#ifndef XMLIMPORT +#define XMLIMPORT +#endif + + +#define XMLPARSEAPI(type) XMLIMPORT type XMLCALL + +#ifdef __cplusplus +extern "C" { +#endif + +#ifdef XML_UNICODE_WCHAR_T +#define XML_UNICODE +#endif + +#ifdef XML_UNICODE /* Information is UTF-16 encoded. */ +#ifdef XML_UNICODE_WCHAR_T +typedef wchar_t XML_Char; +typedef wchar_t XML_LChar; +#else +typedef unsigned short XML_Char; +typedef char XML_LChar; +#endif /* XML_UNICODE_WCHAR_T */ +#else /* Information is UTF-8 encoded. */ +typedef char XML_Char; +typedef char XML_LChar; +#endif /* XML_UNICODE */ + +#ifdef XML_LARGE_SIZE /* Use large integers for file/stream positions. */ +#if defined(XML_USE_MSC_EXTENSIONS) && _MSC_VER < 1400 +typedef __int64 XML_Index; +typedef unsigned __int64 XML_Size; +#else +typedef long long XML_Index; +typedef unsigned long long XML_Size; +#endif +#else +typedef long XML_Index; +typedef unsigned long XML_Size; +#endif /* XML_LARGE_SIZE */ + +#ifdef __cplusplus +} +#endif + +#endif /* not Expat_External_INCLUDED */ diff --git a/xml/details/expat/iasciitab.h b/xml/details/expat/iasciitab.h new file mode 100644 index 0000000..24a1d5c --- /dev/null +++ b/xml/details/expat/iasciitab.h @@ -0,0 +1,37 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +/* Like asciitab.h, except that 0xD has code BT_S rather than BT_CR */ +/* 0x00 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x04 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x08 */ BT_NONXML, BT_S, BT_LF, BT_NONXML, +/* 0x0C */ BT_NONXML, BT_S, BT_NONXML, BT_NONXML, +/* 0x10 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x14 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x18 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x1C */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0x20 */ BT_S, BT_EXCL, BT_QUOT, BT_NUM, +/* 0x24 */ BT_OTHER, BT_PERCNT, BT_AMP, BT_APOS, +/* 0x28 */ BT_LPAR, BT_RPAR, BT_AST, BT_PLUS, +/* 0x2C */ BT_COMMA, BT_MINUS, BT_NAME, BT_SOL, +/* 0x30 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT, +/* 0x34 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT, +/* 0x38 */ BT_DIGIT, BT_DIGIT, BT_COLON, BT_SEMI, +/* 0x3C */ BT_LT, BT_EQUALS, BT_GT, BT_QUEST, +/* 0x40 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX, +/* 0x44 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT, +/* 0x48 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x4C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x50 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x54 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x58 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_LSQB, +/* 0x5C */ BT_OTHER, BT_RSQB, BT_OTHER, BT_NMSTRT, +/* 0x60 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX, +/* 0x64 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT, +/* 0x68 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x6C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x70 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x74 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0x78 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER, +/* 0x7C */ BT_VERBAR, BT_OTHER, BT_OTHER, BT_OTHER, diff --git a/xml/details/expat/internal.h b/xml/details/expat/internal.h new file mode 100644 index 0000000..dd54548 --- /dev/null +++ b/xml/details/expat/internal.h @@ -0,0 +1,73 @@ +/* internal.h + + Internal definitions used by Expat. This is not needed to compile + client code. + + The following calling convention macros are defined for frequently + called functions: + + FASTCALL - Used for those internal functions that have a simple + body and a low number of arguments and local variables. + + PTRCALL - Used for functions called though function pointers. + + PTRFASTCALL - Like PTRCALL, but for low number of arguments. + + inline - Used for selected internal functions for which inlining + may improve performance on some platforms. + + Note: Use of these macros is based on judgement, not hard rules, + and therefore subject to change. +*/ + +#if defined(__GNUC__) && defined(__i386__) && !defined(__MINGW32__) +/* We'll use this version by default only where we know it helps. + + regparm() generates warnings on Solaris boxes. See SF bug #692878. + + Instability reported with egcs on a RedHat Linux 7.3. + Let's comment out: + #define FASTCALL __attribute__((stdcall, regparm(3))) + and let's try this: +*/ +#define FASTCALL __attribute__((regparm(3))) +#define PTRFASTCALL __attribute__((regparm(3))) +#endif + +/* Using __fastcall seems to have an unexpected negative effect under + MS VC++, especially for function pointers, so we won't use it for + now on that platform. It may be reconsidered for a future release + if it can be made more effective. + Likely reason: __fastcall on Windows is like stdcall, therefore + the compiler cannot perform stack optimizations for call clusters. +*/ + +/* Make sure all of these are defined if they aren't already. */ + +#ifndef FASTCALL +#define FASTCALL +#endif + +#ifndef PTRCALL +#define PTRCALL +#endif + +#ifndef PTRFASTCALL +#define PTRFASTCALL +#endif + +#ifndef XML_MIN_SIZE +#if !defined(__cplusplus) && !defined(inline) +#ifdef __GNUC__ +#define inline __inline +#endif /* __GNUC__ */ +#endif +#endif /* XML_MIN_SIZE */ + +#ifdef __cplusplus +#define inline inline +#else +#ifndef inline +#define inline +#endif +#endif diff --git a/xml/details/expat/latin1tab.h b/xml/details/expat/latin1tab.h new file mode 100644 index 0000000..53c25d7 --- /dev/null +++ b/xml/details/expat/latin1tab.h @@ -0,0 +1,36 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +/* 0x80 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0x84 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0x88 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0x8C */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0x90 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0x94 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0x98 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0x9C */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0xA0 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0xA4 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0xA8 */ BT_OTHER, BT_OTHER, BT_NMSTRT, BT_OTHER, +/* 0xAC */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0xB0 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0xB4 */ BT_OTHER, BT_NMSTRT, BT_OTHER, BT_NAME, +/* 0xB8 */ BT_OTHER, BT_OTHER, BT_NMSTRT, BT_OTHER, +/* 0xBC */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER, +/* 0xC0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xC4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xC8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xCC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xD0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xD4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER, +/* 0xD8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xDC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xE0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xE4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xE8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xEC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xF0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xF4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER, +/* 0xF8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, +/* 0xFC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, diff --git a/xml/details/expat/nametab.h b/xml/details/expat/nametab.h new file mode 100644 index 0000000..b05e62c --- /dev/null +++ b/xml/details/expat/nametab.h @@ -0,0 +1,150 @@ +static const unsigned namingBitmap[] = { +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, +0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, +0x00000000, 0x04000000, 0x87FFFFFE, 0x07FFFFFE, +0x00000000, 0x00000000, 0xFF7FFFFF, 0xFF7FFFFF, +0xFFFFFFFF, 0x7FF3FFFF, 0xFFFFFDFE, 0x7FFFFFFF, +0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFE00F, 0xFC31FFFF, +0x00FFFFFF, 0x00000000, 0xFFFF0000, 0xFFFFFFFF, +0xFFFFFFFF, 0xF80001FF, 0x00000003, 0x00000000, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0xFFFFD740, 0xFFFFFFFB, 0x547F7FFF, 0x000FFFFD, +0xFFFFDFFE, 0xFFFFFFFF, 0xDFFEFFFF, 0xFFFFFFFF, +0xFFFF0003, 0xFFFFFFFF, 0xFFFF199F, 0x033FCFFF, +0x00000000, 0xFFFE0000, 0x027FFFFF, 0xFFFFFFFE, +0x0000007F, 0x00000000, 0xFFFF0000, 0x000707FF, +0x00000000, 0x07FFFFFE, 0x000007FE, 0xFFFE0000, +0xFFFFFFFF, 0x7CFFFFFF, 0x002F7FFF, 0x00000060, +0xFFFFFFE0, 0x23FFFFFF, 0xFF000000, 0x00000003, +0xFFF99FE0, 0x03C5FDFF, 0xB0000000, 0x00030003, +0xFFF987E0, 0x036DFDFF, 0x5E000000, 0x001C0000, +0xFFFBAFE0, 0x23EDFDFF, 0x00000000, 0x00000001, +0xFFF99FE0, 0x23CDFDFF, 0xB0000000, 0x00000003, +0xD63DC7E0, 0x03BFC718, 0x00000000, 0x00000000, +0xFFFDDFE0, 0x03EFFDFF, 0x00000000, 0x00000003, +0xFFFDDFE0, 0x03EFFDFF, 0x40000000, 0x00000003, +0xFFFDDFE0, 0x03FFFDFF, 0x00000000, 0x00000003, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0xFFFFFFFE, 0x000D7FFF, 0x0000003F, 0x00000000, +0xFEF02596, 0x200D6CAE, 0x0000001F, 0x00000000, +0x00000000, 0x00000000, 0xFFFFFEFF, 0x000003FF, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0x00000000, 0xFFFFFFFF, 0xFFFF003F, 0x007FFFFF, +0x0007DAED, 0x50000000, 0x82315001, 0x002C62AB, +0x40000000, 0xF580C900, 0x00000007, 0x02010800, +0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, +0x0FFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0x03FFFFFF, +0x3F3FFFFF, 0xFFFFFFFF, 0xAAFF3F3F, 0x3FFFFFFF, +0xFFFFFFFF, 0x5FDFFFFF, 0x0FCF1FDC, 0x1FDC1FFF, +0x00000000, 0x00004C40, 0x00000000, 0x00000000, +0x00000007, 0x00000000, 0x00000000, 0x00000000, +0x00000080, 0x000003FE, 0xFFFFFFFE, 0xFFFFFFFF, +0x001FFFFF, 0xFFFFFFFE, 0xFFFFFFFF, 0x07FFFFFF, +0xFFFFFFE0, 0x00001FFF, 0x00000000, 0x00000000, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, +0xFFFFFFFF, 0x0000003F, 0x00000000, 0x00000000, +0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, +0xFFFFFFFF, 0x0000000F, 0x00000000, 0x00000000, +0x00000000, 0x07FF6000, 0x87FFFFFE, 0x07FFFFFE, +0x00000000, 0x00800000, 0xFF7FFFFF, 0xFF7FFFFF, +0x00FFFFFF, 0x00000000, 0xFFFF0000, 0xFFFFFFFF, +0xFFFFFFFF, 0xF80001FF, 0x00030003, 0x00000000, +0xFFFFFFFF, 0xFFFFFFFF, 0x0000003F, 0x00000003, +0xFFFFD7C0, 0xFFFFFFFB, 0x547F7FFF, 0x000FFFFD, +0xFFFFDFFE, 0xFFFFFFFF, 0xDFFEFFFF, 0xFFFFFFFF, +0xFFFF007B, 0xFFFFFFFF, 0xFFFF199F, 0x033FCFFF, +0x00000000, 0xFFFE0000, 0x027FFFFF, 0xFFFFFFFE, +0xFFFE007F, 0xBBFFFFFB, 0xFFFF0016, 0x000707FF, +0x00000000, 0x07FFFFFE, 0x0007FFFF, 0xFFFF03FF, +0xFFFFFFFF, 0x7CFFFFFF, 0xFFEF7FFF, 0x03FF3DFF, +0xFFFFFFEE, 0xF3FFFFFF, 0xFF1E3FFF, 0x0000FFCF, +0xFFF99FEE, 0xD3C5FDFF, 0xB080399F, 0x0003FFCF, +0xFFF987E4, 0xD36DFDFF, 0x5E003987, 0x001FFFC0, +0xFFFBAFEE, 0xF3EDFDFF, 0x00003BBF, 0x0000FFC1, +0xFFF99FEE, 0xF3CDFDFF, 0xB0C0398F, 0x0000FFC3, +0xD63DC7EC, 0xC3BFC718, 0x00803DC7, 0x0000FF80, +0xFFFDDFEE, 0xC3EFFDFF, 0x00603DDF, 0x0000FFC3, +0xFFFDDFEC, 0xC3EFFDFF, 0x40603DDF, 0x0000FFC3, +0xFFFDDFEC, 0xC3FFFDFF, 0x00803DCF, 0x0000FFC3, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0xFFFFFFFE, 0x07FF7FFF, 0x03FF7FFF, 0x00000000, +0xFEF02596, 0x3BFF6CAE, 0x03FF3F5F, 0x00000000, +0x03000000, 0xC2A003FF, 0xFFFFFEFF, 0xFFFE03FF, +0xFEBF0FDF, 0x02FE3FFF, 0x00000000, 0x00000000, +0x00000000, 0x00000000, 0x00000000, 0x00000000, +0x00000000, 0x00000000, 0x1FFF0000, 0x00000002, +0x000000A0, 0x003EFFFE, 0xFFFFFFFE, 0xFFFFFFFF, +0x661FFFFF, 0xFFFFFFFE, 0xFFFFFFFF, 0x77FFFFFF, +}; +static const unsigned char nmstrtPages[] = { +0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x00, +0x00, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, +0x10, 0x11, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x12, 0x13, +0x00, 0x14, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x15, 0x16, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x17, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x18, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +}; +static const unsigned char namePages[] = { +0x19, 0x03, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x00, +0x00, 0x1F, 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, +0x10, 0x11, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x12, 0x13, +0x26, 0x14, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x27, 0x16, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x17, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, +0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x18, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +}; diff --git a/xml/details/expat/utf8tab.h b/xml/details/expat/utf8tab.h new file mode 100644 index 0000000..7bb3e77 --- /dev/null +++ b/xml/details/expat/utf8tab.h @@ -0,0 +1,37 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + + +/* 0x80 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0x84 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0x88 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0x8C */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0x90 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0x94 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0x98 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0x9C */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xA0 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xA4 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xA8 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xAC */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xB0 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xB4 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xB8 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xBC */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL, +/* 0xC0 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xC4 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xC8 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xCC */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xD0 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xD4 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xD8 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xDC */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2, +/* 0xE0 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3, +/* 0xE4 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3, +/* 0xE8 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3, +/* 0xEC */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3, +/* 0xF0 */ BT_LEAD4, BT_LEAD4, BT_LEAD4, BT_LEAD4, +/* 0xF4 */ BT_LEAD4, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0xF8 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML, +/* 0xFC */ BT_NONXML, BT_NONXML, BT_MALFORM, BT_MALFORM, diff --git a/xml/details/expat/xmlparse.c b/xml/details/expat/xmlparse.c new file mode 100644 index 0000000..b65612c --- /dev/null +++ b/xml/details/expat/xmlparse.c @@ -0,0 +1,6397 @@ +/* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#include <stddef.h> +#include <string.h> /* memset(), memcpy() */ +#include <assert.h> +#include <limits.h> /* UINT_MAX */ +#include <time.h> /* time() */ + +#define XML_BUILDING_EXPAT 1 + +#include <xml/details/expat/config.h> + +#include <xml/details/expat/ascii.h> +#include <xml/details/expat/expat.h> + +#ifdef XML_UNICODE +#define XML_ENCODE_MAX XML_UTF16_ENCODE_MAX +#define XmlConvert XmlUtf16Convert +#define XmlGetInternalEncoding XmlGetUtf16InternalEncoding +#define XmlGetInternalEncodingNS XmlGetUtf16InternalEncodingNS +#define XmlEncode XmlUtf16Encode +/* Using pointer subtraction to convert to integer type. */ +#define MUST_CONVERT(enc, s) (!(enc)->isUtf16 || (((char *)(s) - (char *)NULL) & 1)) +typedef unsigned short ICHAR; +#else +#define XML_ENCODE_MAX XML_UTF8_ENCODE_MAX +#define XmlConvert XmlUtf8Convert +#define XmlGetInternalEncoding XmlGetUtf8InternalEncoding +#define XmlGetInternalEncodingNS XmlGetUtf8InternalEncodingNS +#define XmlEncode XmlUtf8Encode +#define MUST_CONVERT(enc, s) (!(enc)->isUtf8) +typedef char ICHAR; +#endif + + +#ifndef XML_NS + +#define XmlInitEncodingNS XmlInitEncoding +#define XmlInitUnknownEncodingNS XmlInitUnknownEncoding +#undef XmlGetInternalEncodingNS +#define XmlGetInternalEncodingNS XmlGetInternalEncoding +#define XmlParseXmlDeclNS XmlParseXmlDecl + +#endif + +#ifdef XML_UNICODE + +#ifdef XML_UNICODE_WCHAR_T +#define XML_T(x) (const wchar_t)x +#define XML_L(x) L ## x +#else +#define XML_T(x) (const unsigned short)x +#define XML_L(x) x +#endif + +#else + +#define XML_T(x) x +#define XML_L(x) x + +#endif + +/* Round up n to be a multiple of sz, where sz is a power of 2. */ +#define ROUND_UP(n, sz) (((n) + ((sz) - 1)) & ~((sz) - 1)) + +/* Handle the case where memmove() doesn't exist. */ +#ifndef HAVE_MEMMOVE +#ifdef HAVE_BCOPY +#define memmove(d,s,l) bcopy((s),(d),(l)) +#else +#error memmove does not exist on this platform, nor is a substitute available +#endif /* HAVE_BCOPY */ +#endif /* HAVE_MEMMOVE */ + +#include <xml/details/expat/internal.h> +#include <xml/details/expat/xmltok.h> +#include <xml/details/expat/xmlrole.h> + +typedef const XML_Char *KEY; + +typedef struct { + KEY name; +} NAMED; + +typedef struct { + NAMED **v; + unsigned char power; + size_t size; + size_t used; + const XML_Memory_Handling_Suite *mem; +} HASH_TABLE; + +/* Basic character hash algorithm, taken from Python's string hash: + h = h * 1000003 ^ character, the constant being a prime number. + +*/ +#ifdef XML_UNICODE +#define CHAR_HASH(h, c) \ + (((h) * 0xF4243) ^ (unsigned short)(c)) +#else +#define CHAR_HASH(h, c) \ + (((h) * 0xF4243) ^ (unsigned char)(c)) +#endif + +/* For probing (after a collision) we need a step size relative prime + to the hash table size, which is a power of 2. We use double-hashing, + since we can calculate a second hash value cheaply by taking those bits + of the first hash value that were discarded (masked out) when the table + index was calculated: index = hash & mask, where mask = table->size - 1. + We limit the maximum step size to table->size / 4 (mask >> 2) and make + it odd, since odd numbers are always relative prime to a power of 2. +*/ +#define SECOND_HASH(hash, mask, power) \ + ((((hash) & ~(mask)) >> ((power) - 1)) & ((mask) >> 2)) +#define PROBE_STEP(hash, mask, power) \ + ((unsigned char)((SECOND_HASH(hash, mask, power)) | 1)) + +typedef struct { + NAMED **p; + NAMED **end; +} HASH_TABLE_ITER; + +#define INIT_TAG_BUF_SIZE 32 /* must be a multiple of sizeof(XML_Char) */ +#define INIT_DATA_BUF_SIZE 1024 +#define INIT_ATTS_SIZE 16 +#define INIT_ATTS_VERSION 0xFFFFFFFF +#define INIT_BLOCK_SIZE 1024 +#define INIT_BUFFER_SIZE 1024 + +#define EXPAND_SPARE 24 + +typedef struct binding { + struct prefix *prefix; + struct binding *nextTagBinding; + struct binding *prevPrefixBinding; + const struct attribute_id *attId; + XML_Char *uri; + int uriLen; + int uriAlloc; +} BINDING; + +typedef struct prefix { + const XML_Char *name; + BINDING *binding; +} PREFIX; + +typedef struct { + const XML_Char *str; + const XML_Char *localPart; + const XML_Char *prefix; + int strLen; + int uriLen; + int prefixLen; +} TAG_NAME; + +/* TAG represents an open element. + The name of the element is stored in both the document and API + encodings. The memory buffer 'buf' is a separately-allocated + memory area which stores the name. During the XML_Parse()/ + XMLParseBuffer() when the element is open, the memory for the 'raw' + version of the name (in the document encoding) is shared with the + document buffer. If the element is open across calls to + XML_Parse()/XML_ParseBuffer(), the buffer is re-allocated to + contain the 'raw' name as well. + + A parser re-uses these structures, maintaining a list of allocated + TAG objects in a free list. +*/ +typedef struct tag { + struct tag *parent; /* parent of this element */ + const char *rawName; /* tagName in the original encoding */ + int rawNameLength; + TAG_NAME name; /* tagName in the API encoding */ + char *buf; /* buffer for name components */ + char *bufEnd; /* end of the buffer */ + BINDING *bindings; +} TAG; + +typedef struct { + const XML_Char *name; + const XML_Char *textPtr; + int textLen; /* length in XML_Chars */ + int processed; /* # of processed bytes - when suspended */ + const XML_Char *systemId; + const XML_Char *base; + const XML_Char *publicId; + const XML_Char *notation; + XML_Bool open; + XML_Bool is_param; + XML_Bool is_internal; /* true if declared in internal subset outside PE */ +} ENTITY; + +typedef struct { + enum XML_Content_Type type; + enum XML_Content_Quant quant; + const XML_Char * name; + int firstchild; + int lastchild; + int childcnt; + int nextsib; +} CONTENT_SCAFFOLD; + +#define INIT_SCAFFOLD_ELEMENTS 32 + +typedef struct block { + struct block *next; + int size; + XML_Char s[1]; +} BLOCK; + +typedef struct { + BLOCK *blocks; + BLOCK *freeBlocks; + const XML_Char *end; + XML_Char *ptr; + XML_Char *start; + const XML_Memory_Handling_Suite *mem; +} STRING_POOL; + +/* The XML_Char before the name is used to determine whether + an attribute has been specified. */ +typedef struct attribute_id { + XML_Char *name; + PREFIX *prefix; + XML_Bool maybeTokenized; + XML_Bool xmlns; +} ATTRIBUTE_ID; + +typedef struct { + const ATTRIBUTE_ID *id; + XML_Bool isCdata; + const XML_Char *value; +} DEFAULT_ATTRIBUTE; + +typedef struct { + unsigned long version; + unsigned long hash; + const XML_Char *uriName; +} NS_ATT; + +typedef struct { + const XML_Char *name; + PREFIX *prefix; + const ATTRIBUTE_ID *idAtt; + int nDefaultAtts; + int allocDefaultAtts; + DEFAULT_ATTRIBUTE *defaultAtts; +} ELEMENT_TYPE; + +typedef struct { + HASH_TABLE generalEntities; + HASH_TABLE elementTypes; + HASH_TABLE attributeIds; + HASH_TABLE prefixes; + STRING_POOL pool; + STRING_POOL entityValuePool; + /* false once a parameter entity reference has been skipped */ + XML_Bool keepProcessing; + /* true once an internal or external PE reference has been encountered; + this includes the reference to an external subset */ + XML_Bool hasParamEntityRefs; + XML_Bool standalone; +#ifdef XML_DTD + /* indicates if external PE has been read */ + XML_Bool paramEntityRead; + HASH_TABLE paramEntities; +#endif /* XML_DTD */ + PREFIX defaultPrefix; + /* === scaffolding for building content model === */ + XML_Bool in_eldecl; + CONTENT_SCAFFOLD *scaffold; + unsigned contentStringLen; + unsigned scaffSize; + unsigned scaffCount; + int scaffLevel; + int *scaffIndex; +} DTD; + +typedef struct open_internal_entity { + const char *internalEventPtr; + const char *internalEventEndPtr; + struct open_internal_entity *next; + ENTITY *entity; + int startTagLevel; + XML_Bool betweenDecl; /* WFC: PE Between Declarations */ +} OPEN_INTERNAL_ENTITY; + +typedef enum XML_Error PTRCALL Processor(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr); + +static Processor prologProcessor; +static Processor prologInitProcessor; +static Processor contentProcessor; +static Processor cdataSectionProcessor; +#ifdef XML_DTD +static Processor ignoreSectionProcessor; +static Processor externalParEntProcessor; +static Processor externalParEntInitProcessor; +static Processor entityValueProcessor; +static Processor entityValueInitProcessor; +#endif /* XML_DTD */ +static Processor epilogProcessor; +static Processor errorProcessor; +static Processor externalEntityInitProcessor; +static Processor externalEntityInitProcessor2; +static Processor externalEntityInitProcessor3; +static Processor externalEntityContentProcessor; +static Processor internalEntityProcessor; + +static enum XML_Error +handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName); +static enum XML_Error +processXmlDecl(XML_Parser parser, int isGeneralTextEntity, + const char *s, const char *next); +static enum XML_Error +initializeEncoding(XML_Parser parser); +static enum XML_Error +doProlog(XML_Parser parser, const ENCODING *enc, const char *s, + const char *end, int tok, const char *next, const char **nextPtr, + XML_Bool haveMore); +static enum XML_Error +processInternalEntity(XML_Parser parser, ENTITY *entity, + XML_Bool betweenDecl); +static enum XML_Error +doContent(XML_Parser parser, int startTagLevel, const ENCODING *enc, + const char *start, const char *end, const char **endPtr, + XML_Bool haveMore); +static enum XML_Error +doCdataSection(XML_Parser parser, const ENCODING *, const char **startPtr, + const char *end, const char **nextPtr, XML_Bool haveMore); +#ifdef XML_DTD +static enum XML_Error +doIgnoreSection(XML_Parser parser, const ENCODING *, const char **startPtr, + const char *end, const char **nextPtr, XML_Bool haveMore); +#endif /* XML_DTD */ + +static enum XML_Error +storeAtts(XML_Parser parser, const ENCODING *, const char *s, + TAG_NAME *tagNamePtr, BINDING **bindingsPtr); +static enum XML_Error +addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId, + const XML_Char *uri, BINDING **bindingsPtr); +static int +defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *, XML_Bool isCdata, + XML_Bool isId, const XML_Char *dfltValue, XML_Parser parser); +static enum XML_Error +storeAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata, + const char *, const char *, STRING_POOL *); +static enum XML_Error +appendAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata, + const char *, const char *, STRING_POOL *); +static ATTRIBUTE_ID * +getAttributeId(XML_Parser parser, const ENCODING *enc, const char *start, + const char *end); +static int +setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *); +static enum XML_Error +storeEntityValue(XML_Parser parser, const ENCODING *enc, const char *start, + const char *end); +static int +reportProcessingInstruction(XML_Parser parser, const ENCODING *enc, + const char *start, const char *end); +static int +reportComment(XML_Parser parser, const ENCODING *enc, const char *start, + const char *end); +static void +reportDefault(XML_Parser parser, const ENCODING *enc, const char *start, + const char *end); + +static const XML_Char * getContext(XML_Parser parser); +static XML_Bool +setContext(XML_Parser parser, const XML_Char *context); + +static void FASTCALL normalizePublicId(XML_Char *s); + +static DTD * dtdCreate(const XML_Memory_Handling_Suite *ms); +/* do not call if parentParser != NULL */ +static void dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms); +static void +dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms); +static int +dtdCopy(XML_Parser oldParser, + DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms); +static int +copyEntityTable(XML_Parser oldParser, + HASH_TABLE *, STRING_POOL *, const HASH_TABLE *); +static NAMED * +lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize); +static void FASTCALL +hashTableInit(HASH_TABLE *, const XML_Memory_Handling_Suite *ms); +static void FASTCALL hashTableClear(HASH_TABLE *); +static void FASTCALL hashTableDestroy(HASH_TABLE *); +static void FASTCALL +hashTableIterInit(HASH_TABLE_ITER *, const HASH_TABLE *); +static NAMED * FASTCALL hashTableIterNext(HASH_TABLE_ITER *); + +static void FASTCALL +poolInit(STRING_POOL *, const XML_Memory_Handling_Suite *ms); +static void FASTCALL poolClear(STRING_POOL *); +static void FASTCALL poolDestroy(STRING_POOL *); +static XML_Char * +poolAppend(STRING_POOL *pool, const ENCODING *enc, + const char *ptr, const char *end); +static XML_Char * +poolStoreString(STRING_POOL *pool, const ENCODING *enc, + const char *ptr, const char *end); +static XML_Bool FASTCALL poolGrow(STRING_POOL *pool); +static const XML_Char * FASTCALL +poolCopyString(STRING_POOL *pool, const XML_Char *s); +static const XML_Char * +poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n); +static const XML_Char * FASTCALL +poolAppendString(STRING_POOL *pool, const XML_Char *s); + +static int FASTCALL nextScaffoldPart(XML_Parser parser); +static XML_Content * build_model(XML_Parser parser); +static ELEMENT_TYPE * +getElementType(XML_Parser parser, const ENCODING *enc, + const char *ptr, const char *end); + +static unsigned long generate_hash_secret_salt(void); +static XML_Bool startParsing(XML_Parser parser); + +static XML_Parser +parserCreate(const XML_Char *encodingName, + const XML_Memory_Handling_Suite *memsuite, + const XML_Char *nameSep, + DTD *dtd); + +static void +parserInit(XML_Parser parser, const XML_Char *encodingName); + +#define poolStart(pool) ((pool)->start) +#define poolEnd(pool) ((pool)->ptr) +#define poolLength(pool) ((pool)->ptr - (pool)->start) +#define poolChop(pool) ((void)--(pool->ptr)) +#define poolLastChar(pool) (((pool)->ptr)[-1]) +#define poolDiscard(pool) ((pool)->ptr = (pool)->start) +#define poolFinish(pool) ((pool)->start = (pool)->ptr) +#define poolAppendChar(pool, c) \ + (((pool)->ptr == (pool)->end && !poolGrow(pool)) \ + ? 0 \ + : ((*((pool)->ptr)++ = c), 1)) + +struct XML_ParserStruct { + /* The first member must be userData so that the XML_GetUserData + macro works. */ + void *m_userData; + void *m_handlerArg; + char *m_buffer; + const XML_Memory_Handling_Suite m_mem; + /* first character to be parsed */ + const char *m_bufferPtr; + /* past last character to be parsed */ + char *m_bufferEnd; + /* allocated end of buffer */ + const char *m_bufferLim; + XML_Index m_parseEndByteIndex; + const char *m_parseEndPtr; + XML_Char *m_dataBuf; + XML_Char *m_dataBufEnd; + XML_StartElementHandler m_startElementHandler; + XML_EndElementHandler m_endElementHandler; + XML_CharacterDataHandler m_characterDataHandler; + XML_ProcessingInstructionHandler m_processingInstructionHandler; + XML_CommentHandler m_commentHandler; + XML_StartCdataSectionHandler m_startCdataSectionHandler; + XML_EndCdataSectionHandler m_endCdataSectionHandler; + XML_DefaultHandler m_defaultHandler; + XML_StartDoctypeDeclHandler m_startDoctypeDeclHandler; + XML_EndDoctypeDeclHandler m_endDoctypeDeclHandler; + XML_UnparsedEntityDeclHandler m_unparsedEntityDeclHandler; + XML_NotationDeclHandler m_notationDeclHandler; + XML_StartNamespaceDeclHandler m_startNamespaceDeclHandler; + XML_EndNamespaceDeclHandler m_endNamespaceDeclHandler; + XML_NotStandaloneHandler m_notStandaloneHandler; + XML_ExternalEntityRefHandler m_externalEntityRefHandler; + XML_Parser m_externalEntityRefHandlerArg; + XML_SkippedEntityHandler m_skippedEntityHandler; + XML_UnknownEncodingHandler m_unknownEncodingHandler; + XML_ElementDeclHandler m_elementDeclHandler; + XML_AttlistDeclHandler m_attlistDeclHandler; + XML_EntityDeclHandler m_entityDeclHandler; + XML_XmlDeclHandler m_xmlDeclHandler; + const ENCODING *m_encoding; + INIT_ENCODING m_initEncoding; + const ENCODING *m_internalEncoding; + const XML_Char *m_protocolEncodingName; + XML_Bool m_ns; + XML_Bool m_ns_triplets; + void *m_unknownEncodingMem; + void *m_unknownEncodingData; + void *m_unknownEncodingHandlerData; + void (XMLCALL *m_unknownEncodingRelease)(void *); + PROLOG_STATE m_prologState; + Processor *m_processor; + enum XML_Error m_errorCode; + const char *m_eventPtr; + const char *m_eventEndPtr; + const char *m_positionPtr; + OPEN_INTERNAL_ENTITY *m_openInternalEntities; + OPEN_INTERNAL_ENTITY *m_freeInternalEntities; + XML_Bool m_defaultExpandInternalEntities; + int m_tagLevel; + ENTITY *m_declEntity; + const XML_Char *m_doctypeName; + const XML_Char *m_doctypeSysid; + const XML_Char *m_doctypePubid; + const XML_Char *m_declAttributeType; + const XML_Char *m_declNotationName; + const XML_Char *m_declNotationPublicId; + ELEMENT_TYPE *m_declElementType; + ATTRIBUTE_ID *m_declAttributeId; + XML_Bool m_declAttributeIsCdata; + XML_Bool m_declAttributeIsId; + DTD *m_dtd; + const XML_Char *m_curBase; + TAG *m_tagStack; + TAG *m_freeTagList; + BINDING *m_inheritedBindings; + BINDING *m_freeBindingList; + int m_attsSize; + int m_nSpecifiedAtts; + int m_idAttIndex; + ATTRIBUTE *m_atts; + NS_ATT *m_nsAtts; + unsigned long m_nsAttsVersion; + unsigned char m_nsAttsPower; +#ifdef XML_ATTR_INFO + XML_AttrInfo *m_attInfo; +#endif + POSITION m_position; + STRING_POOL m_tempPool; + STRING_POOL m_temp2Pool; + char *m_groupConnector; + unsigned int m_groupSize; + XML_Char m_namespaceSeparator; + XML_Parser m_parentParser; + XML_ParsingStatus m_parsingStatus; +#ifdef XML_DTD + XML_Bool m_isParamEntity; + XML_Bool m_useForeignDTD; + enum XML_ParamEntityParsing m_paramEntityParsing; +#endif + unsigned long m_hash_secret_salt; +}; + +#define MALLOC(s) (parser->m_mem.malloc_fcn((s))) +#define REALLOC(p,s) (parser->m_mem.realloc_fcn((p),(s))) +#define FREE(p) (parser->m_mem.free_fcn((p))) + +#define userData (parser->m_userData) +#define handlerArg (parser->m_handlerArg) +#define startElementHandler (parser->m_startElementHandler) +#define endElementHandler (parser->m_endElementHandler) +#define characterDataHandler (parser->m_characterDataHandler) +#define processingInstructionHandler \ + (parser->m_processingInstructionHandler) +#define commentHandler (parser->m_commentHandler) +#define startCdataSectionHandler \ + (parser->m_startCdataSectionHandler) +#define endCdataSectionHandler (parser->m_endCdataSectionHandler) +#define defaultHandler (parser->m_defaultHandler) +#define startDoctypeDeclHandler (parser->m_startDoctypeDeclHandler) +#define endDoctypeDeclHandler (parser->m_endDoctypeDeclHandler) +#define unparsedEntityDeclHandler \ + (parser->m_unparsedEntityDeclHandler) +#define notationDeclHandler (parser->m_notationDeclHandler) +#define startNamespaceDeclHandler \ + (parser->m_startNamespaceDeclHandler) +#define endNamespaceDeclHandler (parser->m_endNamespaceDeclHandler) +#define notStandaloneHandler (parser->m_notStandaloneHandler) +#define externalEntityRefHandler \ + (parser->m_externalEntityRefHandler) +#define externalEntityRefHandlerArg \ + (parser->m_externalEntityRefHandlerArg) +#define internalEntityRefHandler \ + (parser->m_internalEntityRefHandler) +#define skippedEntityHandler (parser->m_skippedEntityHandler) +#define unknownEncodingHandler (parser->m_unknownEncodingHandler) +#define elementDeclHandler (parser->m_elementDeclHandler) +#define attlistDeclHandler (parser->m_attlistDeclHandler) +#define entityDeclHandler (parser->m_entityDeclHandler) +#define xmlDeclHandler (parser->m_xmlDeclHandler) +#define encoding (parser->m_encoding) +#define initEncoding (parser->m_initEncoding) +#define internalEncoding (parser->m_internalEncoding) +#define unknownEncodingMem (parser->m_unknownEncodingMem) +#define unknownEncodingData (parser->m_unknownEncodingData) +#define unknownEncodingHandlerData \ + (parser->m_unknownEncodingHandlerData) +#define unknownEncodingRelease (parser->m_unknownEncodingRelease) +#define protocolEncodingName (parser->m_protocolEncodingName) +#define ns (parser->m_ns) +#define ns_triplets (parser->m_ns_triplets) +#define prologState (parser->m_prologState) +#define processor (parser->m_processor) +#define errorCode (parser->m_errorCode) +#define eventPtr (parser->m_eventPtr) +#define eventEndPtr (parser->m_eventEndPtr) +#define positionPtr (parser->m_positionPtr) +#define position (parser->m_position) +#define openInternalEntities (parser->m_openInternalEntities) +#define freeInternalEntities (parser->m_freeInternalEntities) +#define defaultExpandInternalEntities \ + (parser->m_defaultExpandInternalEntities) +#define tagLevel (parser->m_tagLevel) +#define buffer (parser->m_buffer) +#define bufferPtr (parser->m_bufferPtr) +#define bufferEnd (parser->m_bufferEnd) +#define parseEndByteIndex (parser->m_parseEndByteIndex) +#define parseEndPtr (parser->m_parseEndPtr) +#define bufferLim (parser->m_bufferLim) +#define dataBuf (parser->m_dataBuf) +#define dataBufEnd (parser->m_dataBufEnd) +#define _dtd (parser->m_dtd) +#define curBase (parser->m_curBase) +#define declEntity (parser->m_declEntity) +#define doctypeName (parser->m_doctypeName) +#define doctypeSysid (parser->m_doctypeSysid) +#define doctypePubid (parser->m_doctypePubid) +#define declAttributeType (parser->m_declAttributeType) +#define declNotationName (parser->m_declNotationName) +#define declNotationPublicId (parser->m_declNotationPublicId) +#define declElementType (parser->m_declElementType) +#define declAttributeId (parser->m_declAttributeId) +#define declAttributeIsCdata (parser->m_declAttributeIsCdata) +#define declAttributeIsId (parser->m_declAttributeIsId) +#define freeTagList (parser->m_freeTagList) +#define freeBindingList (parser->m_freeBindingList) +#define inheritedBindings (parser->m_inheritedBindings) +#define tagStack (parser->m_tagStack) +#define atts (parser->m_atts) +#define attsSize (parser->m_attsSize) +#define nSpecifiedAtts (parser->m_nSpecifiedAtts) +#define idAttIndex (parser->m_idAttIndex) +#define nsAtts (parser->m_nsAtts) +#define nsAttsVersion (parser->m_nsAttsVersion) +#define nsAttsPower (parser->m_nsAttsPower) +#define attInfo (parser->m_attInfo) +#define tempPool (parser->m_tempPool) +#define temp2Pool (parser->m_temp2Pool) +#define groupConnector (parser->m_groupConnector) +#define groupSize (parser->m_groupSize) +#define namespaceSeparator (parser->m_namespaceSeparator) +#define parentParser (parser->m_parentParser) +#define ps_parsing (parser->m_parsingStatus.parsing) +#define ps_finalBuffer (parser->m_parsingStatus.finalBuffer) +#ifdef XML_DTD +#define isParamEntity (parser->m_isParamEntity) +#define useForeignDTD (parser->m_useForeignDTD) +#define paramEntityParsing (parser->m_paramEntityParsing) +#endif /* XML_DTD */ +#define hash_secret_salt (parser->m_hash_secret_salt) + +XML_Parser XMLCALL +XML_ParserCreate(const XML_Char *encodingName) +{ + return XML_ParserCreate_MM(encodingName, NULL, NULL); +} + +XML_Parser XMLCALL +XML_ParserCreateNS(const XML_Char *encodingName, XML_Char nsSep) +{ + XML_Char tmp[2]; + *tmp = nsSep; + return XML_ParserCreate_MM(encodingName, NULL, tmp); +} + +static const XML_Char implicitContext[] = { + ASCII_x, ASCII_m, ASCII_l, ASCII_EQUALS, ASCII_h, ASCII_t, ASCII_t, ASCII_p, + ASCII_COLON, ASCII_SLASH, ASCII_SLASH, ASCII_w, ASCII_w, ASCII_w, + ASCII_PERIOD, ASCII_w, ASCII_3, ASCII_PERIOD, ASCII_o, ASCII_r, ASCII_g, + ASCII_SLASH, ASCII_X, ASCII_M, ASCII_L, ASCII_SLASH, ASCII_1, ASCII_9, + ASCII_9, ASCII_8, ASCII_SLASH, ASCII_n, ASCII_a, ASCII_m, ASCII_e, + ASCII_s, ASCII_p, ASCII_a, ASCII_c, ASCII_e, '\0' +}; + +static unsigned long +generate_hash_secret_salt(void) +{ + unsigned int seed = time(NULL) % UINT_MAX; + srand(seed); + return rand(); +} + +static XML_Bool /* only valid for root parser */ +startParsing(XML_Parser parser) +{ + /* hash functions must be initialized before setContext() is called */ + if (hash_secret_salt == 0) + hash_secret_salt = generate_hash_secret_salt(); + if (ns) { + /* implicit context only set for root parser, since child + parsers (i.e. external entity parsers) will inherit it + */ + return setContext(parser, implicitContext); + } + return XML_TRUE; +} + +XML_Parser XMLCALL +XML_ParserCreate_MM(const XML_Char *encodingName, + const XML_Memory_Handling_Suite *memsuite, + const XML_Char *nameSep) +{ + return parserCreate(encodingName, memsuite, nameSep, NULL); +} + +static XML_Parser +parserCreate(const XML_Char *encodingName, + const XML_Memory_Handling_Suite *memsuite, + const XML_Char *nameSep, + DTD *dtd) +{ + XML_Parser parser; + + if (memsuite) { + XML_Memory_Handling_Suite *mtemp; + parser = (XML_Parser) + memsuite->malloc_fcn(sizeof(struct XML_ParserStruct)); + if (parser != NULL) { + mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem); + mtemp->malloc_fcn = memsuite->malloc_fcn; + mtemp->realloc_fcn = memsuite->realloc_fcn; + mtemp->free_fcn = memsuite->free_fcn; + } + } + else { + XML_Memory_Handling_Suite *mtemp; + parser = (XML_Parser)malloc(sizeof(struct XML_ParserStruct)); + if (parser != NULL) { + mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem); + mtemp->malloc_fcn = malloc; + mtemp->realloc_fcn = realloc; + mtemp->free_fcn = free; + } + } + + if (!parser) + return parser; + + buffer = NULL; + bufferLim = NULL; + + attsSize = INIT_ATTS_SIZE; + atts = (ATTRIBUTE *)MALLOC(attsSize * sizeof(ATTRIBUTE)); + if (atts == NULL) { + FREE(parser); + return NULL; + } +#ifdef XML_ATTR_INFO + attInfo = (XML_AttrInfo*)MALLOC(attsSize * sizeof(XML_AttrInfo)); + if (attInfo == NULL) { + FREE(atts); + FREE(parser); + return NULL; + } +#endif + dataBuf = (XML_Char *)MALLOC(INIT_DATA_BUF_SIZE * sizeof(XML_Char)); + if (dataBuf == NULL) { + FREE(atts); +#ifdef XML_ATTR_INFO + FREE(attInfo); +#endif + FREE(parser); + return NULL; + } + dataBufEnd = dataBuf + INIT_DATA_BUF_SIZE; + + if (dtd) + _dtd = dtd; + else { + _dtd = dtdCreate(&parser->m_mem); + if (_dtd == NULL) { + FREE(dataBuf); + FREE(atts); +#ifdef XML_ATTR_INFO + FREE(attInfo); +#endif + FREE(parser); + return NULL; + } + } + + freeBindingList = NULL; + freeTagList = NULL; + freeInternalEntities = NULL; + + groupSize = 0; + groupConnector = NULL; + + unknownEncodingHandler = NULL; + unknownEncodingHandlerData = NULL; + + namespaceSeparator = ASCII_EXCL; + ns = XML_FALSE; + ns_triplets = XML_FALSE; + + nsAtts = NULL; + nsAttsVersion = 0; + nsAttsPower = 0; + + poolInit(&tempPool, &(parser->m_mem)); + poolInit(&temp2Pool, &(parser->m_mem)); + parserInit(parser, encodingName); + + if (encodingName && !protocolEncodingName) { + XML_ParserFree(parser); + return NULL; + } + + if (nameSep) { + ns = XML_TRUE; + internalEncoding = XmlGetInternalEncodingNS(); + namespaceSeparator = *nameSep; + } + else { + internalEncoding = XmlGetInternalEncoding(); + } + + return parser; +} + +static void +parserInit(XML_Parser parser, const XML_Char *encodingName) +{ + processor = prologInitProcessor; + XmlPrologStateInit(&prologState); + protocolEncodingName = (encodingName != NULL + ? poolCopyString(&tempPool, encodingName) + : NULL); + curBase = NULL; + XmlInitEncoding(&initEncoding, &encoding, 0); + userData = NULL; + handlerArg = NULL; + startElementHandler = NULL; + endElementHandler = NULL; + characterDataHandler = NULL; + processingInstructionHandler = NULL; + commentHandler = NULL; + startCdataSectionHandler = NULL; + endCdataSectionHandler = NULL; + defaultHandler = NULL; + startDoctypeDeclHandler = NULL; + endDoctypeDeclHandler = NULL; + unparsedEntityDeclHandler = NULL; + notationDeclHandler = NULL; + startNamespaceDeclHandler = NULL; + endNamespaceDeclHandler = NULL; + notStandaloneHandler = NULL; + externalEntityRefHandler = NULL; + externalEntityRefHandlerArg = parser; + skippedEntityHandler = NULL; + elementDeclHandler = NULL; + attlistDeclHandler = NULL; + entityDeclHandler = NULL; + xmlDeclHandler = NULL; + bufferPtr = buffer; + bufferEnd = buffer; + parseEndByteIndex = 0; + parseEndPtr = NULL; + declElementType = NULL; + declAttributeId = NULL; + declEntity = NULL; + doctypeName = NULL; + doctypeSysid = NULL; + doctypePubid = NULL; + declAttributeType = NULL; + declNotationName = NULL; + declNotationPublicId = NULL; + declAttributeIsCdata = XML_FALSE; + declAttributeIsId = XML_FALSE; + memset(&position, 0, sizeof(POSITION)); + errorCode = XML_ERROR_NONE; + eventPtr = NULL; + eventEndPtr = NULL; + positionPtr = NULL; + openInternalEntities = NULL; + defaultExpandInternalEntities = XML_TRUE; + tagLevel = 0; + tagStack = NULL; + inheritedBindings = NULL; + nSpecifiedAtts = 0; + unknownEncodingMem = NULL; + unknownEncodingRelease = NULL; + unknownEncodingData = NULL; + parentParser = NULL; + ps_parsing = XML_INITIALIZED; +#ifdef XML_DTD + isParamEntity = XML_FALSE; + useForeignDTD = XML_FALSE; + paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER; +#endif + hash_secret_salt = 0; +} + +/* moves list of bindings to freeBindingList */ +static void FASTCALL +moveToFreeBindingList(XML_Parser parser, BINDING *bindings) +{ + while (bindings) { + BINDING *b = bindings; + bindings = bindings->nextTagBinding; + b->nextTagBinding = freeBindingList; + freeBindingList = b; + } +} + +XML_Bool XMLCALL +XML_ParserReset(XML_Parser parser, const XML_Char *encodingName) +{ + TAG *tStk; + OPEN_INTERNAL_ENTITY *openEntityList; + if (parentParser) + return XML_FALSE; + /* move tagStack to freeTagList */ + tStk = tagStack; + while (tStk) { + TAG *tag = tStk; + tStk = tStk->parent; + tag->parent = freeTagList; + moveToFreeBindingList(parser, tag->bindings); + tag->bindings = NULL; + freeTagList = tag; + } + /* move openInternalEntities to freeInternalEntities */ + openEntityList = openInternalEntities; + while (openEntityList) { + OPEN_INTERNAL_ENTITY *openEntity = openEntityList; + openEntityList = openEntity->next; + openEntity->next = freeInternalEntities; + freeInternalEntities = openEntity; + } + moveToFreeBindingList(parser, inheritedBindings); + FREE(unknownEncodingMem); + if (unknownEncodingRelease) + unknownEncodingRelease(unknownEncodingData); + poolClear(&tempPool); + poolClear(&temp2Pool); + parserInit(parser, encodingName); + dtdReset(_dtd, &parser->m_mem); + return XML_TRUE; +} + +enum XML_Status XMLCALL +XML_SetEncoding(XML_Parser parser, const XML_Char *encodingName) +{ + /* Block after XML_Parse()/XML_ParseBuffer() has been called. + XXX There's no way for the caller to determine which of the + XXX possible error cases caused the XML_STATUS_ERROR return. + */ + if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) + return XML_STATUS_ERROR; + if (encodingName == NULL) + protocolEncodingName = NULL; + else { + protocolEncodingName = poolCopyString(&tempPool, encodingName); + if (!protocolEncodingName) + return XML_STATUS_ERROR; + } + return XML_STATUS_OK; +} + +XML_Parser XMLCALL +XML_ExternalEntityParserCreate(XML_Parser oldParser, + const XML_Char *context, + const XML_Char *encodingName) +{ + XML_Parser parser = oldParser; + DTD *newDtd = NULL; + DTD *oldDtd = _dtd; + XML_StartElementHandler oldStartElementHandler = startElementHandler; + XML_EndElementHandler oldEndElementHandler = endElementHandler; + XML_CharacterDataHandler oldCharacterDataHandler = characterDataHandler; + XML_ProcessingInstructionHandler oldProcessingInstructionHandler + = processingInstructionHandler; + XML_CommentHandler oldCommentHandler = commentHandler; + XML_StartCdataSectionHandler oldStartCdataSectionHandler + = startCdataSectionHandler; + XML_EndCdataSectionHandler oldEndCdataSectionHandler + = endCdataSectionHandler; + XML_DefaultHandler oldDefaultHandler = defaultHandler; + XML_UnparsedEntityDeclHandler oldUnparsedEntityDeclHandler + = unparsedEntityDeclHandler; + XML_NotationDeclHandler oldNotationDeclHandler = notationDeclHandler; + XML_StartNamespaceDeclHandler oldStartNamespaceDeclHandler + = startNamespaceDeclHandler; + XML_EndNamespaceDeclHandler oldEndNamespaceDeclHandler + = endNamespaceDeclHandler; + XML_NotStandaloneHandler oldNotStandaloneHandler = notStandaloneHandler; + XML_ExternalEntityRefHandler oldExternalEntityRefHandler + = externalEntityRefHandler; + XML_SkippedEntityHandler oldSkippedEntityHandler = skippedEntityHandler; + XML_UnknownEncodingHandler oldUnknownEncodingHandler + = unknownEncodingHandler; + XML_ElementDeclHandler oldElementDeclHandler = elementDeclHandler; + XML_AttlistDeclHandler oldAttlistDeclHandler = attlistDeclHandler; + XML_EntityDeclHandler oldEntityDeclHandler = entityDeclHandler; + XML_XmlDeclHandler oldXmlDeclHandler = xmlDeclHandler; + ELEMENT_TYPE * oldDeclElementType = declElementType; + + void *oldUserData = userData; + void *oldHandlerArg = handlerArg; + XML_Bool oldDefaultExpandInternalEntities = defaultExpandInternalEntities; + XML_Parser oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg; +#ifdef XML_DTD + enum XML_ParamEntityParsing oldParamEntityParsing = paramEntityParsing; + int oldInEntityValue = prologState.inEntityValue; +#endif + XML_Bool oldns_triplets = ns_triplets; + /* Note that the new parser shares the same hash secret as the old + parser, so that dtdCopy and copyEntityTable can lookup values + from hash tables associated with either parser without us having + to worry which hash secrets each table has. + */ + unsigned long oldhash_secret_salt = hash_secret_salt; + +#ifdef XML_DTD + if (!context) + newDtd = oldDtd; +#endif /* XML_DTD */ + + /* Note that the magical uses of the pre-processor to make field + access look more like C++ require that `parser' be overwritten + here. This makes this function more painful to follow than it + would be otherwise. + */ + if (ns) { + XML_Char tmp[2]; + *tmp = namespaceSeparator; + parser = parserCreate(encodingName, &parser->m_mem, tmp, newDtd); + } + else { + parser = parserCreate(encodingName, &parser->m_mem, NULL, newDtd); + } + + if (!parser) + return NULL; + + startElementHandler = oldStartElementHandler; + endElementHandler = oldEndElementHandler; + characterDataHandler = oldCharacterDataHandler; + processingInstructionHandler = oldProcessingInstructionHandler; + commentHandler = oldCommentHandler; + startCdataSectionHandler = oldStartCdataSectionHandler; + endCdataSectionHandler = oldEndCdataSectionHandler; + defaultHandler = oldDefaultHandler; + unparsedEntityDeclHandler = oldUnparsedEntityDeclHandler; + notationDeclHandler = oldNotationDeclHandler; + startNamespaceDeclHandler = oldStartNamespaceDeclHandler; + endNamespaceDeclHandler = oldEndNamespaceDeclHandler; + notStandaloneHandler = oldNotStandaloneHandler; + externalEntityRefHandler = oldExternalEntityRefHandler; + skippedEntityHandler = oldSkippedEntityHandler; + unknownEncodingHandler = oldUnknownEncodingHandler; + elementDeclHandler = oldElementDeclHandler; + attlistDeclHandler = oldAttlistDeclHandler; + entityDeclHandler = oldEntityDeclHandler; + xmlDeclHandler = oldXmlDeclHandler; + declElementType = oldDeclElementType; + userData = oldUserData; + if (oldUserData == oldHandlerArg) + handlerArg = userData; + else + handlerArg = parser; + if (oldExternalEntityRefHandlerArg != oldParser) + externalEntityRefHandlerArg = oldExternalEntityRefHandlerArg; + defaultExpandInternalEntities = oldDefaultExpandInternalEntities; + ns_triplets = oldns_triplets; + hash_secret_salt = oldhash_secret_salt; + parentParser = oldParser; +#ifdef XML_DTD + paramEntityParsing = oldParamEntityParsing; + prologState.inEntityValue = oldInEntityValue; + if (context) { +#endif /* XML_DTD */ + if (!dtdCopy(oldParser, _dtd, oldDtd, &parser->m_mem) + || !setContext(parser, context)) { + XML_ParserFree(parser); + return NULL; + } + processor = externalEntityInitProcessor; +#ifdef XML_DTD + } + else { + /* The DTD instance referenced by _dtd is shared between the document's + root parser and external PE parsers, therefore one does not need to + call setContext. In addition, one also *must* not call setContext, + because this would overwrite existing prefix->binding pointers in + _dtd with ones that get destroyed with the external PE parser. + This would leave those prefixes with dangling pointers. + */ + isParamEntity = XML_TRUE; + XmlPrologStateInitExternalEntity(&prologState); + processor = externalParEntInitProcessor; + } +#endif /* XML_DTD */ + return parser; +} + +static void FASTCALL +destroyBindings(BINDING *bindings, XML_Parser parser) +{ + for (;;) { + BINDING *b = bindings; + if (!b) + break; + bindings = b->nextTagBinding; + FREE(b->uri); + FREE(b); + } +} + +void XMLCALL +XML_ParserFree(XML_Parser parser) +{ + TAG *tagList; + OPEN_INTERNAL_ENTITY *entityList; + if (parser == NULL) + return; + /* free tagStack and freeTagList */ + tagList = tagStack; + for (;;) { + TAG *p; + if (tagList == NULL) { + if (freeTagList == NULL) + break; + tagList = freeTagList; + freeTagList = NULL; + } + p = tagList; + tagList = tagList->parent; + FREE(p->buf); + destroyBindings(p->bindings, parser); + FREE(p); + } + /* free openInternalEntities and freeInternalEntities */ + entityList = openInternalEntities; + for (;;) { + OPEN_INTERNAL_ENTITY *openEntity; + if (entityList == NULL) { + if (freeInternalEntities == NULL) + break; + entityList = freeInternalEntities; + freeInternalEntities = NULL; + } + openEntity = entityList; + entityList = entityList->next; + FREE(openEntity); + } + + destroyBindings(freeBindingList, parser); + destroyBindings(inheritedBindings, parser); + poolDestroy(&tempPool); + poolDestroy(&temp2Pool); +#ifdef XML_DTD + /* external parameter entity parsers share the DTD structure + parser->m_dtd with the root parser, so we must not destroy it + */ + if (!isParamEntity && _dtd) +#else + if (_dtd) +#endif /* XML_DTD */ + dtdDestroy(_dtd, (XML_Bool)!parentParser, &parser->m_mem); + FREE((void *)atts); +#ifdef XML_ATTR_INFO + FREE((void *)attInfo); +#endif + FREE(groupConnector); + FREE(buffer); + FREE(dataBuf); + FREE(nsAtts); + FREE(unknownEncodingMem); + if (unknownEncodingRelease) + unknownEncodingRelease(unknownEncodingData); + FREE(parser); +} + +void XMLCALL +XML_UseParserAsHandlerArg(XML_Parser parser) +{ + handlerArg = parser; +} + +enum XML_Error XMLCALL +XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD) +{ +#ifdef XML_DTD + /* block after XML_Parse()/XML_ParseBuffer() has been called */ + if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) + return XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING; + useForeignDTD = useDTD; + return XML_ERROR_NONE; +#else + return XML_ERROR_FEATURE_REQUIRES_XML_DTD; +#endif +} + +void XMLCALL +XML_SetReturnNSTriplet(XML_Parser parser, int do_nst) +{ + /* block after XML_Parse()/XML_ParseBuffer() has been called */ + if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) + return; + ns_triplets = do_nst ? XML_TRUE : XML_FALSE; +} + +void XMLCALL +XML_SetUserData(XML_Parser parser, void *p) +{ + if (handlerArg == userData) + handlerArg = userData = p; + else + userData = p; +} + +enum XML_Status XMLCALL +XML_SetBase(XML_Parser parser, const XML_Char *p) +{ + if (p) { + p = poolCopyString(&_dtd->pool, p); + if (!p) + return XML_STATUS_ERROR; + curBase = p; + } + else + curBase = NULL; + return XML_STATUS_OK; +} + +const XML_Char * XMLCALL +XML_GetBase(XML_Parser parser) +{ + return curBase; +} + +int XMLCALL +XML_GetSpecifiedAttributeCount(XML_Parser parser) +{ + return nSpecifiedAtts; +} + +int XMLCALL +XML_GetIdAttributeIndex(XML_Parser parser) +{ + return idAttIndex; +} + +#ifdef XML_ATTR_INFO +const XML_AttrInfo * XMLCALL +XML_GetAttributeInfo(XML_Parser parser) +{ + return attInfo; +} +#endif + +void XMLCALL +XML_SetElementHandler(XML_Parser parser, + XML_StartElementHandler start, + XML_EndElementHandler end) +{ + startElementHandler = start; + endElementHandler = end; +} + +void XMLCALL +XML_SetStartElementHandler(XML_Parser parser, + XML_StartElementHandler start) { + startElementHandler = start; +} + +void XMLCALL +XML_SetEndElementHandler(XML_Parser parser, + XML_EndElementHandler end) { + endElementHandler = end; +} + +void XMLCALL +XML_SetCharacterDataHandler(XML_Parser parser, + XML_CharacterDataHandler handler) +{ + characterDataHandler = handler; +} + +void XMLCALL +XML_SetProcessingInstructionHandler(XML_Parser parser, + XML_ProcessingInstructionHandler handler) +{ + processingInstructionHandler = handler; +} + +void XMLCALL +XML_SetCommentHandler(XML_Parser parser, + XML_CommentHandler handler) +{ + commentHandler = handler; +} + +void XMLCALL +XML_SetCdataSectionHandler(XML_Parser parser, + XML_StartCdataSectionHandler start, + XML_EndCdataSectionHandler end) +{ + startCdataSectionHandler = start; + endCdataSectionHandler = end; +} + +void XMLCALL +XML_SetStartCdataSectionHandler(XML_Parser parser, + XML_StartCdataSectionHandler start) { + startCdataSectionHandler = start; +} + +void XMLCALL +XML_SetEndCdataSectionHandler(XML_Parser parser, + XML_EndCdataSectionHandler end) { + endCdataSectionHandler = end; +} + +void XMLCALL +XML_SetDefaultHandler(XML_Parser parser, + XML_DefaultHandler handler) +{ + defaultHandler = handler; + defaultExpandInternalEntities = XML_FALSE; +} + +void XMLCALL +XML_SetDefaultHandlerExpand(XML_Parser parser, + XML_DefaultHandler handler) +{ + defaultHandler = handler; + defaultExpandInternalEntities = XML_TRUE; +} + +void XMLCALL +XML_SetDoctypeDeclHandler(XML_Parser parser, + XML_StartDoctypeDeclHandler start, + XML_EndDoctypeDeclHandler end) +{ + startDoctypeDeclHandler = start; + endDoctypeDeclHandler = end; +} + +void XMLCALL +XML_SetStartDoctypeDeclHandler(XML_Parser parser, + XML_StartDoctypeDeclHandler start) { + startDoctypeDeclHandler = start; +} + +void XMLCALL +XML_SetEndDoctypeDeclHandler(XML_Parser parser, + XML_EndDoctypeDeclHandler end) { + endDoctypeDeclHandler = end; +} + +void XMLCALL +XML_SetUnparsedEntityDeclHandler(XML_Parser parser, + XML_UnparsedEntityDeclHandler handler) +{ + unparsedEntityDeclHandler = handler; +} + +void XMLCALL +XML_SetNotationDeclHandler(XML_Parser parser, + XML_NotationDeclHandler handler) +{ + notationDeclHandler = handler; +} + +void XMLCALL +XML_SetNamespaceDeclHandler(XML_Parser parser, + XML_StartNamespaceDeclHandler start, + XML_EndNamespaceDeclHandler end) +{ + startNamespaceDeclHandler = start; + endNamespaceDeclHandler = end; +} + +void XMLCALL +XML_SetStartNamespaceDeclHandler(XML_Parser parser, + XML_StartNamespaceDeclHandler start) { + startNamespaceDeclHandler = start; +} + +void XMLCALL +XML_SetEndNamespaceDeclHandler(XML_Parser parser, + XML_EndNamespaceDeclHandler end) { + endNamespaceDeclHandler = end; +} + +void XMLCALL +XML_SetNotStandaloneHandler(XML_Parser parser, + XML_NotStandaloneHandler handler) +{ + notStandaloneHandler = handler; +} + +void XMLCALL +XML_SetExternalEntityRefHandler(XML_Parser parser, + XML_ExternalEntityRefHandler handler) +{ + externalEntityRefHandler = handler; +} + +void XMLCALL +XML_SetExternalEntityRefHandlerArg(XML_Parser parser, void *arg) +{ + if (arg) + externalEntityRefHandlerArg = (XML_Parser)arg; + else + externalEntityRefHandlerArg = parser; +} + +void XMLCALL +XML_SetSkippedEntityHandler(XML_Parser parser, + XML_SkippedEntityHandler handler) +{ + skippedEntityHandler = handler; +} + +void XMLCALL +XML_SetUnknownEncodingHandler(XML_Parser parser, + XML_UnknownEncodingHandler handler, + void *data) +{ + unknownEncodingHandler = handler; + unknownEncodingHandlerData = data; +} + +void XMLCALL +XML_SetElementDeclHandler(XML_Parser parser, + XML_ElementDeclHandler eldecl) +{ + elementDeclHandler = eldecl; +} + +void XMLCALL +XML_SetAttlistDeclHandler(XML_Parser parser, + XML_AttlistDeclHandler attdecl) +{ + attlistDeclHandler = attdecl; +} + +void XMLCALL +XML_SetEntityDeclHandler(XML_Parser parser, + XML_EntityDeclHandler handler) +{ + entityDeclHandler = handler; +} + +void XMLCALL +XML_SetXmlDeclHandler(XML_Parser parser, + XML_XmlDeclHandler handler) { + xmlDeclHandler = handler; +} + +int XMLCALL +XML_SetParamEntityParsing(XML_Parser parser, + enum XML_ParamEntityParsing peParsing) +{ + /* block after XML_Parse()/XML_ParseBuffer() has been called */ + if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) + return 0; +#ifdef XML_DTD + paramEntityParsing = peParsing; + return 1; +#else + return peParsing == XML_PARAM_ENTITY_PARSING_NEVER; +#endif +} + +int XMLCALL +XML_SetHashSalt(XML_Parser parser, + unsigned long hash_salt) +{ + /* block after XML_Parse()/XML_ParseBuffer() has been called */ + if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) + return 0; + hash_secret_salt = hash_salt; + return 1; +} + +enum XML_Status XMLCALL +XML_Parse(XML_Parser parser, const char *s, int len, int isFinal) +{ + switch (ps_parsing) { + case XML_SUSPENDED: + errorCode = XML_ERROR_SUSPENDED; + return XML_STATUS_ERROR; + case XML_FINISHED: + errorCode = XML_ERROR_FINISHED; + return XML_STATUS_ERROR; + case XML_INITIALIZED: + if (parentParser == NULL && !startParsing(parser)) { + errorCode = XML_ERROR_NO_MEMORY; + return XML_STATUS_ERROR; + } + default: + ps_parsing = XML_PARSING; + } + + if (len == 0) { + ps_finalBuffer = (XML_Bool)isFinal; + if (!isFinal) + return XML_STATUS_OK; + positionPtr = bufferPtr; + parseEndPtr = bufferEnd; + + /* If data are left over from last buffer, and we now know that these + data are the final chunk of input, then we have to check them again + to detect errors based on that fact. + */ + errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr); + + if (errorCode == XML_ERROR_NONE) { + switch (ps_parsing) { + case XML_SUSPENDED: + XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position); + positionPtr = bufferPtr; + return XML_STATUS_SUSPENDED; + case XML_INITIALIZED: + case XML_PARSING: + ps_parsing = XML_FINISHED; + /* fall through */ + default: + return XML_STATUS_OK; + } + } + eventEndPtr = eventPtr; + processor = errorProcessor; + return XML_STATUS_ERROR; + } +#ifndef XML_CONTEXT_BYTES + else if (bufferPtr == bufferEnd) { + const char *end; + int nLeftOver; + enum XML_Error result; + parseEndByteIndex += len; + positionPtr = s; + ps_finalBuffer = (XML_Bool)isFinal; + + errorCode = processor(parser, s, parseEndPtr = s + len, &end); + + if (errorCode != XML_ERROR_NONE) { + eventEndPtr = eventPtr; + processor = errorProcessor; + return XML_STATUS_ERROR; + } + else { + switch (ps_parsing) { + case XML_SUSPENDED: + result = XML_STATUS_SUSPENDED; + break; + case XML_INITIALIZED: + case XML_PARSING: + if (isFinal) { + ps_parsing = XML_FINISHED; + return XML_STATUS_OK; + } + /* fall through */ + default: + result = XML_STATUS_OK; + } + } + + XmlUpdatePosition(encoding, positionPtr, end, &position); + nLeftOver = s + len - end; + if (nLeftOver) { + if (buffer == NULL || nLeftOver > bufferLim - buffer) { + /* FIXME avoid integer overflow */ + char *temp; + temp = (buffer == NULL + ? (char *)MALLOC(len * 2) + : (char *)REALLOC(buffer, len * 2)); + if (temp == NULL) { + errorCode = XML_ERROR_NO_MEMORY; + eventPtr = eventEndPtr = NULL; + processor = errorProcessor; + return XML_STATUS_ERROR; + } + buffer = temp; + bufferLim = buffer + len * 2; + } + memcpy(buffer, end, nLeftOver); + } + bufferPtr = buffer; + bufferEnd = buffer + nLeftOver; + positionPtr = bufferPtr; + parseEndPtr = bufferEnd; + eventPtr = bufferPtr; + eventEndPtr = bufferPtr; + return result; + } +#endif /* not defined XML_CONTEXT_BYTES */ + else { + void *buff = XML_GetBuffer(parser, len); + if (buff == NULL) + return XML_STATUS_ERROR; + else { + memcpy(buff, s, len); + return XML_ParseBuffer(parser, len, isFinal); + } + } +} + +enum XML_Status XMLCALL +XML_ParseBuffer(XML_Parser parser, int len, int isFinal) +{ + const char *start; + enum XML_Status result = XML_STATUS_OK; + + switch (ps_parsing) { + case XML_SUSPENDED: + errorCode = XML_ERROR_SUSPENDED; + return XML_STATUS_ERROR; + case XML_FINISHED: + errorCode = XML_ERROR_FINISHED; + return XML_STATUS_ERROR; + case XML_INITIALIZED: + if (parentParser == NULL && !startParsing(parser)) { + errorCode = XML_ERROR_NO_MEMORY; + return XML_STATUS_ERROR; + } + default: + ps_parsing = XML_PARSING; + } + + start = bufferPtr; + positionPtr = start; + bufferEnd += len; + parseEndPtr = bufferEnd; + parseEndByteIndex += len; + ps_finalBuffer = (XML_Bool)isFinal; + + errorCode = processor(parser, start, parseEndPtr, &bufferPtr); + + if (errorCode != XML_ERROR_NONE) { + eventEndPtr = eventPtr; + processor = errorProcessor; + return XML_STATUS_ERROR; + } + else { + switch (ps_parsing) { + case XML_SUSPENDED: + result = XML_STATUS_SUSPENDED; + break; + case XML_INITIALIZED: + case XML_PARSING: + if (isFinal) { + ps_parsing = XML_FINISHED; + return result; + } + default: ; /* should not happen */ + } + } + + XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position); + positionPtr = bufferPtr; + return result; +} + +void * XMLCALL +XML_GetBuffer(XML_Parser parser, int len) +{ + switch (ps_parsing) { + case XML_SUSPENDED: + errorCode = XML_ERROR_SUSPENDED; + return NULL; + case XML_FINISHED: + errorCode = XML_ERROR_FINISHED; + return NULL; + default: ; + } + + if (len > bufferLim - bufferEnd) { + /* FIXME avoid integer overflow */ + int neededSize = len + (int)(bufferEnd - bufferPtr); +#ifdef XML_CONTEXT_BYTES + int keep = (int)(bufferPtr - buffer); + + if (keep > XML_CONTEXT_BYTES) + keep = XML_CONTEXT_BYTES; + neededSize += keep; +#endif /* defined XML_CONTEXT_BYTES */ + if (neededSize <= bufferLim - buffer) { +#ifdef XML_CONTEXT_BYTES + if (keep < bufferPtr - buffer) { + int offset = (int)(bufferPtr - buffer) - keep; + memmove(buffer, &buffer[offset], bufferEnd - bufferPtr + keep); + bufferEnd -= offset; + bufferPtr -= offset; + } +#else + memmove(buffer, bufferPtr, bufferEnd - bufferPtr); + bufferEnd = buffer + (bufferEnd - bufferPtr); + bufferPtr = buffer; +#endif /* not defined XML_CONTEXT_BYTES */ + } + else { + char *newBuf; + int bufferSize = (int)(bufferLim - bufferPtr); + if (bufferSize == 0) + bufferSize = INIT_BUFFER_SIZE; + do { + bufferSize *= 2; + } while (bufferSize < neededSize); + newBuf = (char *)MALLOC(bufferSize); + if (newBuf == 0) { + errorCode = XML_ERROR_NO_MEMORY; + return NULL; + } + bufferLim = newBuf + bufferSize; +#ifdef XML_CONTEXT_BYTES + if (bufferPtr) { + int keep = (int)(bufferPtr - buffer); + if (keep > XML_CONTEXT_BYTES) + keep = XML_CONTEXT_BYTES; + memcpy(newBuf, &bufferPtr[-keep], bufferEnd - bufferPtr + keep); + FREE(buffer); + buffer = newBuf; + bufferEnd = buffer + (bufferEnd - bufferPtr) + keep; + bufferPtr = buffer + keep; + } + else { + bufferEnd = newBuf + (bufferEnd - bufferPtr); + bufferPtr = buffer = newBuf; + } +#else + if (bufferPtr) { + memcpy(newBuf, bufferPtr, bufferEnd - bufferPtr); + FREE(buffer); + } + bufferEnd = newBuf + (bufferEnd - bufferPtr); + bufferPtr = buffer = newBuf; +#endif /* not defined XML_CONTEXT_BYTES */ + } + eventPtr = eventEndPtr = NULL; + positionPtr = NULL; + } + return bufferEnd; +} + +enum XML_Status XMLCALL +XML_StopParser(XML_Parser parser, XML_Bool resumable) +{ + switch (ps_parsing) { + case XML_SUSPENDED: + if (resumable) { + errorCode = XML_ERROR_SUSPENDED; + return XML_STATUS_ERROR; + } + ps_parsing = XML_FINISHED; + break; + case XML_FINISHED: + errorCode = XML_ERROR_FINISHED; + return XML_STATUS_ERROR; + default: + if (resumable) { +#ifdef XML_DTD + if (isParamEntity) { + errorCode = XML_ERROR_SUSPEND_PE; + return XML_STATUS_ERROR; + } +#endif + ps_parsing = XML_SUSPENDED; + } + else + ps_parsing = XML_FINISHED; + } + return XML_STATUS_OK; +} + +enum XML_Status XMLCALL +XML_ResumeParser(XML_Parser parser) +{ + enum XML_Status result = XML_STATUS_OK; + + if (ps_parsing != XML_SUSPENDED) { + errorCode = XML_ERROR_NOT_SUSPENDED; + return XML_STATUS_ERROR; + } + ps_parsing = XML_PARSING; + + errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr); + + if (errorCode != XML_ERROR_NONE) { + eventEndPtr = eventPtr; + processor = errorProcessor; + return XML_STATUS_ERROR; + } + else { + switch (ps_parsing) { + case XML_SUSPENDED: + result = XML_STATUS_SUSPENDED; + break; + case XML_INITIALIZED: + case XML_PARSING: + if (ps_finalBuffer) { + ps_parsing = XML_FINISHED; + return result; + } + default: ; + } + } + + XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position); + positionPtr = bufferPtr; + return result; +} + +void XMLCALL +XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status) +{ + assert(status != NULL); + *status = parser->m_parsingStatus; +} + +enum XML_Error XMLCALL +XML_GetErrorCode(XML_Parser parser) +{ + return errorCode; +} + +XML_Index XMLCALL +XML_GetCurrentByteIndex(XML_Parser parser) +{ + if (eventPtr) + return parseEndByteIndex - (parseEndPtr - eventPtr); + return -1; +} + +int XMLCALL +XML_GetCurrentByteCount(XML_Parser parser) +{ + if (eventEndPtr && eventPtr) + return (int)(eventEndPtr - eventPtr); + return 0; +} + +const char * XMLCALL +XML_GetInputContext(XML_Parser parser, int *offset, int *size) +{ +#ifdef XML_CONTEXT_BYTES + if (eventPtr && buffer) { + *offset = (int)(eventPtr - buffer); + *size = (int)(bufferEnd - buffer); + return buffer; + } +#endif /* defined XML_CONTEXT_BYTES */ + return (char *) 0; +} + +XML_Size XMLCALL +XML_GetCurrentLineNumber(XML_Parser parser) +{ + if (eventPtr && eventPtr >= positionPtr) { + XmlUpdatePosition(encoding, positionPtr, eventPtr, &position); + positionPtr = eventPtr; + } + return position.lineNumber + 1; +} + +XML_Size XMLCALL +XML_GetCurrentColumnNumber(XML_Parser parser) +{ + if (eventPtr && eventPtr >= positionPtr) { + XmlUpdatePosition(encoding, positionPtr, eventPtr, &position); + positionPtr = eventPtr; + } + return position.columnNumber; +} + +void XMLCALL +XML_FreeContentModel(XML_Parser parser, XML_Content *model) +{ + FREE(model); +} + +void * XMLCALL +XML_MemMalloc(XML_Parser parser, size_t size) +{ + return MALLOC(size); +} + +void * XMLCALL +XML_MemRealloc(XML_Parser parser, void *ptr, size_t size) +{ + return REALLOC(ptr, size); +} + +void XMLCALL +XML_MemFree(XML_Parser parser, void *ptr) +{ + FREE(ptr); +} + +void XMLCALL +XML_DefaultCurrent(XML_Parser parser) +{ + if (defaultHandler) { + if (openInternalEntities) + reportDefault(parser, + internalEncoding, + openInternalEntities->internalEventPtr, + openInternalEntities->internalEventEndPtr); + else + reportDefault(parser, encoding, eventPtr, eventEndPtr); + } +} + +const XML_LChar * XMLCALL +XML_ErrorString(enum XML_Error code) +{ + static const XML_LChar* const message[] = { + 0, + XML_L("out of memory"), + XML_L("syntax error"), + XML_L("no element found"), + XML_L("not well-formed (invalid token)"), + XML_L("unclosed token"), + XML_L("partial character"), + XML_L("mismatched tag"), + XML_L("duplicate attribute"), + XML_L("junk after document element"), + XML_L("illegal parameter entity reference"), + XML_L("undefined entity"), + XML_L("recursive entity reference"), + XML_L("asynchronous entity"), + XML_L("reference to invalid character number"), + XML_L("reference to binary entity"), + XML_L("reference to external entity in attribute"), + XML_L("XML or text declaration not at start of entity"), + XML_L("unknown encoding"), + XML_L("encoding specified in XML declaration is incorrect"), + XML_L("unclosed CDATA section"), + XML_L("error in processing external entity reference"), + XML_L("document is not standalone"), + XML_L("unexpected parser state - please send a bug report"), + XML_L("entity declared in parameter entity"), + XML_L("requested feature requires XML_DTD support in Expat"), + XML_L("cannot change setting once parsing has begun"), + XML_L("unbound prefix"), + XML_L("must not undeclare prefix"), + XML_L("incomplete markup in parameter entity"), + XML_L("XML declaration not well-formed"), + XML_L("text declaration not well-formed"), + XML_L("illegal character(s) in public id"), + XML_L("parser suspended"), + XML_L("parser not suspended"), + XML_L("parsing aborted"), + XML_L("parsing finished"), + XML_L("cannot suspend in external parameter entity"), + XML_L("reserved prefix (xml) must not be undeclared or bound to another namespace name"), + XML_L("reserved prefix (xmlns) must not be declared or undeclared"), + XML_L("prefix must not be bound to one of the reserved namespace names") + }; + if (code > 0 && code < sizeof(message)/sizeof(message[0])) + return message[code]; + return NULL; +} + +const XML_LChar * XMLCALL +XML_ExpatVersion(void) { + + /* V1 is used to string-ize the version number. However, it would + string-ize the actual version macro *names* unless we get them + substituted before being passed to V1. CPP is defined to expand + a macro, then rescan for more expansions. Thus, we use V2 to expand + the version macros, then CPP will expand the resulting V1() macro + with the correct numerals. */ + /* ### I'm assuming cpp is portable in this respect... */ + +#define V1(a,b,c) XML_L(#a)XML_L(".")XML_L(#b)XML_L(".")XML_L(#c) +#define V2(a,b,c) XML_L("expat_")V1(a,b,c) + + return V2(XML_MAJOR_VERSION, XML_MINOR_VERSION, XML_MICRO_VERSION); + +#undef V1 +#undef V2 +} + +XML_Expat_Version XMLCALL +XML_ExpatVersionInfo(void) +{ + XML_Expat_Version version; + + version.major = XML_MAJOR_VERSION; + version.minor = XML_MINOR_VERSION; + version.micro = XML_MICRO_VERSION; + + return version; +} + +const XML_Feature * XMLCALL +XML_GetFeatureList(void) +{ + static const XML_Feature features[] = { + {XML_FEATURE_SIZEOF_XML_CHAR, XML_L("sizeof(XML_Char)"), + sizeof(XML_Char)}, + {XML_FEATURE_SIZEOF_XML_LCHAR, XML_L("sizeof(XML_LChar)"), + sizeof(XML_LChar)}, +#ifdef XML_UNICODE + {XML_FEATURE_UNICODE, XML_L("XML_UNICODE"), 0}, +#endif +#ifdef XML_UNICODE_WCHAR_T + {XML_FEATURE_UNICODE_WCHAR_T, XML_L("XML_UNICODE_WCHAR_T"), 0}, +#endif +#ifdef XML_DTD + {XML_FEATURE_DTD, XML_L("XML_DTD"), 0}, +#endif +#ifdef XML_CONTEXT_BYTES + {XML_FEATURE_CONTEXT_BYTES, XML_L("XML_CONTEXT_BYTES"), + XML_CONTEXT_BYTES}, +#endif +#ifdef XML_MIN_SIZE + {XML_FEATURE_MIN_SIZE, XML_L("XML_MIN_SIZE"), 0}, +#endif +#ifdef XML_NS + {XML_FEATURE_NS, XML_L("XML_NS"), 0}, +#endif +#ifdef XML_LARGE_SIZE + {XML_FEATURE_LARGE_SIZE, XML_L("XML_LARGE_SIZE"), 0}, +#endif +#ifdef XML_ATTR_INFO + {XML_FEATURE_ATTR_INFO, XML_L("XML_ATTR_INFO"), 0}, +#endif + {XML_FEATURE_END, NULL, 0} + }; + + return features; +} + +/* Initially tag->rawName always points into the parse buffer; + for those TAG instances opened while the current parse buffer was + processed, and not yet closed, we need to store tag->rawName in a more + permanent location, since the parse buffer is about to be discarded. +*/ +static XML_Bool +storeRawNames(XML_Parser parser) +{ + TAG *tag = tagStack; + while (tag) { + int bufSize; + int nameLen = sizeof(XML_Char) * (tag->name.strLen + 1); + char *rawNameBuf = tag->buf + nameLen; + /* Stop if already stored. Since tagStack is a stack, we can stop + at the first entry that has already been copied; everything + below it in the stack is already been accounted for in a + previous call to this function. + */ + if (tag->rawName == rawNameBuf) + break; + /* For re-use purposes we need to ensure that the + size of tag->buf is a multiple of sizeof(XML_Char). + */ + bufSize = nameLen + ROUND_UP(tag->rawNameLength, sizeof(XML_Char)); + if (bufSize > tag->bufEnd - tag->buf) { + char *temp = (char *)REALLOC(tag->buf, bufSize); + if (temp == NULL) + return XML_FALSE; + /* if tag->name.str points to tag->buf (only when namespace + processing is off) then we have to update it + */ + if (tag->name.str == (XML_Char *)tag->buf) + tag->name.str = (XML_Char *)temp; + /* if tag->name.localPart is set (when namespace processing is on) + then update it as well, since it will always point into tag->buf + */ + if (tag->name.localPart) + tag->name.localPart = (XML_Char *)temp + (tag->name.localPart - + (XML_Char *)tag->buf); + tag->buf = temp; + tag->bufEnd = temp + bufSize; + rawNameBuf = temp + nameLen; + } + memcpy(rawNameBuf, tag->rawName, tag->rawNameLength); + tag->rawName = rawNameBuf; + tag = tag->parent; + } + return XML_TRUE; +} + +static enum XML_Error PTRCALL +contentProcessor(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr) +{ + enum XML_Error result = doContent(parser, 0, encoding, start, end, + endPtr, (XML_Bool)!ps_finalBuffer); + if (result == XML_ERROR_NONE) { + if (!storeRawNames(parser)) + return XML_ERROR_NO_MEMORY; + } + return result; +} + +static enum XML_Error PTRCALL +externalEntityInitProcessor(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr) +{ + enum XML_Error result = initializeEncoding(parser); + if (result != XML_ERROR_NONE) + return result; + processor = externalEntityInitProcessor2; + return externalEntityInitProcessor2(parser, start, end, endPtr); +} + +static enum XML_Error PTRCALL +externalEntityInitProcessor2(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr) +{ + const char *next = start; /* XmlContentTok doesn't always set the last arg */ + int tok = XmlContentTok(encoding, start, end, &next); + switch (tok) { + case XML_TOK_BOM: + /* If we are at the end of the buffer, this would cause the next stage, + i.e. externalEntityInitProcessor3, to pass control directly to + doContent (by detecting XML_TOK_NONE) without processing any xml text + declaration - causing the error XML_ERROR_MISPLACED_XML_PI in doContent. + */ + if (next == end && !ps_finalBuffer) { + *endPtr = next; + return XML_ERROR_NONE; + } + start = next; + break; + case XML_TOK_PARTIAL: + if (!ps_finalBuffer) { + *endPtr = start; + return XML_ERROR_NONE; + } + eventPtr = start; + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + if (!ps_finalBuffer) { + *endPtr = start; + return XML_ERROR_NONE; + } + eventPtr = start; + return XML_ERROR_PARTIAL_CHAR; + } + processor = externalEntityInitProcessor3; + return externalEntityInitProcessor3(parser, start, end, endPtr); +} + +static enum XML_Error PTRCALL +externalEntityInitProcessor3(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr) +{ + int tok; + const char *next = start; /* XmlContentTok doesn't always set the last arg */ + eventPtr = start; + tok = XmlContentTok(encoding, start, end, &next); + eventEndPtr = next; + + switch (tok) { + case XML_TOK_XML_DECL: + { + enum XML_Error result; + result = processXmlDecl(parser, 1, start, next); + if (result != XML_ERROR_NONE) + return result; + switch (ps_parsing) { + case XML_SUSPENDED: + *endPtr = next; + return XML_ERROR_NONE; + case XML_FINISHED: + return XML_ERROR_ABORTED; + default: + start = next; + } + } + break; + case XML_TOK_PARTIAL: + if (!ps_finalBuffer) { + *endPtr = start; + return XML_ERROR_NONE; + } + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + if (!ps_finalBuffer) { + *endPtr = start; + return XML_ERROR_NONE; + } + return XML_ERROR_PARTIAL_CHAR; + } + processor = externalEntityContentProcessor; + tagLevel = 1; + return externalEntityContentProcessor(parser, start, end, endPtr); +} + +static enum XML_Error PTRCALL +externalEntityContentProcessor(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr) +{ + enum XML_Error result = doContent(parser, 1, encoding, start, end, + endPtr, (XML_Bool)!ps_finalBuffer); + if (result == XML_ERROR_NONE) { + if (!storeRawNames(parser)) + return XML_ERROR_NO_MEMORY; + } + return result; +} + +static enum XML_Error +doContent(XML_Parser parser, + int startTagLevel, + const ENCODING *enc, + const char *s, + const char *end, + const char **nextPtr, + XML_Bool haveMore) +{ + /* save one level of indirection */ + DTD * const dtd = _dtd; + + const char **eventPP; + const char **eventEndPP; + if (enc == encoding) { + eventPP = &eventPtr; + eventEndPP = &eventEndPtr; + } + else { + eventPP = &(openInternalEntities->internalEventPtr); + eventEndPP = &(openInternalEntities->internalEventEndPtr); + } + *eventPP = s; + + for (;;) { + const char *next = s; /* XmlContentTok doesn't always set the last arg */ + int tok = XmlContentTok(enc, s, end, &next); + *eventEndPP = next; + switch (tok) { + case XML_TOK_TRAILING_CR: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + *eventEndPP = end; + if (characterDataHandler) { + XML_Char c = 0xA; + characterDataHandler(handlerArg, &c, 1); + } + else if (defaultHandler) + reportDefault(parser, enc, s, end); + /* We are at the end of the final buffer, should we check for + XML_SUSPENDED, XML_FINISHED? + */ + if (startTagLevel == 0) + return XML_ERROR_NO_ELEMENTS; + if (tagLevel != startTagLevel) + return XML_ERROR_ASYNC_ENTITY; + *nextPtr = end; + return XML_ERROR_NONE; + case XML_TOK_NONE: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + if (startTagLevel > 0) { + if (tagLevel != startTagLevel) + return XML_ERROR_ASYNC_ENTITY; + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_NO_ELEMENTS; + case XML_TOK_INVALID: + *eventPP = next; + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_PARTIAL_CHAR; + case XML_TOK_ENTITY_REF: + { + const XML_Char *name; + ENTITY *entity; + XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (ch) { + if (characterDataHandler) + characterDataHandler(handlerArg, &ch, 1); + else if (defaultHandler) + reportDefault(parser, enc, s, next); + break; + } + name = poolStoreString(&dtd->pool, enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!name) + return XML_ERROR_NO_MEMORY; + entity = (ENTITY *)lookup(parser, &dtd->generalEntities, name, 0); + poolDiscard(&dtd->pool); + /* First, determine if a check for an existing declaration is needed; + if yes, check that the entity exists, and that it is internal, + otherwise call the skipped entity or default handler. + */ + if (!dtd->hasParamEntityRefs || dtd->standalone) { + if (!entity) + return XML_ERROR_UNDEFINED_ENTITY; + else if (!entity->is_internal) + return XML_ERROR_ENTITY_DECLARED_IN_PE; + } + else if (!entity) { + if (skippedEntityHandler) + skippedEntityHandler(handlerArg, name, 0); + else if (defaultHandler) + reportDefault(parser, enc, s, next); + break; + } + if (entity->open) + return XML_ERROR_RECURSIVE_ENTITY_REF; + if (entity->notation) + return XML_ERROR_BINARY_ENTITY_REF; + if (entity->textPtr) { + enum XML_Error result; + if (!defaultExpandInternalEntities) { + if (skippedEntityHandler) + skippedEntityHandler(handlerArg, entity->name, 0); + else if (defaultHandler) + reportDefault(parser, enc, s, next); + break; + } + result = processInternalEntity(parser, entity, XML_FALSE); + if (result != XML_ERROR_NONE) + return result; + } + else if (externalEntityRefHandler) { + const XML_Char *context; + entity->open = XML_TRUE; + context = getContext(parser); + entity->open = XML_FALSE; + if (!context) + return XML_ERROR_NO_MEMORY; + if (!externalEntityRefHandler(externalEntityRefHandlerArg, + context, + entity->base, + entity->systemId, + entity->publicId)) + return XML_ERROR_EXTERNAL_ENTITY_HANDLING; + poolDiscard(&tempPool); + } + else if (defaultHandler) + reportDefault(parser, enc, s, next); + break; + } + case XML_TOK_START_TAG_NO_ATTS: + /* fall through */ + case XML_TOK_START_TAG_WITH_ATTS: + { + TAG *tag; + enum XML_Error result; + XML_Char *toPtr; + if (freeTagList) { + tag = freeTagList; + freeTagList = freeTagList->parent; + } + else { + tag = (TAG *)MALLOC(sizeof(TAG)); + if (!tag) + return XML_ERROR_NO_MEMORY; + tag->buf = (char *)MALLOC(INIT_TAG_BUF_SIZE); + if (!tag->buf) { + FREE(tag); + return XML_ERROR_NO_MEMORY; + } + tag->bufEnd = tag->buf + INIT_TAG_BUF_SIZE; + } + tag->bindings = NULL; + tag->parent = tagStack; + tagStack = tag; + tag->name.localPart = NULL; + tag->name.prefix = NULL; + tag->rawName = s + enc->minBytesPerChar; + tag->rawNameLength = XmlNameLength(enc, tag->rawName); + ++tagLevel; + { + const char *rawNameEnd = tag->rawName + tag->rawNameLength; + const char *fromPtr = tag->rawName; + toPtr = (XML_Char *)tag->buf; + for (;;) { + int bufSize; + int convLen; + XmlConvert(enc, + &fromPtr, rawNameEnd, + (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1); + convLen = (int)(toPtr - (XML_Char *)tag->buf); + if (fromPtr == rawNameEnd) { + tag->name.strLen = convLen; + break; + } + bufSize = (int)(tag->bufEnd - tag->buf) << 1; + { + char *temp = (char *)REALLOC(tag->buf, bufSize); + if (temp == NULL) + return XML_ERROR_NO_MEMORY; + tag->buf = temp; + tag->bufEnd = temp + bufSize; + toPtr = (XML_Char *)temp + convLen; + } + } + } + tag->name.str = (XML_Char *)tag->buf; + *toPtr = XML_T('\0'); + result = storeAtts(parser, enc, s, &(tag->name), &(tag->bindings)); + if (result) + return result; + if (startElementHandler) + startElementHandler(handlerArg, tag->name.str, + (const XML_Char **)atts); + else if (defaultHandler) + reportDefault(parser, enc, s, next); + poolClear(&tempPool); + break; + } + case XML_TOK_EMPTY_ELEMENT_NO_ATTS: + /* fall through */ + case XML_TOK_EMPTY_ELEMENT_WITH_ATTS: + { + const char *rawName = s + enc->minBytesPerChar; + enum XML_Error result; + BINDING *bindings = NULL; + XML_Bool noElmHandlers = XML_TRUE; + TAG_NAME name; + name.str = poolStoreString(&tempPool, enc, rawName, + rawName + XmlNameLength(enc, rawName)); + if (!name.str) + return XML_ERROR_NO_MEMORY; + poolFinish(&tempPool); + result = storeAtts(parser, enc, s, &name, &bindings); + if (result) + return result; + poolFinish(&tempPool); + if (startElementHandler) { + startElementHandler(handlerArg, name.str, (const XML_Char **)atts); + noElmHandlers = XML_FALSE; + } + if (endElementHandler) { + if (startElementHandler) + *eventPP = *eventEndPP; + endElementHandler(handlerArg, name.str); + noElmHandlers = XML_FALSE; + } + if (noElmHandlers && defaultHandler) + reportDefault(parser, enc, s, next); + poolClear(&tempPool); + while (bindings) { + BINDING *b = bindings; + if (endNamespaceDeclHandler) + endNamespaceDeclHandler(handlerArg, b->prefix->name); + bindings = bindings->nextTagBinding; + b->nextTagBinding = freeBindingList; + freeBindingList = b; + b->prefix->binding = b->prevPrefixBinding; + } + } + if (tagLevel == 0) + return epilogProcessor(parser, next, end, nextPtr); + break; + case XML_TOK_END_TAG: + if (tagLevel == startTagLevel) + return XML_ERROR_ASYNC_ENTITY; + else { + int len; + const char *rawName; + TAG *tag = tagStack; + tagStack = tag->parent; + tag->parent = freeTagList; + freeTagList = tag; + rawName = s + enc->minBytesPerChar*2; + len = XmlNameLength(enc, rawName); + if (len != tag->rawNameLength + || memcmp(tag->rawName, rawName, len) != 0) { + *eventPP = rawName; + return XML_ERROR_TAG_MISMATCH; + } + --tagLevel; + if (endElementHandler) { + const XML_Char *localPart; + const XML_Char *prefix; + XML_Char *uri; + localPart = tag->name.localPart; + if (ns && localPart) { + /* localPart and prefix may have been overwritten in + tag->name.str, since this points to the binding->uri + buffer which gets re-used; so we have to add them again + */ + uri = (XML_Char *)tag->name.str + tag->name.uriLen; + /* don't need to check for space - already done in storeAtts() */ + while (*localPart) *uri++ = *localPart++; + prefix = (XML_Char *)tag->name.prefix; + if (ns_triplets && prefix) { + *uri++ = namespaceSeparator; + while (*prefix) *uri++ = *prefix++; + } + *uri = XML_T('\0'); + } + endElementHandler(handlerArg, tag->name.str); + } + else if (defaultHandler) + reportDefault(parser, enc, s, next); + while (tag->bindings) { + BINDING *b = tag->bindings; + if (endNamespaceDeclHandler) + endNamespaceDeclHandler(handlerArg, b->prefix->name); + tag->bindings = tag->bindings->nextTagBinding; + b->nextTagBinding = freeBindingList; + freeBindingList = b; + b->prefix->binding = b->prevPrefixBinding; + } + if (tagLevel == 0) + return epilogProcessor(parser, next, end, nextPtr); + } + break; + case XML_TOK_CHAR_REF: + { + int n = XmlCharRefNumber(enc, s); + if (n < 0) + return XML_ERROR_BAD_CHAR_REF; + if (characterDataHandler) { + XML_Char buf[XML_ENCODE_MAX]; + characterDataHandler(handlerArg, buf, XmlEncode(n, (ICHAR *)buf)); + } + else if (defaultHandler) + reportDefault(parser, enc, s, next); + } + break; + case XML_TOK_XML_DECL: + return XML_ERROR_MISPLACED_XML_PI; + case XML_TOK_DATA_NEWLINE: + if (characterDataHandler) { + XML_Char c = 0xA; + characterDataHandler(handlerArg, &c, 1); + } + else if (defaultHandler) + reportDefault(parser, enc, s, next); + break; + case XML_TOK_CDATA_SECT_OPEN: + { + enum XML_Error result; + if (startCdataSectionHandler) + startCdataSectionHandler(handlerArg); +#if 0 + /* Suppose you doing a transformation on a document that involves + changing only the character data. You set up a defaultHandler + and a characterDataHandler. The defaultHandler simply copies + characters through. The characterDataHandler does the + transformation and writes the characters out escaping them as + necessary. This case will fail to work if we leave out the + following two lines (because & and < inside CDATA sections will + be incorrectly escaped). + + However, now we have a start/endCdataSectionHandler, so it seems + easier to let the user deal with this. + */ + else if (characterDataHandler) + characterDataHandler(handlerArg, dataBuf, 0); +#endif + else if (defaultHandler) + reportDefault(parser, enc, s, next); + result = doCdataSection(parser, enc, &next, end, nextPtr, haveMore); + if (result != XML_ERROR_NONE) + return result; + else if (!next) { + processor = cdataSectionProcessor; + return result; + } + } + break; + case XML_TOK_TRAILING_RSQB: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + if (characterDataHandler) { + if (MUST_CONVERT(enc, s)) { + ICHAR *dataPtr = (ICHAR *)dataBuf; + XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd); + characterDataHandler(handlerArg, dataBuf, + (int)(dataPtr - (ICHAR *)dataBuf)); + } + else + characterDataHandler(handlerArg, + (XML_Char *)s, + (int)((XML_Char *)end - (XML_Char *)s)); + } + else if (defaultHandler) + reportDefault(parser, enc, s, end); + /* We are at the end of the final buffer, should we check for + XML_SUSPENDED, XML_FINISHED? + */ + if (startTagLevel == 0) { + *eventPP = end; + return XML_ERROR_NO_ELEMENTS; + } + if (tagLevel != startTagLevel) { + *eventPP = end; + return XML_ERROR_ASYNC_ENTITY; + } + *nextPtr = end; + return XML_ERROR_NONE; + case XML_TOK_DATA_CHARS: + { + XML_CharacterDataHandler charDataHandler = characterDataHandler; + if (charDataHandler) { + if (MUST_CONVERT(enc, s)) { + for (;;) { + ICHAR *dataPtr = (ICHAR *)dataBuf; + XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); + *eventEndPP = s; + charDataHandler(handlerArg, dataBuf, + (int)(dataPtr - (ICHAR *)dataBuf)); + if (s == next) + break; + *eventPP = s; + } + } + else + charDataHandler(handlerArg, + (XML_Char *)s, + (int)((XML_Char *)next - (XML_Char *)s)); + } + else if (defaultHandler) + reportDefault(parser, enc, s, next); + } + break; + case XML_TOK_PI: + if (!reportProcessingInstruction(parser, enc, s, next)) + return XML_ERROR_NO_MEMORY; + break; + case XML_TOK_COMMENT: + if (!reportComment(parser, enc, s, next)) + return XML_ERROR_NO_MEMORY; + break; + default: + if (defaultHandler) + reportDefault(parser, enc, s, next); + break; + } + *eventPP = s = next; + switch (ps_parsing) { + case XML_SUSPENDED: + *nextPtr = next; + return XML_ERROR_NONE; + case XML_FINISHED: + return XML_ERROR_ABORTED; + default: ; + } + } + /* not reached */ +} + +/* Precondition: all arguments must be non-NULL; + Purpose: + - normalize attributes + - check attributes for well-formedness + - generate namespace aware attribute names (URI, prefix) + - build list of attributes for startElementHandler + - default attributes + - process namespace declarations (check and report them) + - generate namespace aware element name (URI, prefix) +*/ +static enum XML_Error +storeAtts(XML_Parser parser, const ENCODING *enc, + const char *attStr, TAG_NAME *tagNamePtr, + BINDING **bindingsPtr) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + ELEMENT_TYPE *elementType; + int nDefaultAtts; + const XML_Char **appAtts; /* the attribute list for the application */ + int attIndex = 0; + int prefixLen; + int i; + int n; + XML_Char *uri; + int nPrefixes = 0; + BINDING *binding; + const XML_Char *localPart; + + /* lookup the element type name */ + elementType = (ELEMENT_TYPE *)lookup(parser, &dtd->elementTypes, tagNamePtr->str,0); + if (!elementType) { + const XML_Char *name = poolCopyString(&dtd->pool, tagNamePtr->str); + if (!name) + return XML_ERROR_NO_MEMORY; + elementType = (ELEMENT_TYPE *)lookup(parser, &dtd->elementTypes, name, + sizeof(ELEMENT_TYPE)); + if (!elementType) + return XML_ERROR_NO_MEMORY; + if (ns && !setElementTypePrefix(parser, elementType)) + return XML_ERROR_NO_MEMORY; + } + nDefaultAtts = elementType->nDefaultAtts; + + /* get the attributes from the tokenizer */ + n = XmlGetAttributes(enc, attStr, attsSize, atts); + if (n + nDefaultAtts > attsSize) { + int oldAttsSize = attsSize; + ATTRIBUTE *temp; +#ifdef XML_ATTR_INFO + XML_AttrInfo *temp2; +#endif + attsSize = n + nDefaultAtts + INIT_ATTS_SIZE; + temp = (ATTRIBUTE *)REALLOC((void *)atts, attsSize * sizeof(ATTRIBUTE)); + if (temp == NULL) + return XML_ERROR_NO_MEMORY; + atts = temp; +#ifdef XML_ATTR_INFO + temp2 = (XML_AttrInfo *)REALLOC((void *)attInfo, attsSize * sizeof(XML_AttrInfo)); + if (temp2 == NULL) + return XML_ERROR_NO_MEMORY; + attInfo = temp2; +#endif + if (n > oldAttsSize) + XmlGetAttributes(enc, attStr, n, atts); + } + + appAtts = (const XML_Char **)atts; + for (i = 0; i < n; i++) { + ATTRIBUTE *currAtt = &atts[i]; +#ifdef XML_ATTR_INFO + XML_AttrInfo *currAttInfo = &attInfo[i]; +#endif + /* add the name and value to the attribute list */ + ATTRIBUTE_ID *attId = getAttributeId(parser, enc, currAtt->name, + currAtt->name + + XmlNameLength(enc, currAtt->name)); + if (!attId) + return XML_ERROR_NO_MEMORY; +#ifdef XML_ATTR_INFO + currAttInfo->nameStart = parseEndByteIndex - (parseEndPtr - currAtt->name); + currAttInfo->nameEnd = currAttInfo->nameStart + + XmlNameLength(enc, currAtt->name); + currAttInfo->valueStart = parseEndByteIndex - + (parseEndPtr - currAtt->valuePtr); + currAttInfo->valueEnd = parseEndByteIndex - (parseEndPtr - currAtt->valueEnd); +#endif + /* Detect duplicate attributes by their QNames. This does not work when + namespace processing is turned on and different prefixes for the same + namespace are used. For this case we have a check further down. + */ + if ((attId->name)[-1]) { + if (enc == encoding) + eventPtr = atts[i].name; + return XML_ERROR_DUPLICATE_ATTRIBUTE; + } + (attId->name)[-1] = 1; + appAtts[attIndex++] = attId->name; + if (!atts[i].normalized) { + enum XML_Error result; + XML_Bool isCdata = XML_TRUE; + + /* figure out whether declared as other than CDATA */ + if (attId->maybeTokenized) { + int j; + for (j = 0; j < nDefaultAtts; j++) { + if (attId == elementType->defaultAtts[j].id) { + isCdata = elementType->defaultAtts[j].isCdata; + break; + } + } + } + + /* normalize the attribute value */ + result = storeAttributeValue(parser, enc, isCdata, + atts[i].valuePtr, atts[i].valueEnd, + &tempPool); + if (result) + return result; + appAtts[attIndex] = poolStart(&tempPool); + poolFinish(&tempPool); + } + else { + /* the value did not need normalizing */ + appAtts[attIndex] = poolStoreString(&tempPool, enc, atts[i].valuePtr, + atts[i].valueEnd); + if (appAtts[attIndex] == 0) + return XML_ERROR_NO_MEMORY; + poolFinish(&tempPool); + } + /* handle prefixed attribute names */ + if (attId->prefix) { + if (attId->xmlns) { + /* deal with namespace declarations here */ + enum XML_Error result = addBinding(parser, attId->prefix, attId, + appAtts[attIndex], bindingsPtr); + if (result) + return result; + --attIndex; + } + else { + /* deal with other prefixed names later */ + attIndex++; + nPrefixes++; + (attId->name)[-1] = 2; + } + } + else + attIndex++; + } + + /* set-up for XML_GetSpecifiedAttributeCount and XML_GetIdAttributeIndex */ + nSpecifiedAtts = attIndex; + if (elementType->idAtt && (elementType->idAtt->name)[-1]) { + for (i = 0; i < attIndex; i += 2) + if (appAtts[i] == elementType->idAtt->name) { + idAttIndex = i; + break; + } + } + else + idAttIndex = -1; + + /* do attribute defaulting */ + for (i = 0; i < nDefaultAtts; i++) { + const DEFAULT_ATTRIBUTE *da = elementType->defaultAtts + i; + if (!(da->id->name)[-1] && da->value) { + if (da->id->prefix) { + if (da->id->xmlns) { + enum XML_Error result = addBinding(parser, da->id->prefix, da->id, + da->value, bindingsPtr); + if (result) + return result; + } + else { + (da->id->name)[-1] = 2; + nPrefixes++; + appAtts[attIndex++] = da->id->name; + appAtts[attIndex++] = da->value; + } + } + else { + (da->id->name)[-1] = 1; + appAtts[attIndex++] = da->id->name; + appAtts[attIndex++] = da->value; + } + } + } + appAtts[attIndex] = 0; + + /* expand prefixed attribute names, check for duplicates, + and clear flags that say whether attributes were specified */ + i = 0; + if (nPrefixes) { + int j; /* hash table index */ + unsigned long version = nsAttsVersion; + int nsAttsSize = (int)1 << nsAttsPower; + /* size of hash table must be at least 2 * (# of prefixed attributes) */ + if ((nPrefixes << 1) >> nsAttsPower) { /* true for nsAttsPower = 0 */ + NS_ATT *temp; + /* hash table size must also be a power of 2 and >= 8 */ + while (nPrefixes >> nsAttsPower++); + if (nsAttsPower < 3) + nsAttsPower = 3; + nsAttsSize = (int)1 << nsAttsPower; + temp = (NS_ATT *)REALLOC(nsAtts, nsAttsSize * sizeof(NS_ATT)); + if (!temp) + return XML_ERROR_NO_MEMORY; + nsAtts = temp; + version = 0; /* force re-initialization of nsAtts hash table */ + } + /* using a version flag saves us from initializing nsAtts every time */ + if (!version) { /* initialize version flags when version wraps around */ + version = INIT_ATTS_VERSION; + for (j = nsAttsSize; j != 0; ) + nsAtts[--j].version = version; + } + nsAttsVersion = --version; + + /* expand prefixed names and check for duplicates */ + for (; i < attIndex; i += 2) { + const XML_Char *s = appAtts[i]; + if (s[-1] == 2) { /* prefixed */ + ATTRIBUTE_ID *id; + const BINDING *b; + unsigned long uriHash = hash_secret_salt; + ((XML_Char *)s)[-1] = 0; /* clear flag */ + id = (ATTRIBUTE_ID *)lookup(parser, &dtd->attributeIds, s, 0); + b = id->prefix->binding; + if (!b) + return XML_ERROR_UNBOUND_PREFIX; + + /* as we expand the name we also calculate its hash value */ + for (j = 0; j < b->uriLen; j++) { + const XML_Char c = b->uri[j]; + if (!poolAppendChar(&tempPool, c)) + return XML_ERROR_NO_MEMORY; + uriHash = CHAR_HASH(uriHash, c); + } + while (*s++ != XML_T(ASCII_COLON)) + ; + do { /* copies null terminator */ + const XML_Char c = *s; + if (!poolAppendChar(&tempPool, *s)) + return XML_ERROR_NO_MEMORY; + uriHash = CHAR_HASH(uriHash, c); + } while (*s++); + + { /* Check hash table for duplicate of expanded name (uriName). + Derived from code in lookup(parser, HASH_TABLE *table, ...). + */ + unsigned char step = 0; + unsigned long mask = nsAttsSize - 1; + j = uriHash & mask; /* index into hash table */ + while (nsAtts[j].version == version) { + /* for speed we compare stored hash values first */ + if (uriHash == nsAtts[j].hash) { + const XML_Char *s1 = poolStart(&tempPool); + const XML_Char *s2 = nsAtts[j].uriName; + /* s1 is null terminated, but not s2 */ + for (; *s1 == *s2 && *s1 != 0; s1++, s2++); + if (*s1 == 0) + return XML_ERROR_DUPLICATE_ATTRIBUTE; + } + if (!step) + step = PROBE_STEP(uriHash, mask, nsAttsPower); + j < step ? (j += nsAttsSize - step) : (j -= step); + } + } + + if (ns_triplets) { /* append namespace separator and prefix */ + tempPool.ptr[-1] = namespaceSeparator; + s = b->prefix->name; + do { + if (!poolAppendChar(&tempPool, *s)) + return XML_ERROR_NO_MEMORY; + } while (*s++); + } + + /* store expanded name in attribute list */ + s = poolStart(&tempPool); + poolFinish(&tempPool); + appAtts[i] = s; + + /* fill empty slot with new version, uriName and hash value */ + nsAtts[j].version = version; + nsAtts[j].hash = uriHash; + nsAtts[j].uriName = s; + + if (!--nPrefixes) { + i += 2; + break; + } + } + else /* not prefixed */ + ((XML_Char *)s)[-1] = 0; /* clear flag */ + } + } + /* clear flags for the remaining attributes */ + for (; i < attIndex; i += 2) + ((XML_Char *)(appAtts[i]))[-1] = 0; + for (binding = *bindingsPtr; binding; binding = binding->nextTagBinding) + binding->attId->name[-1] = 0; + + if (!ns) + return XML_ERROR_NONE; + + /* expand the element type name */ + if (elementType->prefix) { + binding = elementType->prefix->binding; + if (!binding) + return XML_ERROR_UNBOUND_PREFIX; + localPart = tagNamePtr->str; + while (*localPart++ != XML_T(ASCII_COLON)) + ; + } + else if (dtd->defaultPrefix.binding) { + binding = dtd->defaultPrefix.binding; + localPart = tagNamePtr->str; + } + else + return XML_ERROR_NONE; + prefixLen = 0; + if (ns_triplets && binding->prefix->name) { + for (; binding->prefix->name[prefixLen++];) + ; /* prefixLen includes null terminator */ + } + tagNamePtr->localPart = localPart; + tagNamePtr->uriLen = binding->uriLen; + tagNamePtr->prefix = binding->prefix->name; + tagNamePtr->prefixLen = prefixLen; + for (i = 0; localPart[i++];) + ; /* i includes null terminator */ + n = i + binding->uriLen + prefixLen; + if (n > binding->uriAlloc) { + TAG *p; + uri = (XML_Char *)MALLOC((n + EXPAND_SPARE) * sizeof(XML_Char)); + if (!uri) + return XML_ERROR_NO_MEMORY; + binding->uriAlloc = n + EXPAND_SPARE; + memcpy(uri, binding->uri, binding->uriLen * sizeof(XML_Char)); + for (p = tagStack; p; p = p->parent) + if (p->name.str == binding->uri) + p->name.str = uri; + FREE(binding->uri); + binding->uri = uri; + } + /* if namespaceSeparator != '\0' then uri includes it already */ + uri = binding->uri + binding->uriLen; + memcpy(uri, localPart, i * sizeof(XML_Char)); + /* we always have a namespace separator between localPart and prefix */ + if (prefixLen) { + uri += i - 1; + *uri = namespaceSeparator; /* replace null terminator */ + memcpy(uri + 1, binding->prefix->name, prefixLen * sizeof(XML_Char)); + } + tagNamePtr->str = binding->uri; + return XML_ERROR_NONE; +} + +/* addBinding() overwrites the value of prefix->binding without checking. + Therefore one must keep track of the old value outside of addBinding(). +*/ +static enum XML_Error +addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId, + const XML_Char *uri, BINDING **bindingsPtr) +{ + static const XML_Char xmlNamespace[] = { + ASCII_h, ASCII_t, ASCII_t, ASCII_p, ASCII_COLON, ASCII_SLASH, ASCII_SLASH, + ASCII_w, ASCII_w, ASCII_w, ASCII_PERIOD, ASCII_w, ASCII_3, ASCII_PERIOD, + ASCII_o, ASCII_r, ASCII_g, ASCII_SLASH, ASCII_X, ASCII_M, ASCII_L, + ASCII_SLASH, ASCII_1, ASCII_9, ASCII_9, ASCII_8, ASCII_SLASH, + ASCII_n, ASCII_a, ASCII_m, ASCII_e, ASCII_s, ASCII_p, ASCII_a, ASCII_c, + ASCII_e, '\0' + }; + static const int xmlLen = + (int)sizeof(xmlNamespace)/sizeof(XML_Char) - 1; + static const XML_Char xmlnsNamespace[] = { + ASCII_h, ASCII_t, ASCII_t, ASCII_p, ASCII_COLON, ASCII_SLASH, ASCII_SLASH, + ASCII_w, ASCII_w, ASCII_w, ASCII_PERIOD, ASCII_w, ASCII_3, ASCII_PERIOD, + ASCII_o, ASCII_r, ASCII_g, ASCII_SLASH, ASCII_2, ASCII_0, ASCII_0, + ASCII_0, ASCII_SLASH, ASCII_x, ASCII_m, ASCII_l, ASCII_n, ASCII_s, + ASCII_SLASH, '\0' + }; + static const int xmlnsLen = + (int)sizeof(xmlnsNamespace)/sizeof(XML_Char) - 1; + + XML_Bool mustBeXML = XML_FALSE; + XML_Bool isXML = XML_TRUE; + XML_Bool isXMLNS = XML_TRUE; + + BINDING *b; + int len; + + /* empty URI is only valid for default namespace per XML NS 1.0 (not 1.1) */ + if (*uri == XML_T('\0') && prefix->name) + return XML_ERROR_UNDECLARING_PREFIX; + + if (prefix->name + && prefix->name[0] == XML_T(ASCII_x) + && prefix->name[1] == XML_T(ASCII_m) + && prefix->name[2] == XML_T(ASCII_l)) { + + /* Not allowed to bind xmlns */ + if (prefix->name[3] == XML_T(ASCII_n) + && prefix->name[4] == XML_T(ASCII_s) + && prefix->name[5] == XML_T('\0')) + return XML_ERROR_RESERVED_PREFIX_XMLNS; + + if (prefix->name[3] == XML_T('\0')) + mustBeXML = XML_TRUE; + } + + for (len = 0; uri[len]; len++) { + if (isXML && (len > xmlLen || uri[len] != xmlNamespace[len])) + isXML = XML_FALSE; + + if (!mustBeXML && isXMLNS + && (len > xmlnsLen || uri[len] != xmlnsNamespace[len])) + isXMLNS = XML_FALSE; + } + isXML = isXML && len == xmlLen; + isXMLNS = isXMLNS && len == xmlnsLen; + + if (mustBeXML != isXML) + return mustBeXML ? XML_ERROR_RESERVED_PREFIX_XML + : XML_ERROR_RESERVED_NAMESPACE_URI; + + if (isXMLNS) + return XML_ERROR_RESERVED_NAMESPACE_URI; + + if (namespaceSeparator) + len++; + if (freeBindingList) { + b = freeBindingList; + if (len > b->uriAlloc) { + XML_Char *temp = (XML_Char *)REALLOC(b->uri, + sizeof(XML_Char) * (len + EXPAND_SPARE)); + if (temp == NULL) + return XML_ERROR_NO_MEMORY; + b->uri = temp; + b->uriAlloc = len + EXPAND_SPARE; + } + freeBindingList = b->nextTagBinding; + } + else { + b = (BINDING *)MALLOC(sizeof(BINDING)); + if (!b) + return XML_ERROR_NO_MEMORY; + b->uri = (XML_Char *)MALLOC(sizeof(XML_Char) * (len + EXPAND_SPARE)); + if (!b->uri) { + FREE(b); + return XML_ERROR_NO_MEMORY; + } + b->uriAlloc = len + EXPAND_SPARE; + } + b->uriLen = len; + memcpy(b->uri, uri, len * sizeof(XML_Char)); + if (namespaceSeparator) + b->uri[len - 1] = namespaceSeparator; + b->prefix = prefix; + b->attId = attId; + b->prevPrefixBinding = prefix->binding; + /* NULL binding when default namespace undeclared */ + if (*uri == XML_T('\0') && prefix == &_dtd->defaultPrefix) + prefix->binding = NULL; + else + prefix->binding = b; + b->nextTagBinding = *bindingsPtr; + *bindingsPtr = b; + /* if attId == NULL then we are not starting a namespace scope */ + if (attId && startNamespaceDeclHandler) + startNamespaceDeclHandler(handlerArg, prefix->name, + prefix->binding ? uri : 0); + return XML_ERROR_NONE; +} + +/* The idea here is to avoid using stack for each CDATA section when + the whole file is parsed with one call. +*/ +static enum XML_Error PTRCALL +cdataSectionProcessor(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr) +{ + enum XML_Error result = doCdataSection(parser, encoding, &start, end, + endPtr, (XML_Bool)!ps_finalBuffer); + if (result != XML_ERROR_NONE) + return result; + if (start) { + if (parentParser) { /* we are parsing an external entity */ + processor = externalEntityContentProcessor; + return externalEntityContentProcessor(parser, start, end, endPtr); + } + else { + processor = contentProcessor; + return contentProcessor(parser, start, end, endPtr); + } + } + return result; +} + +/* startPtr gets set to non-null if the section is closed, and to null if + the section is not yet closed. +*/ +static enum XML_Error +doCdataSection(XML_Parser parser, + const ENCODING *enc, + const char **startPtr, + const char *end, + const char **nextPtr, + XML_Bool haveMore) +{ + const char *s = *startPtr; + const char **eventPP; + const char **eventEndPP; + if (enc == encoding) { + eventPP = &eventPtr; + *eventPP = s; + eventEndPP = &eventEndPtr; + } + else { + eventPP = &(openInternalEntities->internalEventPtr); + eventEndPP = &(openInternalEntities->internalEventEndPtr); + } + *eventPP = s; + *startPtr = NULL; + + for (;;) { + const char *next; + int tok = XmlCdataSectionTok(enc, s, end, &next); + *eventEndPP = next; + switch (tok) { + case XML_TOK_CDATA_SECT_CLOSE: + if (endCdataSectionHandler) + endCdataSectionHandler(handlerArg); +#if 0 + /* see comment under XML_TOK_CDATA_SECT_OPEN */ + else if (characterDataHandler) + characterDataHandler(handlerArg, dataBuf, 0); +#endif + else if (defaultHandler) + reportDefault(parser, enc, s, next); + *startPtr = next; + *nextPtr = next; + if (ps_parsing == XML_FINISHED) + return XML_ERROR_ABORTED; + else + return XML_ERROR_NONE; + case XML_TOK_DATA_NEWLINE: + if (characterDataHandler) { + XML_Char c = 0xA; + characterDataHandler(handlerArg, &c, 1); + } + else if (defaultHandler) + reportDefault(parser, enc, s, next); + break; + case XML_TOK_DATA_CHARS: + { + XML_CharacterDataHandler charDataHandler = characterDataHandler; + if (charDataHandler) { + if (MUST_CONVERT(enc, s)) { + for (;;) { + ICHAR *dataPtr = (ICHAR *)dataBuf; + XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); + *eventEndPP = next; + charDataHandler(handlerArg, dataBuf, + (int)(dataPtr - (ICHAR *)dataBuf)); + if (s == next) + break; + *eventPP = s; + } + } + else + charDataHandler(handlerArg, + (XML_Char *)s, + (int)((XML_Char *)next - (XML_Char *)s)); + } + else if (defaultHandler) + reportDefault(parser, enc, s, next); + } + break; + case XML_TOK_INVALID: + *eventPP = next; + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL_CHAR: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_PARTIAL_CHAR; + case XML_TOK_PARTIAL: + case XML_TOK_NONE: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_UNCLOSED_CDATA_SECTION; + default: + *eventPP = next; + return XML_ERROR_UNEXPECTED_STATE; + } + + *eventPP = s = next; + switch (ps_parsing) { + case XML_SUSPENDED: + *nextPtr = next; + return XML_ERROR_NONE; + case XML_FINISHED: + return XML_ERROR_ABORTED; + default: ; + } + } + /* not reached */ +} + +#ifdef XML_DTD + +/* The idea here is to avoid using stack for each IGNORE section when + the whole file is parsed with one call. +*/ +static enum XML_Error PTRCALL +ignoreSectionProcessor(XML_Parser parser, + const char *start, + const char *end, + const char **endPtr) +{ + enum XML_Error result = doIgnoreSection(parser, encoding, &start, end, + endPtr, (XML_Bool)!ps_finalBuffer); + if (result != XML_ERROR_NONE) + return result; + if (start) { + processor = prologProcessor; + return prologProcessor(parser, start, end, endPtr); + } + return result; +} + +/* startPtr gets set to non-null is the section is closed, and to null + if the section is not yet closed. +*/ +static enum XML_Error +doIgnoreSection(XML_Parser parser, + const ENCODING *enc, + const char **startPtr, + const char *end, + const char **nextPtr, + XML_Bool haveMore) +{ + const char *next; + int tok; + const char *s = *startPtr; + const char **eventPP; + const char **eventEndPP; + if (enc == encoding) { + eventPP = &eventPtr; + *eventPP = s; + eventEndPP = &eventEndPtr; + } + else { + eventPP = &(openInternalEntities->internalEventPtr); + eventEndPP = &(openInternalEntities->internalEventEndPtr); + } + *eventPP = s; + *startPtr = NULL; + tok = XmlIgnoreSectionTok(enc, s, end, &next); + *eventEndPP = next; + switch (tok) { + case XML_TOK_IGNORE_SECT: + if (defaultHandler) + reportDefault(parser, enc, s, next); + *startPtr = next; + *nextPtr = next; + if (ps_parsing == XML_FINISHED) + return XML_ERROR_ABORTED; + else + return XML_ERROR_NONE; + case XML_TOK_INVALID: + *eventPP = next; + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL_CHAR: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_PARTIAL_CHAR; + case XML_TOK_PARTIAL: + case XML_TOK_NONE: + if (haveMore) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_SYNTAX; /* XML_ERROR_UNCLOSED_IGNORE_SECTION */ + default: + *eventPP = next; + return XML_ERROR_UNEXPECTED_STATE; + } + /* not reached */ +} + +#endif /* XML_DTD */ + +static enum XML_Error +initializeEncoding(XML_Parser parser) +{ + const char *s; +#ifdef XML_UNICODE + char encodingBuf[128]; + if (!protocolEncodingName) + s = NULL; + else { + int i; + for (i = 0; protocolEncodingName[i]; i++) { + if (i == sizeof(encodingBuf) - 1 + || (protocolEncodingName[i] & ~0x7f) != 0) { + encodingBuf[0] = '\0'; + break; + } + encodingBuf[i] = (char)protocolEncodingName[i]; + } + encodingBuf[i] = '\0'; + s = encodingBuf; + } +#else + s = protocolEncodingName; +#endif + if ((ns ? XmlInitEncodingNS : XmlInitEncoding)(&initEncoding, &encoding, s)) + return XML_ERROR_NONE; + return handleUnknownEncoding(parser, protocolEncodingName); +} + +static enum XML_Error +processXmlDecl(XML_Parser parser, int isGeneralTextEntity, + const char *s, const char *next) +{ + const char *encodingName = NULL; + const XML_Char *storedEncName = NULL; + const ENCODING *newEncoding = NULL; + const char *version = NULL; + const char *versionend; + const XML_Char *storedversion = NULL; + int standalone = -1; + if (!(ns + ? XmlParseXmlDeclNS + : XmlParseXmlDecl)(isGeneralTextEntity, + encoding, + s, + next, + &eventPtr, + &version, + &versionend, + &encodingName, + &newEncoding, + &standalone)) { + if (isGeneralTextEntity) + return XML_ERROR_TEXT_DECL; + else + return XML_ERROR_XML_DECL; + } + if (!isGeneralTextEntity && standalone == 1) { + _dtd->standalone = XML_TRUE; +#ifdef XML_DTD + if (paramEntityParsing == XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE) + paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER; +#endif /* XML_DTD */ + } + if (xmlDeclHandler) { + if (encodingName != NULL) { + storedEncName = poolStoreString(&temp2Pool, + encoding, + encodingName, + encodingName + + XmlNameLength(encoding, encodingName)); + if (!storedEncName) + return XML_ERROR_NO_MEMORY; + poolFinish(&temp2Pool); + } + if (version) { + storedversion = poolStoreString(&temp2Pool, + encoding, + version, + versionend - encoding->minBytesPerChar); + if (!storedversion) + return XML_ERROR_NO_MEMORY; + } + xmlDeclHandler(handlerArg, storedversion, storedEncName, standalone); + } + else if (defaultHandler) + reportDefault(parser, encoding, s, next); + if (protocolEncodingName == NULL) { + if (newEncoding) { + if (newEncoding->minBytesPerChar != encoding->minBytesPerChar) { + eventPtr = encodingName; + return XML_ERROR_INCORRECT_ENCODING; + } + encoding = newEncoding; + } + else if (encodingName) { + enum XML_Error result; + if (!storedEncName) { + storedEncName = poolStoreString( + &temp2Pool, encoding, encodingName, + encodingName + XmlNameLength(encoding, encodingName)); + if (!storedEncName) + return XML_ERROR_NO_MEMORY; + } + result = handleUnknownEncoding(parser, storedEncName); + poolClear(&temp2Pool); + if (result == XML_ERROR_UNKNOWN_ENCODING) + eventPtr = encodingName; + return result; + } + } + + if (storedEncName || storedversion) + poolClear(&temp2Pool); + + return XML_ERROR_NONE; +} + +static enum XML_Error +handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName) +{ + if (unknownEncodingHandler) { + XML_Encoding info; + int i; + for (i = 0; i < 256; i++) + info.map[i] = -1; + info.convert = NULL; + info.data = NULL; + info.release = NULL; + if (unknownEncodingHandler(unknownEncodingHandlerData, encodingName, + &info)) { + ENCODING *enc; + unknownEncodingMem = MALLOC(XmlSizeOfUnknownEncoding()); + if (!unknownEncodingMem) { + if (info.release) + info.release(info.data); + return XML_ERROR_NO_MEMORY; + } + enc = (ns + ? XmlInitUnknownEncodingNS + : XmlInitUnknownEncoding)(unknownEncodingMem, + info.map, + info.convert, + info.data); + if (enc) { + unknownEncodingData = info.data; + unknownEncodingRelease = info.release; + encoding = enc; + return XML_ERROR_NONE; + } + } + if (info.release != NULL) + info.release(info.data); + } + return XML_ERROR_UNKNOWN_ENCODING; +} + +static enum XML_Error PTRCALL +prologInitProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + enum XML_Error result = initializeEncoding(parser); + if (result != XML_ERROR_NONE) + return result; + processor = prologProcessor; + return prologProcessor(parser, s, end, nextPtr); +} + +#ifdef XML_DTD + +static enum XML_Error PTRCALL +externalParEntInitProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + enum XML_Error result = initializeEncoding(parser); + if (result != XML_ERROR_NONE) + return result; + + /* we know now that XML_Parse(Buffer) has been called, + so we consider the external parameter entity read */ + _dtd->paramEntityRead = XML_TRUE; + + if (prologState.inEntityValue) { + processor = entityValueInitProcessor; + return entityValueInitProcessor(parser, s, end, nextPtr); + } + else { + processor = externalParEntProcessor; + return externalParEntProcessor(parser, s, end, nextPtr); + } +} + +static enum XML_Error PTRCALL +entityValueInitProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + int tok; + const char *start = s; + const char *next = start; + eventPtr = start; + + for (;;) { + tok = XmlPrologTok(encoding, start, end, &next); + eventEndPtr = next; + if (tok <= 0) { + if (!ps_finalBuffer && tok != XML_TOK_INVALID) { + *nextPtr = s; + return XML_ERROR_NONE; + } + switch (tok) { + case XML_TOK_INVALID: + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL: + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + return XML_ERROR_PARTIAL_CHAR; + case XML_TOK_NONE: /* start == end */ + default: + break; + } + /* found end of entity value - can store it now */ + return storeEntityValue(parser, encoding, s, end); + } + else if (tok == XML_TOK_XML_DECL) { + enum XML_Error result; + result = processXmlDecl(parser, 0, start, next); + if (result != XML_ERROR_NONE) + return result; + switch (ps_parsing) { + case XML_SUSPENDED: + *nextPtr = next; + return XML_ERROR_NONE; + case XML_FINISHED: + return XML_ERROR_ABORTED; + default: + *nextPtr = next; + } + /* stop scanning for text declaration - we found one */ + processor = entityValueProcessor; + return entityValueProcessor(parser, next, end, nextPtr); + } + /* If we are at the end of the buffer, this would cause XmlPrologTok to + return XML_TOK_NONE on the next call, which would then cause the + function to exit with *nextPtr set to s - that is what we want for other + tokens, but not for the BOM - we would rather like to skip it; + then, when this routine is entered the next time, XmlPrologTok will + return XML_TOK_INVALID, since the BOM is still in the buffer + */ + else if (tok == XML_TOK_BOM && next == end && !ps_finalBuffer) { + *nextPtr = next; + return XML_ERROR_NONE; + } + start = next; + eventPtr = start; + } +} + +static enum XML_Error PTRCALL +externalParEntProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + const char *next = s; + int tok; + + tok = XmlPrologTok(encoding, s, end, &next); + if (tok <= 0) { + if (!ps_finalBuffer && tok != XML_TOK_INVALID) { + *nextPtr = s; + return XML_ERROR_NONE; + } + switch (tok) { + case XML_TOK_INVALID: + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL: + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + return XML_ERROR_PARTIAL_CHAR; + case XML_TOK_NONE: /* start == end */ + default: + break; + } + } + /* This would cause the next stage, i.e. doProlog to be passed XML_TOK_BOM. + However, when parsing an external subset, doProlog will not accept a BOM + as valid, and report a syntax error, so we have to skip the BOM + */ + else if (tok == XML_TOK_BOM) { + s = next; + tok = XmlPrologTok(encoding, s, end, &next); + } + + processor = prologProcessor; + return doProlog(parser, encoding, s, end, tok, next, + nextPtr, (XML_Bool)!ps_finalBuffer); +} + +static enum XML_Error PTRCALL +entityValueProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + const char *start = s; + const char *next = s; + const ENCODING *enc = encoding; + int tok; + + for (;;) { + tok = XmlPrologTok(enc, start, end, &next); + if (tok <= 0) { + if (!ps_finalBuffer && tok != XML_TOK_INVALID) { + *nextPtr = s; + return XML_ERROR_NONE; + } + switch (tok) { + case XML_TOK_INVALID: + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL: + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + return XML_ERROR_PARTIAL_CHAR; + case XML_TOK_NONE: /* start == end */ + default: + break; + } + /* found end of entity value - can store it now */ + return storeEntityValue(parser, enc, s, end); + } + start = next; + } +} + +#endif /* XML_DTD */ + +static enum XML_Error PTRCALL +prologProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + const char *next = s; + int tok = XmlPrologTok(encoding, s, end, &next); + return doProlog(parser, encoding, s, end, tok, next, + nextPtr, (XML_Bool)!ps_finalBuffer); +} + +static enum XML_Error +doProlog(XML_Parser parser, + const ENCODING *enc, + const char *s, + const char *end, + int tok, + const char *next, + const char **nextPtr, + XML_Bool haveMore) +{ +#ifdef XML_DTD + static const XML_Char externalSubsetName[] = { ASCII_HASH , '\0' }; +#endif /* XML_DTD */ + static const XML_Char atypeCDATA[] = + { ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' }; + static const XML_Char atypeID[] = { ASCII_I, ASCII_D, '\0' }; + static const XML_Char atypeIDREF[] = + { ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, '\0' }; + static const XML_Char atypeIDREFS[] = + { ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, ASCII_S, '\0' }; + static const XML_Char atypeENTITY[] = + { ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_Y, '\0' }; + static const XML_Char atypeENTITIES[] = { ASCII_E, ASCII_N, + ASCII_T, ASCII_I, ASCII_T, ASCII_I, ASCII_E, ASCII_S, '\0' }; + static const XML_Char atypeNMTOKEN[] = { + ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, '\0' }; + static const XML_Char atypeNMTOKENS[] = { ASCII_N, ASCII_M, ASCII_T, + ASCII_O, ASCII_K, ASCII_E, ASCII_N, ASCII_S, '\0' }; + static const XML_Char notationPrefix[] = { ASCII_N, ASCII_O, ASCII_T, + ASCII_A, ASCII_T, ASCII_I, ASCII_O, ASCII_N, ASCII_LPAREN, '\0' }; + static const XML_Char enumValueSep[] = { ASCII_PIPE, '\0' }; + static const XML_Char enumValueStart[] = { ASCII_LPAREN, '\0' }; + + /* save one level of indirection */ + DTD * const dtd = _dtd; + + const char **eventPP; + const char **eventEndPP; + enum XML_Content_Quant quant; + + if (enc == encoding) { + eventPP = &eventPtr; + eventEndPP = &eventEndPtr; + } + else { + eventPP = &(openInternalEntities->internalEventPtr); + eventEndPP = &(openInternalEntities->internalEventEndPtr); + } + + for (;;) { + int role; + XML_Bool handleDefault = XML_TRUE; + *eventPP = s; + *eventEndPP = next; + if (tok <= 0) { + if (haveMore && tok != XML_TOK_INVALID) { + *nextPtr = s; + return XML_ERROR_NONE; + } + switch (tok) { + case XML_TOK_INVALID: + *eventPP = next; + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL: + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + return XML_ERROR_PARTIAL_CHAR; + case -XML_TOK_PROLOG_S: + tok = -tok; + break; + case XML_TOK_NONE: +#ifdef XML_DTD + /* for internal PE NOT referenced between declarations */ + if (enc != encoding && !openInternalEntities->betweenDecl) { + *nextPtr = s; + return XML_ERROR_NONE; + } + /* WFC: PE Between Declarations - must check that PE contains + complete markup, not only for external PEs, but also for + internal PEs if the reference occurs between declarations. + */ + if (isParamEntity || enc != encoding) { + if (XmlTokenRole(&prologState, XML_TOK_NONE, end, end, enc) + == XML_ROLE_ERROR) + return XML_ERROR_INCOMPLETE_PE; + *nextPtr = s; + return XML_ERROR_NONE; + } +#endif /* XML_DTD */ + return XML_ERROR_NO_ELEMENTS; + default: + tok = -tok; + next = end; + break; + } + } + role = XmlTokenRole(&prologState, tok, s, next, enc); + switch (role) { + case XML_ROLE_XML_DECL: + { + enum XML_Error result = processXmlDecl(parser, 0, s, next); + if (result != XML_ERROR_NONE) + return result; + enc = encoding; + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_DOCTYPE_NAME: + if (startDoctypeDeclHandler) { + doctypeName = poolStoreString(&tempPool, enc, s, next); + if (!doctypeName) + return XML_ERROR_NO_MEMORY; + poolFinish(&tempPool); + doctypePubid = NULL; + handleDefault = XML_FALSE; + } + doctypeSysid = NULL; /* always initialize to NULL */ + break; + case XML_ROLE_DOCTYPE_INTERNAL_SUBSET: + if (startDoctypeDeclHandler) { + startDoctypeDeclHandler(handlerArg, doctypeName, doctypeSysid, + doctypePubid, 1); + doctypeName = NULL; + poolClear(&tempPool); + handleDefault = XML_FALSE; + } + break; +#ifdef XML_DTD + case XML_ROLE_TEXT_DECL: + { + enum XML_Error result = processXmlDecl(parser, 1, s, next); + if (result != XML_ERROR_NONE) + return result; + enc = encoding; + handleDefault = XML_FALSE; + } + break; +#endif /* XML_DTD */ + case XML_ROLE_DOCTYPE_PUBLIC_ID: +#ifdef XML_DTD + useForeignDTD = XML_FALSE; + declEntity = (ENTITY *)lookup(parser, + &dtd->paramEntities, + externalSubsetName, + sizeof(ENTITY)); + if (!declEntity) + return XML_ERROR_NO_MEMORY; +#endif /* XML_DTD */ + dtd->hasParamEntityRefs = XML_TRUE; + if (startDoctypeDeclHandler) { + XML_Char *pubId; + if (!XmlIsPublicId(enc, s, next, eventPP)) + return XML_ERROR_PUBLICID; + pubId = poolStoreString(&tempPool, enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!pubId) + return XML_ERROR_NO_MEMORY; + normalizePublicId(pubId); + poolFinish(&tempPool); + doctypePubid = pubId; + handleDefault = XML_FALSE; + goto alreadyChecked; + } + /* fall through */ + case XML_ROLE_ENTITY_PUBLIC_ID: + if (!XmlIsPublicId(enc, s, next, eventPP)) + return XML_ERROR_PUBLICID; + alreadyChecked: + if (dtd->keepProcessing && declEntity) { + XML_Char *tem = poolStoreString(&dtd->pool, + enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!tem) + return XML_ERROR_NO_MEMORY; + normalizePublicId(tem); + declEntity->publicId = tem; + poolFinish(&dtd->pool); + if (entityDeclHandler) + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_DOCTYPE_CLOSE: + if (doctypeName) { + startDoctypeDeclHandler(handlerArg, doctypeName, + doctypeSysid, doctypePubid, 0); + poolClear(&tempPool); + handleDefault = XML_FALSE; + } + /* doctypeSysid will be non-NULL in the case of a previous + XML_ROLE_DOCTYPE_SYSTEM_ID, even if startDoctypeDeclHandler + was not set, indicating an external subset + */ +#ifdef XML_DTD + if (doctypeSysid || useForeignDTD) { + XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs; + dtd->hasParamEntityRefs = XML_TRUE; + if (paramEntityParsing && externalEntityRefHandler) { + ENTITY *entity = (ENTITY *)lookup(parser, + &dtd->paramEntities, + externalSubsetName, + sizeof(ENTITY)); + if (!entity) + return XML_ERROR_NO_MEMORY; + if (useForeignDTD) + entity->base = curBase; + dtd->paramEntityRead = XML_FALSE; + if (!externalEntityRefHandler(externalEntityRefHandlerArg, + 0, + entity->base, + entity->systemId, + entity->publicId)) + return XML_ERROR_EXTERNAL_ENTITY_HANDLING; + if (dtd->paramEntityRead) { + if (!dtd->standalone && + notStandaloneHandler && + !notStandaloneHandler(handlerArg)) + return XML_ERROR_NOT_STANDALONE; + } + /* if we didn't read the foreign DTD then this means that there + is no external subset and we must reset dtd->hasParamEntityRefs + */ + else if (!doctypeSysid) + dtd->hasParamEntityRefs = hadParamEntityRefs; + /* end of DTD - no need to update dtd->keepProcessing */ + } + useForeignDTD = XML_FALSE; + } +#endif /* XML_DTD */ + if (endDoctypeDeclHandler) { + endDoctypeDeclHandler(handlerArg); + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_INSTANCE_START: +#ifdef XML_DTD + /* if there is no DOCTYPE declaration then now is the + last chance to read the foreign DTD + */ + if (useForeignDTD) { + XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs; + dtd->hasParamEntityRefs = XML_TRUE; + if (paramEntityParsing && externalEntityRefHandler) { + ENTITY *entity = (ENTITY *)lookup(parser, &dtd->paramEntities, + externalSubsetName, + sizeof(ENTITY)); + if (!entity) + return XML_ERROR_NO_MEMORY; + entity->base = curBase; + dtd->paramEntityRead = XML_FALSE; + if (!externalEntityRefHandler(externalEntityRefHandlerArg, + 0, + entity->base, + entity->systemId, + entity->publicId)) + return XML_ERROR_EXTERNAL_ENTITY_HANDLING; + if (dtd->paramEntityRead) { + if (!dtd->standalone && + notStandaloneHandler && + !notStandaloneHandler(handlerArg)) + return XML_ERROR_NOT_STANDALONE; + } + /* if we didn't read the foreign DTD then this means that there + is no external subset and we must reset dtd->hasParamEntityRefs + */ + else + dtd->hasParamEntityRefs = hadParamEntityRefs; + /* end of DTD - no need to update dtd->keepProcessing */ + } + } +#endif /* XML_DTD */ + processor = contentProcessor; + return contentProcessor(parser, s, end, nextPtr); + case XML_ROLE_ATTLIST_ELEMENT_NAME: + declElementType = getElementType(parser, enc, s, next); + if (!declElementType) + return XML_ERROR_NO_MEMORY; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_NAME: + declAttributeId = getAttributeId(parser, enc, s, next); + if (!declAttributeId) + return XML_ERROR_NO_MEMORY; + declAttributeIsCdata = XML_FALSE; + declAttributeType = NULL; + declAttributeIsId = XML_FALSE; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_CDATA: + declAttributeIsCdata = XML_TRUE; + declAttributeType = atypeCDATA; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_ID: + declAttributeIsId = XML_TRUE; + declAttributeType = atypeID; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_IDREF: + declAttributeType = atypeIDREF; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_IDREFS: + declAttributeType = atypeIDREFS; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_ENTITY: + declAttributeType = atypeENTITY; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_ENTITIES: + declAttributeType = atypeENTITIES; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN: + declAttributeType = atypeNMTOKEN; + goto checkAttListDeclHandler; + case XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS: + declAttributeType = atypeNMTOKENS; + checkAttListDeclHandler: + if (dtd->keepProcessing && attlistDeclHandler) + handleDefault = XML_FALSE; + break; + case XML_ROLE_ATTRIBUTE_ENUM_VALUE: + case XML_ROLE_ATTRIBUTE_NOTATION_VALUE: + if (dtd->keepProcessing && attlistDeclHandler) { + const XML_Char *prefix; + if (declAttributeType) { + prefix = enumValueSep; + } + else { + prefix = (role == XML_ROLE_ATTRIBUTE_NOTATION_VALUE + ? notationPrefix + : enumValueStart); + } + if (!poolAppendString(&tempPool, prefix)) + return XML_ERROR_NO_MEMORY; + if (!poolAppend(&tempPool, enc, s, next)) + return XML_ERROR_NO_MEMORY; + declAttributeType = tempPool.start; + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_IMPLIED_ATTRIBUTE_VALUE: + case XML_ROLE_REQUIRED_ATTRIBUTE_VALUE: + if (dtd->keepProcessing) { + if (!defineAttribute(declElementType, declAttributeId, + declAttributeIsCdata, declAttributeIsId, + 0, parser)) + return XML_ERROR_NO_MEMORY; + if (attlistDeclHandler && declAttributeType) { + if (*declAttributeType == XML_T(ASCII_LPAREN) + || (*declAttributeType == XML_T(ASCII_N) + && declAttributeType[1] == XML_T(ASCII_O))) { + /* Enumerated or Notation type */ + if (!poolAppendChar(&tempPool, XML_T(ASCII_RPAREN)) + || !poolAppendChar(&tempPool, XML_T('\0'))) + return XML_ERROR_NO_MEMORY; + declAttributeType = tempPool.start; + poolFinish(&tempPool); + } + *eventEndPP = s; + attlistDeclHandler(handlerArg, declElementType->name, + declAttributeId->name, declAttributeType, + 0, role == XML_ROLE_REQUIRED_ATTRIBUTE_VALUE); + poolClear(&tempPool); + handleDefault = XML_FALSE; + } + } + break; + case XML_ROLE_DEFAULT_ATTRIBUTE_VALUE: + case XML_ROLE_FIXED_ATTRIBUTE_VALUE: + if (dtd->keepProcessing) { + const XML_Char *attVal; + enum XML_Error result = + storeAttributeValue(parser, enc, declAttributeIsCdata, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar, + &dtd->pool); + if (result) + return result; + attVal = poolStart(&dtd->pool); + poolFinish(&dtd->pool); + /* ID attributes aren't allowed to have a default */ + if (!defineAttribute(declElementType, declAttributeId, + declAttributeIsCdata, XML_FALSE, attVal, parser)) + return XML_ERROR_NO_MEMORY; + if (attlistDeclHandler && declAttributeType) { + if (*declAttributeType == XML_T(ASCII_LPAREN) + || (*declAttributeType == XML_T(ASCII_N) + && declAttributeType[1] == XML_T(ASCII_O))) { + /* Enumerated or Notation type */ + if (!poolAppendChar(&tempPool, XML_T(ASCII_RPAREN)) + || !poolAppendChar(&tempPool, XML_T('\0'))) + return XML_ERROR_NO_MEMORY; + declAttributeType = tempPool.start; + poolFinish(&tempPool); + } + *eventEndPP = s; + attlistDeclHandler(handlerArg, declElementType->name, + declAttributeId->name, declAttributeType, + attVal, + role == XML_ROLE_FIXED_ATTRIBUTE_VALUE); + poolClear(&tempPool); + handleDefault = XML_FALSE; + } + } + break; + case XML_ROLE_ENTITY_VALUE: + if (dtd->keepProcessing) { + enum XML_Error result = storeEntityValue(parser, enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (declEntity) { + declEntity->textPtr = poolStart(&dtd->entityValuePool); + declEntity->textLen = (int)(poolLength(&dtd->entityValuePool)); + poolFinish(&dtd->entityValuePool); + if (entityDeclHandler) { + *eventEndPP = s; + entityDeclHandler(handlerArg, + declEntity->name, + declEntity->is_param, + declEntity->textPtr, + declEntity->textLen, + curBase, 0, 0, 0); + handleDefault = XML_FALSE; + } + } + else + poolDiscard(&dtd->entityValuePool); + if (result != XML_ERROR_NONE) + return result; + } + break; + case XML_ROLE_DOCTYPE_SYSTEM_ID: +#ifdef XML_DTD + useForeignDTD = XML_FALSE; +#endif /* XML_DTD */ + dtd->hasParamEntityRefs = XML_TRUE; + if (startDoctypeDeclHandler) { + doctypeSysid = poolStoreString(&tempPool, enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (doctypeSysid == NULL) + return XML_ERROR_NO_MEMORY; + poolFinish(&tempPool); + handleDefault = XML_FALSE; + } +#ifdef XML_DTD + else + /* use externalSubsetName to make doctypeSysid non-NULL + for the case where no startDoctypeDeclHandler is set */ + doctypeSysid = externalSubsetName; +#endif /* XML_DTD */ + if (!dtd->standalone +#ifdef XML_DTD + && !paramEntityParsing +#endif /* XML_DTD */ + && notStandaloneHandler + && !notStandaloneHandler(handlerArg)) + return XML_ERROR_NOT_STANDALONE; +#ifndef XML_DTD + break; +#else /* XML_DTD */ + if (!declEntity) { + declEntity = (ENTITY *)lookup(parser, + &dtd->paramEntities, + externalSubsetName, + sizeof(ENTITY)); + if (!declEntity) + return XML_ERROR_NO_MEMORY; + declEntity->publicId = NULL; + } + /* fall through */ +#endif /* XML_DTD */ + case XML_ROLE_ENTITY_SYSTEM_ID: + if (dtd->keepProcessing && declEntity) { + declEntity->systemId = poolStoreString(&dtd->pool, enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!declEntity->systemId) + return XML_ERROR_NO_MEMORY; + declEntity->base = curBase; + poolFinish(&dtd->pool); + if (entityDeclHandler) + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_ENTITY_COMPLETE: + if (dtd->keepProcessing && declEntity && entityDeclHandler) { + *eventEndPP = s; + entityDeclHandler(handlerArg, + declEntity->name, + declEntity->is_param, + 0,0, + declEntity->base, + declEntity->systemId, + declEntity->publicId, + 0); + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_ENTITY_NOTATION_NAME: + if (dtd->keepProcessing && declEntity) { + declEntity->notation = poolStoreString(&dtd->pool, enc, s, next); + if (!declEntity->notation) + return XML_ERROR_NO_MEMORY; + poolFinish(&dtd->pool); + if (unparsedEntityDeclHandler) { + *eventEndPP = s; + unparsedEntityDeclHandler(handlerArg, + declEntity->name, + declEntity->base, + declEntity->systemId, + declEntity->publicId, + declEntity->notation); + handleDefault = XML_FALSE; + } + else if (entityDeclHandler) { + *eventEndPP = s; + entityDeclHandler(handlerArg, + declEntity->name, + 0,0,0, + declEntity->base, + declEntity->systemId, + declEntity->publicId, + declEntity->notation); + handleDefault = XML_FALSE; + } + } + break; + case XML_ROLE_GENERAL_ENTITY_NAME: + { + if (XmlPredefinedEntityName(enc, s, next)) { + declEntity = NULL; + break; + } + if (dtd->keepProcessing) { + const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next); + if (!name) + return XML_ERROR_NO_MEMORY; + declEntity = (ENTITY *)lookup(parser, &dtd->generalEntities, name, + sizeof(ENTITY)); + if (!declEntity) + return XML_ERROR_NO_MEMORY; + if (declEntity->name != name) { + poolDiscard(&dtd->pool); + declEntity = NULL; + } + else { + poolFinish(&dtd->pool); + declEntity->publicId = NULL; + declEntity->is_param = XML_FALSE; + /* if we have a parent parser or are reading an internal parameter + entity, then the entity declaration is not considered "internal" + */ + declEntity->is_internal = !(parentParser || openInternalEntities); + if (entityDeclHandler) + handleDefault = XML_FALSE; + } + } + else { + poolDiscard(&dtd->pool); + declEntity = NULL; + } + } + break; + case XML_ROLE_PARAM_ENTITY_NAME: +#ifdef XML_DTD + if (dtd->keepProcessing) { + const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next); + if (!name) + return XML_ERROR_NO_MEMORY; + declEntity = (ENTITY *)lookup(parser, &dtd->paramEntities, + name, sizeof(ENTITY)); + if (!declEntity) + return XML_ERROR_NO_MEMORY; + if (declEntity->name != name) { + poolDiscard(&dtd->pool); + declEntity = NULL; + } + else { + poolFinish(&dtd->pool); + declEntity->publicId = NULL; + declEntity->is_param = XML_TRUE; + /* if we have a parent parser or are reading an internal parameter + entity, then the entity declaration is not considered "internal" + */ + declEntity->is_internal = !(parentParser || openInternalEntities); + if (entityDeclHandler) + handleDefault = XML_FALSE; + } + } + else { + poolDiscard(&dtd->pool); + declEntity = NULL; + } +#else /* not XML_DTD */ + declEntity = NULL; +#endif /* XML_DTD */ + break; + case XML_ROLE_NOTATION_NAME: + declNotationPublicId = NULL; + declNotationName = NULL; + if (notationDeclHandler) { + declNotationName = poolStoreString(&tempPool, enc, s, next); + if (!declNotationName) + return XML_ERROR_NO_MEMORY; + poolFinish(&tempPool); + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_NOTATION_PUBLIC_ID: + if (!XmlIsPublicId(enc, s, next, eventPP)) + return XML_ERROR_PUBLICID; + if (declNotationName) { /* means notationDeclHandler != NULL */ + XML_Char *tem = poolStoreString(&tempPool, + enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!tem) + return XML_ERROR_NO_MEMORY; + normalizePublicId(tem); + declNotationPublicId = tem; + poolFinish(&tempPool); + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_NOTATION_SYSTEM_ID: + if (declNotationName && notationDeclHandler) { + const XML_Char *systemId + = poolStoreString(&tempPool, enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!systemId) + return XML_ERROR_NO_MEMORY; + *eventEndPP = s; + notationDeclHandler(handlerArg, + declNotationName, + curBase, + systemId, + declNotationPublicId); + handleDefault = XML_FALSE; + } + poolClear(&tempPool); + break; + case XML_ROLE_NOTATION_NO_SYSTEM_ID: + if (declNotationPublicId && notationDeclHandler) { + *eventEndPP = s; + notationDeclHandler(handlerArg, + declNotationName, + curBase, + 0, + declNotationPublicId); + handleDefault = XML_FALSE; + } + poolClear(&tempPool); + break; + case XML_ROLE_ERROR: + switch (tok) { + case XML_TOK_PARAM_ENTITY_REF: + /* PE references in internal subset are + not allowed within declarations. */ + return XML_ERROR_PARAM_ENTITY_REF; + case XML_TOK_XML_DECL: + return XML_ERROR_MISPLACED_XML_PI; + default: + return XML_ERROR_SYNTAX; + } +#ifdef XML_DTD + case XML_ROLE_IGNORE_SECT: + { + enum XML_Error result; + if (defaultHandler) + reportDefault(parser, enc, s, next); + handleDefault = XML_FALSE; + result = doIgnoreSection(parser, enc, &next, end, nextPtr, haveMore); + if (result != XML_ERROR_NONE) + return result; + else if (!next) { + processor = ignoreSectionProcessor; + return result; + } + } + break; +#endif /* XML_DTD */ + case XML_ROLE_GROUP_OPEN: + if (prologState.level >= groupSize) { + if (groupSize) { + char *temp = (char *)REALLOC(groupConnector, groupSize *= 2); + if (temp == NULL) + return XML_ERROR_NO_MEMORY; + groupConnector = temp; + if (dtd->scaffIndex) { + int *temp = (int *)REALLOC(dtd->scaffIndex, + groupSize * sizeof(int)); + if (temp == NULL) + return XML_ERROR_NO_MEMORY; + dtd->scaffIndex = temp; + } + } + else { + groupConnector = (char *)MALLOC(groupSize = 32); + if (!groupConnector) + return XML_ERROR_NO_MEMORY; + } + } + groupConnector[prologState.level] = 0; + if (dtd->in_eldecl) { + int myindex = nextScaffoldPart(parser); + if (myindex < 0) + return XML_ERROR_NO_MEMORY; + dtd->scaffIndex[dtd->scaffLevel] = myindex; + dtd->scaffLevel++; + dtd->scaffold[myindex].type = XML_CTYPE_SEQ; + if (elementDeclHandler) + handleDefault = XML_FALSE; + } + break; + case XML_ROLE_GROUP_SEQUENCE: + if (groupConnector[prologState.level] == ASCII_PIPE) + return XML_ERROR_SYNTAX; + groupConnector[prologState.level] = ASCII_COMMA; + if (dtd->in_eldecl && elementDeclHandler) + handleDefault = XML_FALSE; + break; + case XML_ROLE_GROUP_CHOICE: + if (groupConnector[prologState.level] == ASCII_COMMA) + return XML_ERROR_SYNTAX; + if (dtd->in_eldecl + && !groupConnector[prologState.level] + && (dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type + != XML_CTYPE_MIXED) + ) { + dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type + = XML_CTYPE_CHOICE; + if (elementDeclHandler) + handleDefault = XML_FALSE; + } + groupConnector[prologState.level] = ASCII_PIPE; + break; + case XML_ROLE_PARAM_ENTITY_REF: +#ifdef XML_DTD + case XML_ROLE_INNER_PARAM_ENTITY_REF: + dtd->hasParamEntityRefs = XML_TRUE; + if (!paramEntityParsing) + dtd->keepProcessing = dtd->standalone; + else { + const XML_Char *name; + ENTITY *entity; + name = poolStoreString(&dtd->pool, enc, + s + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!name) + return XML_ERROR_NO_MEMORY; + entity = (ENTITY *)lookup(parser, &dtd->paramEntities, name, 0); + poolDiscard(&dtd->pool); + /* first, determine if a check for an existing declaration is needed; + if yes, check that the entity exists, and that it is internal, + otherwise call the skipped entity handler + */ + if (prologState.documentEntity && + (dtd->standalone + ? !openInternalEntities + : !dtd->hasParamEntityRefs)) { + if (!entity) + return XML_ERROR_UNDEFINED_ENTITY; + else if (!entity->is_internal) + return XML_ERROR_ENTITY_DECLARED_IN_PE; + } + else if (!entity) { + dtd->keepProcessing = dtd->standalone; + /* cannot report skipped entities in declarations */ + if ((role == XML_ROLE_PARAM_ENTITY_REF) && skippedEntityHandler) { + skippedEntityHandler(handlerArg, name, 1); + handleDefault = XML_FALSE; + } + break; + } + if (entity->open) + return XML_ERROR_RECURSIVE_ENTITY_REF; + if (entity->textPtr) { + enum XML_Error result; + XML_Bool betweenDecl = + (role == XML_ROLE_PARAM_ENTITY_REF ? XML_TRUE : XML_FALSE); + result = processInternalEntity(parser, entity, betweenDecl); + if (result != XML_ERROR_NONE) + return result; + handleDefault = XML_FALSE; + break; + } + if (externalEntityRefHandler) { + dtd->paramEntityRead = XML_FALSE; + entity->open = XML_TRUE; + if (!externalEntityRefHandler(externalEntityRefHandlerArg, + 0, + entity->base, + entity->systemId, + entity->publicId)) { + entity->open = XML_FALSE; + return XML_ERROR_EXTERNAL_ENTITY_HANDLING; + } + entity->open = XML_FALSE; + handleDefault = XML_FALSE; + if (!dtd->paramEntityRead) { + dtd->keepProcessing = dtd->standalone; + break; + } + } + else { + dtd->keepProcessing = dtd->standalone; + break; + } + } +#endif /* XML_DTD */ + if (!dtd->standalone && + notStandaloneHandler && + !notStandaloneHandler(handlerArg)) + return XML_ERROR_NOT_STANDALONE; + break; + + /* Element declaration stuff */ + + case XML_ROLE_ELEMENT_NAME: + if (elementDeclHandler) { + declElementType = getElementType(parser, enc, s, next); + if (!declElementType) + return XML_ERROR_NO_MEMORY; + dtd->scaffLevel = 0; + dtd->scaffCount = 0; + dtd->in_eldecl = XML_TRUE; + handleDefault = XML_FALSE; + } + break; + + case XML_ROLE_CONTENT_ANY: + case XML_ROLE_CONTENT_EMPTY: + if (dtd->in_eldecl) { + if (elementDeclHandler) { + XML_Content * content = (XML_Content *) MALLOC(sizeof(XML_Content)); + if (!content) + return XML_ERROR_NO_MEMORY; + content->quant = XML_CQUANT_NONE; + content->name = NULL; + content->numchildren = 0; + content->children = NULL; + content->type = ((role == XML_ROLE_CONTENT_ANY) ? + XML_CTYPE_ANY : + XML_CTYPE_EMPTY); + *eventEndPP = s; + elementDeclHandler(handlerArg, declElementType->name, content); + handleDefault = XML_FALSE; + } + dtd->in_eldecl = XML_FALSE; + } + break; + + case XML_ROLE_CONTENT_PCDATA: + if (dtd->in_eldecl) { + dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type + = XML_CTYPE_MIXED; + if (elementDeclHandler) + handleDefault = XML_FALSE; + } + break; + + case XML_ROLE_CONTENT_ELEMENT: + quant = XML_CQUANT_NONE; + goto elementContent; + case XML_ROLE_CONTENT_ELEMENT_OPT: + quant = XML_CQUANT_OPT; + goto elementContent; + case XML_ROLE_CONTENT_ELEMENT_REP: + quant = XML_CQUANT_REP; + goto elementContent; + case XML_ROLE_CONTENT_ELEMENT_PLUS: + quant = XML_CQUANT_PLUS; + elementContent: + if (dtd->in_eldecl) { + ELEMENT_TYPE *el; + const XML_Char *name; + int nameLen; + const char *nxt = (quant == XML_CQUANT_NONE + ? next + : next - enc->minBytesPerChar); + int myindex = nextScaffoldPart(parser); + if (myindex < 0) + return XML_ERROR_NO_MEMORY; + dtd->scaffold[myindex].type = XML_CTYPE_NAME; + dtd->scaffold[myindex].quant = quant; + el = getElementType(parser, enc, s, nxt); + if (!el) + return XML_ERROR_NO_MEMORY; + name = el->name; + dtd->scaffold[myindex].name = name; + nameLen = 0; + for (; name[nameLen++]; ); + dtd->contentStringLen += nameLen; + if (elementDeclHandler) + handleDefault = XML_FALSE; + } + break; + + case XML_ROLE_GROUP_CLOSE: + quant = XML_CQUANT_NONE; + goto closeGroup; + case XML_ROLE_GROUP_CLOSE_OPT: + quant = XML_CQUANT_OPT; + goto closeGroup; + case XML_ROLE_GROUP_CLOSE_REP: + quant = XML_CQUANT_REP; + goto closeGroup; + case XML_ROLE_GROUP_CLOSE_PLUS: + quant = XML_CQUANT_PLUS; + closeGroup: + if (dtd->in_eldecl) { + if (elementDeclHandler) + handleDefault = XML_FALSE; + dtd->scaffLevel--; + dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel]].quant = quant; + if (dtd->scaffLevel == 0) { + if (!handleDefault) { + XML_Content *model = build_model(parser); + if (!model) + return XML_ERROR_NO_MEMORY; + *eventEndPP = s; + elementDeclHandler(handlerArg, declElementType->name, model); + } + dtd->in_eldecl = XML_FALSE; + dtd->contentStringLen = 0; + } + } + break; + /* End element declaration stuff */ + + case XML_ROLE_PI: + if (!reportProcessingInstruction(parser, enc, s, next)) + return XML_ERROR_NO_MEMORY; + handleDefault = XML_FALSE; + break; + case XML_ROLE_COMMENT: + if (!reportComment(parser, enc, s, next)) + return XML_ERROR_NO_MEMORY; + handleDefault = XML_FALSE; + break; + case XML_ROLE_NONE: + switch (tok) { + case XML_TOK_BOM: + handleDefault = XML_FALSE; + break; + } + break; + case XML_ROLE_DOCTYPE_NONE: + if (startDoctypeDeclHandler) + handleDefault = XML_FALSE; + break; + case XML_ROLE_ENTITY_NONE: + if (dtd->keepProcessing && entityDeclHandler) + handleDefault = XML_FALSE; + break; + case XML_ROLE_NOTATION_NONE: + if (notationDeclHandler) + handleDefault = XML_FALSE; + break; + case XML_ROLE_ATTLIST_NONE: + if (dtd->keepProcessing && attlistDeclHandler) + handleDefault = XML_FALSE; + break; + case XML_ROLE_ELEMENT_NONE: + if (elementDeclHandler) + handleDefault = XML_FALSE; + break; + } /* end of big switch */ + + if (handleDefault && defaultHandler) + reportDefault(parser, enc, s, next); + + switch (ps_parsing) { + case XML_SUSPENDED: + *nextPtr = next; + return XML_ERROR_NONE; + case XML_FINISHED: + return XML_ERROR_ABORTED; + default: + s = next; + tok = XmlPrologTok(enc, s, end, &next); + } + } + /* not reached */ +} + +static enum XML_Error PTRCALL +epilogProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + processor = epilogProcessor; + eventPtr = s; + for (;;) { + const char *next = NULL; + int tok = XmlPrologTok(encoding, s, end, &next); + eventEndPtr = next; + switch (tok) { + /* report partial linebreak - it might be the last token */ + case -XML_TOK_PROLOG_S: + if (defaultHandler) { + reportDefault(parser, encoding, s, next); + if (ps_parsing == XML_FINISHED) + return XML_ERROR_ABORTED; + } + *nextPtr = next; + return XML_ERROR_NONE; + case XML_TOK_NONE: + *nextPtr = s; + return XML_ERROR_NONE; + case XML_TOK_PROLOG_S: + if (defaultHandler) + reportDefault(parser, encoding, s, next); + break; + case XML_TOK_PI: + if (!reportProcessingInstruction(parser, encoding, s, next)) + return XML_ERROR_NO_MEMORY; + break; + case XML_TOK_COMMENT: + if (!reportComment(parser, encoding, s, next)) + return XML_ERROR_NO_MEMORY; + break; + case XML_TOK_INVALID: + eventPtr = next; + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL: + if (!ps_finalBuffer) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_UNCLOSED_TOKEN; + case XML_TOK_PARTIAL_CHAR: + if (!ps_finalBuffer) { + *nextPtr = s; + return XML_ERROR_NONE; + } + return XML_ERROR_PARTIAL_CHAR; + default: + return XML_ERROR_JUNK_AFTER_DOC_ELEMENT; + } + eventPtr = s = next; + switch (ps_parsing) { + case XML_SUSPENDED: + *nextPtr = next; + return XML_ERROR_NONE; + case XML_FINISHED: + return XML_ERROR_ABORTED; + default: ; + } + } +} + +static enum XML_Error +processInternalEntity(XML_Parser parser, ENTITY *entity, + XML_Bool betweenDecl) +{ + const char *textStart, *textEnd; + const char *next; + enum XML_Error result; + OPEN_INTERNAL_ENTITY *openEntity; + + if (freeInternalEntities) { + openEntity = freeInternalEntities; + freeInternalEntities = openEntity->next; + } + else { + openEntity = (OPEN_INTERNAL_ENTITY *)MALLOC(sizeof(OPEN_INTERNAL_ENTITY)); + if (!openEntity) + return XML_ERROR_NO_MEMORY; + } + entity->open = XML_TRUE; + entity->processed = 0; + openEntity->next = openInternalEntities; + openInternalEntities = openEntity; + openEntity->entity = entity; + openEntity->startTagLevel = tagLevel; + openEntity->betweenDecl = betweenDecl; + openEntity->internalEventPtr = NULL; + openEntity->internalEventEndPtr = NULL; + textStart = (char *)entity->textPtr; + textEnd = (char *)(entity->textPtr + entity->textLen); + +#ifdef XML_DTD + if (entity->is_param) { + int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next); + result = doProlog(parser, internalEncoding, textStart, textEnd, tok, + next, &next, XML_FALSE); + } + else +#endif /* XML_DTD */ + result = doContent(parser, tagLevel, internalEncoding, textStart, + textEnd, &next, XML_FALSE); + + if (result == XML_ERROR_NONE) { + if (textEnd != next && ps_parsing == XML_SUSPENDED) { + entity->processed = (int)(next - textStart); + processor = internalEntityProcessor; + } + else { + entity->open = XML_FALSE; + openInternalEntities = openEntity->next; + /* put openEntity back in list of free instances */ + openEntity->next = freeInternalEntities; + freeInternalEntities = openEntity; + } + } + return result; +} + +static enum XML_Error PTRCALL +internalEntityProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + ENTITY *entity; + const char *textStart, *textEnd; + const char *next; + enum XML_Error result; + OPEN_INTERNAL_ENTITY *openEntity = openInternalEntities; + if (!openEntity) + return XML_ERROR_UNEXPECTED_STATE; + + entity = openEntity->entity; + textStart = ((char *)entity->textPtr) + entity->processed; + textEnd = (char *)(entity->textPtr + entity->textLen); + +#ifdef XML_DTD + if (entity->is_param) { + int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next); + result = doProlog(parser, internalEncoding, textStart, textEnd, tok, + next, &next, XML_FALSE); + } + else +#endif /* XML_DTD */ + result = doContent(parser, openEntity->startTagLevel, internalEncoding, + textStart, textEnd, &next, XML_FALSE); + + if (result != XML_ERROR_NONE) + return result; + else if (textEnd != next && ps_parsing == XML_SUSPENDED) { + entity->processed = (int)(next - (char *)entity->textPtr); + return result; + } + else { + entity->open = XML_FALSE; + openInternalEntities = openEntity->next; + /* put openEntity back in list of free instances */ + openEntity->next = freeInternalEntities; + freeInternalEntities = openEntity; + } + +#ifdef XML_DTD + if (entity->is_param) { + int tok; + processor = prologProcessor; + tok = XmlPrologTok(encoding, s, end, &next); + return doProlog(parser, encoding, s, end, tok, next, nextPtr, + (XML_Bool)!ps_finalBuffer); + } + else +#endif /* XML_DTD */ + { + processor = contentProcessor; + /* see externalEntityContentProcessor vs contentProcessor */ + return doContent(parser, parentParser ? 1 : 0, encoding, s, end, + nextPtr, (XML_Bool)!ps_finalBuffer); + } +} + +static enum XML_Error PTRCALL +errorProcessor(XML_Parser parser, + const char *s, + const char *end, + const char **nextPtr) +{ + UNUSED(s); + UNUSED(end); + UNUSED(nextPtr); + + return errorCode; +} + +static enum XML_Error +storeAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata, + const char *ptr, const char *end, + STRING_POOL *pool) +{ + enum XML_Error result = appendAttributeValue(parser, enc, isCdata, ptr, + end, pool); + if (result) + return result; + if (!isCdata && poolLength(pool) && poolLastChar(pool) == 0x20) + poolChop(pool); + if (!poolAppendChar(pool, XML_T('\0'))) + return XML_ERROR_NO_MEMORY; + return XML_ERROR_NONE; +} + +static enum XML_Error +appendAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata, + const char *ptr, const char *end, + STRING_POOL *pool) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + for (;;) { + const char *next; + int tok = XmlAttributeValueTok(enc, ptr, end, &next); + switch (tok) { + case XML_TOK_NONE: + return XML_ERROR_NONE; + case XML_TOK_INVALID: + if (enc == encoding) + eventPtr = next; + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_PARTIAL: + if (enc == encoding) + eventPtr = ptr; + return XML_ERROR_INVALID_TOKEN; + case XML_TOK_CHAR_REF: + { + XML_Char buf[XML_ENCODE_MAX]; + int i; + int n = XmlCharRefNumber(enc, ptr); + if (n < 0) { + if (enc == encoding) + eventPtr = ptr; + return XML_ERROR_BAD_CHAR_REF; + } + if (!isCdata + && n == 0x20 /* space */ + && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20)) + break; + n = XmlEncode(n, (ICHAR *)buf); + if (!n) { + if (enc == encoding) + eventPtr = ptr; + return XML_ERROR_BAD_CHAR_REF; + } + for (i = 0; i < n; i++) { + if (!poolAppendChar(pool, buf[i])) + return XML_ERROR_NO_MEMORY; + } + } + break; + case XML_TOK_DATA_CHARS: + if (!poolAppend(pool, enc, ptr, next)) + return XML_ERROR_NO_MEMORY; + break; + case XML_TOK_TRAILING_CR: + next = ptr + enc->minBytesPerChar; + /* fall through */ + case XML_TOK_ATTRIBUTE_VALUE_S: + case XML_TOK_DATA_NEWLINE: + if (!isCdata && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20)) + break; + if (!poolAppendChar(pool, 0x20)) + return XML_ERROR_NO_MEMORY; + break; + case XML_TOK_ENTITY_REF: + { + const XML_Char *name; + ENTITY *entity; + char checkEntityDecl; + XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc, + ptr + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (ch) { + if (!poolAppendChar(pool, ch)) + return XML_ERROR_NO_MEMORY; + break; + } + name = poolStoreString(&temp2Pool, enc, + ptr + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!name) + return XML_ERROR_NO_MEMORY; + entity = (ENTITY *)lookup(parser, &dtd->generalEntities, name, 0); + poolDiscard(&temp2Pool); + /* First, determine if a check for an existing declaration is needed; + if yes, check that the entity exists, and that it is internal. + */ + if (pool == &dtd->pool) /* are we called from prolog? */ + checkEntityDecl = +#ifdef XML_DTD + prologState.documentEntity && +#endif /* XML_DTD */ + (dtd->standalone + ? !openInternalEntities + : !dtd->hasParamEntityRefs); + else /* if (pool == &tempPool): we are called from content */ + checkEntityDecl = !dtd->hasParamEntityRefs || dtd->standalone; + if (checkEntityDecl) { + if (!entity) + return XML_ERROR_UNDEFINED_ENTITY; + else if (!entity->is_internal) + return XML_ERROR_ENTITY_DECLARED_IN_PE; + } + else if (!entity) { + /* Cannot report skipped entity here - see comments on + skippedEntityHandler. + if (skippedEntityHandler) + skippedEntityHandler(handlerArg, name, 0); + */ + /* Cannot call the default handler because this would be + out of sync with the call to the startElementHandler. + if ((pool == &tempPool) && defaultHandler) + reportDefault(parser, enc, ptr, next); + */ + break; + } + if (entity->open) { + if (enc == encoding) + eventPtr = ptr; + return XML_ERROR_RECURSIVE_ENTITY_REF; + } + if (entity->notation) { + if (enc == encoding) + eventPtr = ptr; + return XML_ERROR_BINARY_ENTITY_REF; + } + if (!entity->textPtr) { + if (enc == encoding) + eventPtr = ptr; + return XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF; + } + else { + enum XML_Error result; + const XML_Char *textEnd = entity->textPtr + entity->textLen; + entity->open = XML_TRUE; + result = appendAttributeValue(parser, internalEncoding, isCdata, + (char *)entity->textPtr, + (char *)textEnd, pool); + entity->open = XML_FALSE; + if (result) + return result; + } + } + break; + default: + if (enc == encoding) + eventPtr = ptr; + return XML_ERROR_UNEXPECTED_STATE; + } + ptr = next; + } + /* not reached */ +} + +static enum XML_Error +storeEntityValue(XML_Parser parser, + const ENCODING *enc, + const char *entityTextPtr, + const char *entityTextEnd) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + STRING_POOL *pool = &(dtd->entityValuePool); + enum XML_Error result = XML_ERROR_NONE; +#ifdef XML_DTD + int oldInEntityValue = prologState.inEntityValue; + prologState.inEntityValue = 1; +#endif /* XML_DTD */ + /* never return Null for the value argument in EntityDeclHandler, + since this would indicate an external entity; therefore we + have to make sure that entityValuePool.start is not null */ + if (!pool->blocks) { + if (!poolGrow(pool)) + return XML_ERROR_NO_MEMORY; + } + + for (;;) { + const char *next; + int tok = XmlEntityValueTok(enc, entityTextPtr, entityTextEnd, &next); + switch (tok) { + case XML_TOK_PARAM_ENTITY_REF: +#ifdef XML_DTD + if (isParamEntity || enc != encoding) { + const XML_Char *name; + ENTITY *entity; + name = poolStoreString(&tempPool, enc, + entityTextPtr + enc->minBytesPerChar, + next - enc->minBytesPerChar); + if (!name) { + result = XML_ERROR_NO_MEMORY; + goto endEntityValue; + } + entity = (ENTITY *)lookup(parser, &dtd->paramEntities, name, 0); + poolDiscard(&tempPool); + if (!entity) { + /* not a well-formedness error - see XML 1.0: WFC Entity Declared */ + /* cannot report skipped entity here - see comments on + skippedEntityHandler + if (skippedEntityHandler) + skippedEntityHandler(handlerArg, name, 0); + */ + dtd->keepProcessing = dtd->standalone; + goto endEntityValue; + } + if (entity->open) { + if (enc == encoding) + eventPtr = entityTextPtr; + result = XML_ERROR_RECURSIVE_ENTITY_REF; + goto endEntityValue; + } + if (entity->systemId) { + if (externalEntityRefHandler) { + dtd->paramEntityRead = XML_FALSE; + entity->open = XML_TRUE; + if (!externalEntityRefHandler(externalEntityRefHandlerArg, + 0, + entity->base, + entity->systemId, + entity->publicId)) { + entity->open = XML_FALSE; + result = XML_ERROR_EXTERNAL_ENTITY_HANDLING; + goto endEntityValue; + } + entity->open = XML_FALSE; + if (!dtd->paramEntityRead) + dtd->keepProcessing = dtd->standalone; + } + else + dtd->keepProcessing = dtd->standalone; + } + else { + entity->open = XML_TRUE; + result = storeEntityValue(parser, + internalEncoding, + (char *)entity->textPtr, + (char *)(entity->textPtr + + entity->textLen)); + entity->open = XML_FALSE; + if (result) + goto endEntityValue; + } + break; + } +#endif /* XML_DTD */ + /* In the internal subset, PE references are not legal + within markup declarations, e.g entity values in this case. */ + eventPtr = entityTextPtr; + result = XML_ERROR_PARAM_ENTITY_REF; + goto endEntityValue; + case XML_TOK_NONE: + result = XML_ERROR_NONE; + goto endEntityValue; + case XML_TOK_ENTITY_REF: + case XML_TOK_DATA_CHARS: + if (!poolAppend(pool, enc, entityTextPtr, next)) { + result = XML_ERROR_NO_MEMORY; + goto endEntityValue; + } + break; + case XML_TOK_TRAILING_CR: + next = entityTextPtr + enc->minBytesPerChar; + /* fall through */ + case XML_TOK_DATA_NEWLINE: + if (pool->end == pool->ptr && !poolGrow(pool)) { + result = XML_ERROR_NO_MEMORY; + goto endEntityValue; + } + *(pool->ptr)++ = 0xA; + break; + case XML_TOK_CHAR_REF: + { + XML_Char buf[XML_ENCODE_MAX]; + int i; + int n = XmlCharRefNumber(enc, entityTextPtr); + if (n < 0) { + if (enc == encoding) + eventPtr = entityTextPtr; + result = XML_ERROR_BAD_CHAR_REF; + goto endEntityValue; + } + n = XmlEncode(n, (ICHAR *)buf); + if (!n) { + if (enc == encoding) + eventPtr = entityTextPtr; + result = XML_ERROR_BAD_CHAR_REF; + goto endEntityValue; + } + for (i = 0; i < n; i++) { + if (pool->end == pool->ptr && !poolGrow(pool)) { + result = XML_ERROR_NO_MEMORY; + goto endEntityValue; + } + *(pool->ptr)++ = buf[i]; + } + } + break; + case XML_TOK_PARTIAL: + if (enc == encoding) + eventPtr = entityTextPtr; + result = XML_ERROR_INVALID_TOKEN; + goto endEntityValue; + case XML_TOK_INVALID: + if (enc == encoding) + eventPtr = next; + result = XML_ERROR_INVALID_TOKEN; + goto endEntityValue; + default: + if (enc == encoding) + eventPtr = entityTextPtr; + result = XML_ERROR_UNEXPECTED_STATE; + goto endEntityValue; + } + entityTextPtr = next; + } +endEntityValue: +#ifdef XML_DTD + prologState.inEntityValue = oldInEntityValue; +#endif /* XML_DTD */ + return result; +} + +static void FASTCALL +normalizeLines(XML_Char *s) +{ + XML_Char *p; + for (;; s++) { + if (*s == XML_T('\0')) + return; + if (*s == 0xD) + break; + } + p = s; + do { + if (*s == 0xD) { + *p++ = 0xA; + if (*++s == 0xA) + s++; + } + else + *p++ = *s++; + } while (*s); + *p = XML_T('\0'); +} + +static int +reportProcessingInstruction(XML_Parser parser, const ENCODING *enc, + const char *start, const char *end) +{ + const XML_Char *target; + XML_Char *data; + const char *tem; + if (!processingInstructionHandler) { + if (defaultHandler) + reportDefault(parser, enc, start, end); + return 1; + } + start += enc->minBytesPerChar * 2; + tem = start + XmlNameLength(enc, start); + target = poolStoreString(&tempPool, enc, start, tem); + if (!target) + return 0; + poolFinish(&tempPool); + data = poolStoreString(&tempPool, enc, + XmlSkipS(enc, tem), + end - enc->minBytesPerChar*2); + if (!data) + return 0; + normalizeLines(data); + processingInstructionHandler(handlerArg, target, data); + poolClear(&tempPool); + return 1; +} + +static int +reportComment(XML_Parser parser, const ENCODING *enc, + const char *start, const char *end) +{ + XML_Char *data; + if (!commentHandler) { + if (defaultHandler) + reportDefault(parser, enc, start, end); + return 1; + } + data = poolStoreString(&tempPool, + enc, + start + enc->minBytesPerChar * 4, + end - enc->minBytesPerChar * 3); + if (!data) + return 0; + normalizeLines(data); + commentHandler(handlerArg, data); + poolClear(&tempPool); + return 1; +} + +static void +reportDefault(XML_Parser parser, const ENCODING *enc, + const char *s, const char *end) +{ + if (MUST_CONVERT(enc, s)) { + const char **eventPP; + const char **eventEndPP; + if (enc == encoding) { + eventPP = &eventPtr; + eventEndPP = &eventEndPtr; + } + else { + eventPP = &(openInternalEntities->internalEventPtr); + eventEndPP = &(openInternalEntities->internalEventEndPtr); + } + do { + ICHAR *dataPtr = (ICHAR *)dataBuf; + XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd); + *eventEndPP = s; + defaultHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf)); + *eventPP = s; + } while (s != end); + } + else + defaultHandler(handlerArg, (XML_Char *)s, (int)((XML_Char *)end - (XML_Char *)s)); +} + + +static int +defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *attId, XML_Bool isCdata, + XML_Bool isId, const XML_Char *value, XML_Parser parser) +{ + DEFAULT_ATTRIBUTE *att; + if (value || isId) { + /* The handling of default attributes gets messed up if we have + a default which duplicates a non-default. */ + int i; + for (i = 0; i < type->nDefaultAtts; i++) + if (attId == type->defaultAtts[i].id) + return 1; + if (isId && !type->idAtt && !attId->xmlns) + type->idAtt = attId; + } + if (type->nDefaultAtts == type->allocDefaultAtts) { + if (type->allocDefaultAtts == 0) { + type->allocDefaultAtts = 8; + type->defaultAtts = (DEFAULT_ATTRIBUTE *)MALLOC(type->allocDefaultAtts + * sizeof(DEFAULT_ATTRIBUTE)); + if (!type->defaultAtts) + return 0; + } + else { + DEFAULT_ATTRIBUTE *temp; + int count = type->allocDefaultAtts * 2; + temp = (DEFAULT_ATTRIBUTE *) + REALLOC(type->defaultAtts, (count * sizeof(DEFAULT_ATTRIBUTE))); + if (temp == NULL) + return 0; + type->allocDefaultAtts = count; + type->defaultAtts = temp; + } + } + att = type->defaultAtts + type->nDefaultAtts; + att->id = attId; + att->value = value; + att->isCdata = isCdata; + if (!isCdata) + attId->maybeTokenized = XML_TRUE; + type->nDefaultAtts += 1; + return 1; +} + +static int +setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *elementType) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + const XML_Char *name; + for (name = elementType->name; *name; name++) { + if (*name == XML_T(ASCII_COLON)) { + PREFIX *prefix; + const XML_Char *s; + for (s = elementType->name; s != name; s++) { + if (!poolAppendChar(&dtd->pool, *s)) + return 0; + } + if (!poolAppendChar(&dtd->pool, XML_T('\0'))) + return 0; + prefix = (PREFIX *)lookup(parser, &dtd->prefixes, poolStart(&dtd->pool), + sizeof(PREFIX)); + if (!prefix) + return 0; + if (prefix->name == poolStart(&dtd->pool)) + poolFinish(&dtd->pool); + else + poolDiscard(&dtd->pool); + elementType->prefix = prefix; + + } + } + return 1; +} + +static ATTRIBUTE_ID * +getAttributeId(XML_Parser parser, const ENCODING *enc, + const char *start, const char *end) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + ATTRIBUTE_ID *id; + const XML_Char *name; + if (!poolAppendChar(&dtd->pool, XML_T('\0'))) + return NULL; + name = poolStoreString(&dtd->pool, enc, start, end); + if (!name) + return NULL; + /* skip quotation mark - its storage will be re-used (like in name[-1]) */ + ++name; + id = (ATTRIBUTE_ID *)lookup(parser, &dtd->attributeIds, name, sizeof(ATTRIBUTE_ID)); + if (!id) + return NULL; + if (id->name != name) + poolDiscard(&dtd->pool); + else { + poolFinish(&dtd->pool); + if (!ns) + ; + else if (name[0] == XML_T(ASCII_x) + && name[1] == XML_T(ASCII_m) + && name[2] == XML_T(ASCII_l) + && name[3] == XML_T(ASCII_n) + && name[4] == XML_T(ASCII_s) + && (name[5] == XML_T('\0') || name[5] == XML_T(ASCII_COLON))) { + if (name[5] == XML_T('\0')) + id->prefix = &dtd->defaultPrefix; + else + id->prefix = (PREFIX *)lookup(parser, &dtd->prefixes, name + 6, sizeof(PREFIX)); + id->xmlns = XML_TRUE; + } + else { + int i; + for (i = 0; name[i]; i++) { + /* attributes without prefix are *not* in the default namespace */ + if (name[i] == XML_T(ASCII_COLON)) { + int j; + for (j = 0; j < i; j++) { + if (!poolAppendChar(&dtd->pool, name[j])) + return NULL; + } + if (!poolAppendChar(&dtd->pool, XML_T('\0'))) + return NULL; + id->prefix = (PREFIX *)lookup(parser, &dtd->prefixes, poolStart(&dtd->pool), + sizeof(PREFIX)); + if (id->prefix->name == poolStart(&dtd->pool)) + poolFinish(&dtd->pool); + else + poolDiscard(&dtd->pool); + break; + } + } + } + } + return id; +} + +#define CONTEXT_SEP XML_T(ASCII_FF) + +static const XML_Char * +getContext(XML_Parser parser) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + HASH_TABLE_ITER iter; + XML_Bool needSep = XML_FALSE; + + if (dtd->defaultPrefix.binding) { + int i; + int len; + if (!poolAppendChar(&tempPool, XML_T(ASCII_EQUALS))) + return NULL; + len = dtd->defaultPrefix.binding->uriLen; + if (namespaceSeparator) + len--; + for (i = 0; i < len; i++) + if (!poolAppendChar(&tempPool, dtd->defaultPrefix.binding->uri[i])) + return NULL; + needSep = XML_TRUE; + } + + hashTableIterInit(&iter, &(dtd->prefixes)); + for (;;) { + int i; + int len; + const XML_Char *s; + PREFIX *prefix = (PREFIX *)hashTableIterNext(&iter); + if (!prefix) + break; + if (!prefix->binding) + continue; + if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP)) + return NULL; + for (s = prefix->name; *s; s++) + if (!poolAppendChar(&tempPool, *s)) + return NULL; + if (!poolAppendChar(&tempPool, XML_T(ASCII_EQUALS))) + return NULL; + len = prefix->binding->uriLen; + if (namespaceSeparator) + len--; + for (i = 0; i < len; i++) + if (!poolAppendChar(&tempPool, prefix->binding->uri[i])) + return NULL; + needSep = XML_TRUE; + } + + + hashTableIterInit(&iter, &(dtd->generalEntities)); + for (;;) { + const XML_Char *s; + ENTITY *e = (ENTITY *)hashTableIterNext(&iter); + if (!e) + break; + if (!e->open) + continue; + if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP)) + return NULL; + for (s = e->name; *s; s++) + if (!poolAppendChar(&tempPool, *s)) + return 0; + needSep = XML_TRUE; + } + + if (!poolAppendChar(&tempPool, XML_T('\0'))) + return NULL; + return tempPool.start; +} + +static XML_Bool +setContext(XML_Parser parser, const XML_Char *context) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + const XML_Char *s = context; + + while (*context != XML_T('\0')) { + if (*s == CONTEXT_SEP || *s == XML_T('\0')) { + ENTITY *e; + if (!poolAppendChar(&tempPool, XML_T('\0'))) + return XML_FALSE; + e = (ENTITY *)lookup(parser, &dtd->generalEntities, poolStart(&tempPool), 0); + if (e) + e->open = XML_TRUE; + if (*s != XML_T('\0')) + s++; + context = s; + poolDiscard(&tempPool); + } + else if (*s == XML_T(ASCII_EQUALS)) { + PREFIX *prefix; + if (poolLength(&tempPool) == 0) + prefix = &dtd->defaultPrefix; + else { + if (!poolAppendChar(&tempPool, XML_T('\0'))) + return XML_FALSE; + prefix = (PREFIX *)lookup(parser, &dtd->prefixes, poolStart(&tempPool), + sizeof(PREFIX)); + if (!prefix) + return XML_FALSE; + if (prefix->name == poolStart(&tempPool)) { + prefix->name = poolCopyString(&dtd->pool, prefix->name); + if (!prefix->name) + return XML_FALSE; + } + poolDiscard(&tempPool); + } + for (context = s + 1; + *context != CONTEXT_SEP && *context != XML_T('\0'); + context++) + if (!poolAppendChar(&tempPool, *context)) + return XML_FALSE; + if (!poolAppendChar(&tempPool, XML_T('\0'))) + return XML_FALSE; + if (addBinding(parser, prefix, NULL, poolStart(&tempPool), + &inheritedBindings) != XML_ERROR_NONE) + return XML_FALSE; + poolDiscard(&tempPool); + if (*context != XML_T('\0')) + ++context; + s = context; + } + else { + if (!poolAppendChar(&tempPool, *s)) + return XML_FALSE; + s++; + } + } + return XML_TRUE; +} + +static void FASTCALL +normalizePublicId(XML_Char *publicId) +{ + XML_Char *p = publicId; + XML_Char *s; + for (s = publicId; *s; s++) { + switch (*s) { + case 0x20: + case 0xD: + case 0xA: + if (p != publicId && p[-1] != 0x20) + *p++ = 0x20; + break; + default: + *p++ = *s; + } + } + if (p != publicId && p[-1] == 0x20) + --p; + *p = XML_T('\0'); +} + +static DTD * +dtdCreate(const XML_Memory_Handling_Suite *ms) +{ + DTD *p = (DTD *)ms->malloc_fcn(sizeof(DTD)); + if (p == NULL) + return p; + poolInit(&(p->pool), ms); + poolInit(&(p->entityValuePool), ms); + hashTableInit(&(p->generalEntities), ms); + hashTableInit(&(p->elementTypes), ms); + hashTableInit(&(p->attributeIds), ms); + hashTableInit(&(p->prefixes), ms); +#ifdef XML_DTD + p->paramEntityRead = XML_FALSE; + hashTableInit(&(p->paramEntities), ms); +#endif /* XML_DTD */ + p->defaultPrefix.name = NULL; + p->defaultPrefix.binding = NULL; + + p->in_eldecl = XML_FALSE; + p->scaffIndex = NULL; + p->scaffold = NULL; + p->scaffLevel = 0; + p->scaffSize = 0; + p->scaffCount = 0; + p->contentStringLen = 0; + + p->keepProcessing = XML_TRUE; + p->hasParamEntityRefs = XML_FALSE; + p->standalone = XML_FALSE; + return p; +} + +static void +dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms) +{ + HASH_TABLE_ITER iter; + hashTableIterInit(&iter, &(p->elementTypes)); + for (;;) { + ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter); + if (!e) + break; + if (e->allocDefaultAtts != 0) + ms->free_fcn(e->defaultAtts); + } + hashTableClear(&(p->generalEntities)); +#ifdef XML_DTD + p->paramEntityRead = XML_FALSE; + hashTableClear(&(p->paramEntities)); +#endif /* XML_DTD */ + hashTableClear(&(p->elementTypes)); + hashTableClear(&(p->attributeIds)); + hashTableClear(&(p->prefixes)); + poolClear(&(p->pool)); + poolClear(&(p->entityValuePool)); + p->defaultPrefix.name = NULL; + p->defaultPrefix.binding = NULL; + + p->in_eldecl = XML_FALSE; + + ms->free_fcn(p->scaffIndex); + p->scaffIndex = NULL; + ms->free_fcn(p->scaffold); + p->scaffold = NULL; + + p->scaffLevel = 0; + p->scaffSize = 0; + p->scaffCount = 0; + p->contentStringLen = 0; + + p->keepProcessing = XML_TRUE; + p->hasParamEntityRefs = XML_FALSE; + p->standalone = XML_FALSE; +} + +static void +dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms) +{ + HASH_TABLE_ITER iter; + hashTableIterInit(&iter, &(p->elementTypes)); + for (;;) { + ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter); + if (!e) + break; + if (e->allocDefaultAtts != 0) + ms->free_fcn(e->defaultAtts); + } + hashTableDestroy(&(p->generalEntities)); +#ifdef XML_DTD + hashTableDestroy(&(p->paramEntities)); +#endif /* XML_DTD */ + hashTableDestroy(&(p->elementTypes)); + hashTableDestroy(&(p->attributeIds)); + hashTableDestroy(&(p->prefixes)); + poolDestroy(&(p->pool)); + poolDestroy(&(p->entityValuePool)); + if (isDocEntity) { + ms->free_fcn(p->scaffIndex); + ms->free_fcn(p->scaffold); + } + ms->free_fcn(p); +} + +/* Do a deep copy of the DTD. Return 0 for out of memory, non-zero otherwise. + The new DTD has already been initialized. +*/ +static int +dtdCopy(XML_Parser oldParser, DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms) +{ + HASH_TABLE_ITER iter; + + /* Copy the prefix table. */ + + hashTableIterInit(&iter, &(oldDtd->prefixes)); + for (;;) { + const XML_Char *name; + const PREFIX *oldP = (PREFIX *)hashTableIterNext(&iter); + if (!oldP) + break; + name = poolCopyString(&(newDtd->pool), oldP->name); + if (!name) + return 0; + if (!lookup(oldParser, &(newDtd->prefixes), name, sizeof(PREFIX))) + return 0; + } + + hashTableIterInit(&iter, &(oldDtd->attributeIds)); + + /* Copy the attribute id table. */ + + for (;;) { + ATTRIBUTE_ID *newA; + const XML_Char *name; + const ATTRIBUTE_ID *oldA = (ATTRIBUTE_ID *)hashTableIterNext(&iter); + + if (!oldA) + break; + /* Remember to allocate the scratch byte before the name. */ + if (!poolAppendChar(&(newDtd->pool), XML_T('\0'))) + return 0; + name = poolCopyString(&(newDtd->pool), oldA->name); + if (!name) + return 0; + ++name; + newA = (ATTRIBUTE_ID *)lookup(oldParser, &(newDtd->attributeIds), name, + sizeof(ATTRIBUTE_ID)); + if (!newA) + return 0; + newA->maybeTokenized = oldA->maybeTokenized; + if (oldA->prefix) { + newA->xmlns = oldA->xmlns; + if (oldA->prefix == &oldDtd->defaultPrefix) + newA->prefix = &newDtd->defaultPrefix; + else + newA->prefix = (PREFIX *)lookup(oldParser, &(newDtd->prefixes), + oldA->prefix->name, 0); + } + } + + /* Copy the element type table. */ + + hashTableIterInit(&iter, &(oldDtd->elementTypes)); + + for (;;) { + int i; + ELEMENT_TYPE *newE; + const XML_Char *name; + const ELEMENT_TYPE *oldE = (ELEMENT_TYPE *)hashTableIterNext(&iter); + if (!oldE) + break; + name = poolCopyString(&(newDtd->pool), oldE->name); + if (!name) + return 0; + newE = (ELEMENT_TYPE *)lookup(oldParser, &(newDtd->elementTypes), name, + sizeof(ELEMENT_TYPE)); + if (!newE) + return 0; + if (oldE->nDefaultAtts) { + newE->defaultAtts = (DEFAULT_ATTRIBUTE *) + ms->malloc_fcn(oldE->nDefaultAtts * sizeof(DEFAULT_ATTRIBUTE)); + if (!newE->defaultAtts) { + ms->free_fcn(newE); + return 0; + } + } + if (oldE->idAtt) + newE->idAtt = (ATTRIBUTE_ID *) + lookup(oldParser, &(newDtd->attributeIds), oldE->idAtt->name, 0); + newE->allocDefaultAtts = newE->nDefaultAtts = oldE->nDefaultAtts; + if (oldE->prefix) + newE->prefix = (PREFIX *)lookup(oldParser, &(newDtd->prefixes), + oldE->prefix->name, 0); + for (i = 0; i < newE->nDefaultAtts; i++) { + newE->defaultAtts[i].id = (ATTRIBUTE_ID *) + lookup(oldParser, &(newDtd->attributeIds), oldE->defaultAtts[i].id->name, 0); + newE->defaultAtts[i].isCdata = oldE->defaultAtts[i].isCdata; + if (oldE->defaultAtts[i].value) { + newE->defaultAtts[i].value + = poolCopyString(&(newDtd->pool), oldE->defaultAtts[i].value); + if (!newE->defaultAtts[i].value) + return 0; + } + else + newE->defaultAtts[i].value = NULL; + } + } + + /* Copy the entity tables. */ + if (!copyEntityTable(oldParser, + &(newDtd->generalEntities), + &(newDtd->pool), + &(oldDtd->generalEntities))) + return 0; + +#ifdef XML_DTD + if (!copyEntityTable(oldParser, + &(newDtd->paramEntities), + &(newDtd->pool), + &(oldDtd->paramEntities))) + return 0; + newDtd->paramEntityRead = oldDtd->paramEntityRead; +#endif /* XML_DTD */ + + newDtd->keepProcessing = oldDtd->keepProcessing; + newDtd->hasParamEntityRefs = oldDtd->hasParamEntityRefs; + newDtd->standalone = oldDtd->standalone; + + /* Don't want deep copying for scaffolding */ + newDtd->in_eldecl = oldDtd->in_eldecl; + newDtd->scaffold = oldDtd->scaffold; + newDtd->contentStringLen = oldDtd->contentStringLen; + newDtd->scaffSize = oldDtd->scaffSize; + newDtd->scaffLevel = oldDtd->scaffLevel; + newDtd->scaffIndex = oldDtd->scaffIndex; + + return 1; +} /* End dtdCopy */ + +static int +copyEntityTable(XML_Parser oldParser, + HASH_TABLE *newTable, + STRING_POOL *newPool, + const HASH_TABLE *oldTable) +{ + HASH_TABLE_ITER iter; + const XML_Char *cachedOldBase = NULL; + const XML_Char *cachedNewBase = NULL; + + hashTableIterInit(&iter, oldTable); + + for (;;) { + ENTITY *newE; + const XML_Char *name; + const ENTITY *oldE = (ENTITY *)hashTableIterNext(&iter); + if (!oldE) + break; + name = poolCopyString(newPool, oldE->name); + if (!name) + return 0; + newE = (ENTITY *)lookup(oldParser, newTable, name, sizeof(ENTITY)); + if (!newE) + return 0; + if (oldE->systemId) { + const XML_Char *tem = poolCopyString(newPool, oldE->systemId); + if (!tem) + return 0; + newE->systemId = tem; + if (oldE->base) { + if (oldE->base == cachedOldBase) + newE->base = cachedNewBase; + else { + cachedOldBase = oldE->base; + tem = poolCopyString(newPool, cachedOldBase); + if (!tem) + return 0; + cachedNewBase = newE->base = tem; + } + } + if (oldE->publicId) { + tem = poolCopyString(newPool, oldE->publicId); + if (!tem) + return 0; + newE->publicId = tem; + } + } + else { + const XML_Char *tem = poolCopyStringN(newPool, oldE->textPtr, + oldE->textLen); + if (!tem) + return 0; + newE->textPtr = tem; + newE->textLen = oldE->textLen; + } + if (oldE->notation) { + const XML_Char *tem = poolCopyString(newPool, oldE->notation); + if (!tem) + return 0; + newE->notation = tem; + } + newE->is_param = oldE->is_param; + newE->is_internal = oldE->is_internal; + } + return 1; +} + +#define INIT_POWER 6 + +static XML_Bool FASTCALL +keyeq(KEY s1, KEY s2) +{ + for (; *s1 == *s2; s1++, s2++) + if (*s1 == 0) + return XML_TRUE; + return XML_FALSE; +} + +static unsigned long FASTCALL +hash(XML_Parser parser, KEY s) +{ + unsigned long h = hash_secret_salt; + while (*s) + h = CHAR_HASH(h, *s++); + return h; +} + +static NAMED * +lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize) +{ + size_t i; + if (table->size == 0) { + size_t tsize; + if (!createSize) + return NULL; + table->power = INIT_POWER; + /* table->size is a power of 2 */ + table->size = (size_t)1 << INIT_POWER; + tsize = table->size * sizeof(NAMED *); + table->v = (NAMED **)table->mem->malloc_fcn(tsize); + if (!table->v) { + table->size = 0; + return NULL; + } + memset(table->v, 0, tsize); + i = hash(parser, name) & ((unsigned long)table->size - 1); + } + else { + unsigned long h = hash(parser, name); + unsigned long mask = (unsigned long)table->size - 1; + unsigned char step = 0; + i = h & mask; + while (table->v[i]) { + if (keyeq(name, table->v[i]->name)) + return table->v[i]; + if (!step) + step = PROBE_STEP(h, mask, table->power); + i < step ? (i += table->size - step) : (i -= step); + } + if (!createSize) + return NULL; + + /* check for overflow (table is half full) */ + if (table->used >> (table->power - 1)) { + unsigned char newPower = table->power + 1; + size_t newSize = (size_t)1 << newPower; + unsigned long newMask = (unsigned long)newSize - 1; + size_t tsize = newSize * sizeof(NAMED *); + NAMED **newV = (NAMED **)table->mem->malloc_fcn(tsize); + if (!newV) + return NULL; + memset(newV, 0, tsize); + for (i = 0; i < table->size; i++) + if (table->v[i]) { + unsigned long newHash = hash(parser, table->v[i]->name); + size_t j = newHash & newMask; + step = 0; + while (newV[j]) { + if (!step) + step = PROBE_STEP(newHash, newMask, newPower); + j < step ? (j += newSize - step) : (j -= step); + } + newV[j] = table->v[i]; + } + table->mem->free_fcn(table->v); + table->v = newV; + table->power = newPower; + table->size = newSize; + i = h & newMask; + step = 0; + while (table->v[i]) { + if (!step) + step = PROBE_STEP(h, newMask, newPower); + i < step ? (i += newSize - step) : (i -= step); + } + } + } + table->v[i] = (NAMED *)table->mem->malloc_fcn(createSize); + if (!table->v[i]) + return NULL; + memset(table->v[i], 0, createSize); + table->v[i]->name = name; + (table->used)++; + return table->v[i]; +} + +static void FASTCALL +hashTableClear(HASH_TABLE *table) +{ + size_t i; + for (i = 0; i < table->size; i++) { + table->mem->free_fcn(table->v[i]); + table->v[i] = NULL; + } + table->used = 0; +} + +static void FASTCALL +hashTableDestroy(HASH_TABLE *table) +{ + size_t i; + for (i = 0; i < table->size; i++) + table->mem->free_fcn(table->v[i]); + table->mem->free_fcn(table->v); +} + +static void FASTCALL +hashTableInit(HASH_TABLE *p, const XML_Memory_Handling_Suite *ms) +{ + p->power = 0; + p->size = 0; + p->used = 0; + p->v = NULL; + p->mem = ms; +} + +static void FASTCALL +hashTableIterInit(HASH_TABLE_ITER *iter, const HASH_TABLE *table) +{ + iter->p = table->v; + iter->end = iter->p + table->size; +} + +static NAMED * FASTCALL +hashTableIterNext(HASH_TABLE_ITER *iter) +{ + while (iter->p != iter->end) { + NAMED *tem = *(iter->p)++; + if (tem) + return tem; + } + return NULL; +} + +static void FASTCALL +poolInit(STRING_POOL *pool, const XML_Memory_Handling_Suite *ms) +{ + pool->blocks = NULL; + pool->freeBlocks = NULL; + pool->start = NULL; + pool->ptr = NULL; + pool->end = NULL; + pool->mem = ms; +} + +static void FASTCALL +poolClear(STRING_POOL *pool) +{ + if (!pool->freeBlocks) + pool->freeBlocks = pool->blocks; + else { + BLOCK *p = pool->blocks; + while (p) { + BLOCK *tem = p->next; + p->next = pool->freeBlocks; + pool->freeBlocks = p; + p = tem; + } + } + pool->blocks = NULL; + pool->start = NULL; + pool->ptr = NULL; + pool->end = NULL; +} + +static void FASTCALL +poolDestroy(STRING_POOL *pool) +{ + BLOCK *p = pool->blocks; + while (p) { + BLOCK *tem = p->next; + pool->mem->free_fcn(p); + p = tem; + } + p = pool->freeBlocks; + while (p) { + BLOCK *tem = p->next; + pool->mem->free_fcn(p); + p = tem; + } +} + +static XML_Char * +poolAppend(STRING_POOL *pool, const ENCODING *enc, + const char *ptr, const char *end) +{ + if (!pool->ptr && !poolGrow(pool)) + return NULL; + for (;;) { + XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end); + if (ptr == end) + break; + if (!poolGrow(pool)) + return NULL; + } + return pool->start; +} + +static const XML_Char * FASTCALL +poolCopyString(STRING_POOL *pool, const XML_Char *s) +{ + do { + if (!poolAppendChar(pool, *s)) + return NULL; + } while (*s++); + s = pool->start; + poolFinish(pool); + return s; +} + +static const XML_Char * +poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n) +{ + if (!pool->ptr && !poolGrow(pool)) + return NULL; + for (; n > 0; --n, s++) { + if (!poolAppendChar(pool, *s)) + return NULL; + } + s = pool->start; + poolFinish(pool); + return s; +} + +static const XML_Char * FASTCALL +poolAppendString(STRING_POOL *pool, const XML_Char *s) +{ + while (*s) { + if (!poolAppendChar(pool, *s)) + return NULL; + s++; + } + return pool->start; +} + +static XML_Char * +poolStoreString(STRING_POOL *pool, const ENCODING *enc, + const char *ptr, const char *end) +{ + if (!poolAppend(pool, enc, ptr, end)) + return NULL; + if (pool->ptr == pool->end && !poolGrow(pool)) + return NULL; + *(pool->ptr)++ = 0; + return pool->start; +} + +static XML_Bool FASTCALL +poolGrow(STRING_POOL *pool) +{ + if (pool->freeBlocks) { + if (pool->start == 0) { + pool->blocks = pool->freeBlocks; + pool->freeBlocks = pool->freeBlocks->next; + pool->blocks->next = NULL; + pool->start = pool->blocks->s; + pool->end = pool->start + pool->blocks->size; + pool->ptr = pool->start; + return XML_TRUE; + } + if (pool->end - pool->start < pool->freeBlocks->size) { + BLOCK *tem = pool->freeBlocks->next; + pool->freeBlocks->next = pool->blocks; + pool->blocks = pool->freeBlocks; + pool->freeBlocks = tem; + memcpy(pool->blocks->s, pool->start, + (pool->end - pool->start) * sizeof(XML_Char)); + pool->ptr = pool->blocks->s + (pool->ptr - pool->start); + pool->start = pool->blocks->s; + pool->end = pool->start + pool->blocks->size; + return XML_TRUE; + } + } + if (pool->blocks && pool->start == pool->blocks->s) { + int blockSize = (int)(pool->end - pool->start)*2; + BLOCK *temp = (BLOCK *) + pool->mem->realloc_fcn(pool->blocks, + (offsetof(BLOCK, s) + + blockSize * sizeof(XML_Char))); + if (temp == NULL) + return XML_FALSE; + pool->blocks = temp; + pool->blocks->size = blockSize; + pool->ptr = pool->blocks->s + (pool->ptr - pool->start); + pool->start = pool->blocks->s; + pool->end = pool->start + blockSize; + } + else { + BLOCK *tem; + int blockSize = (int)(pool->end - pool->start); + if (blockSize < INIT_BLOCK_SIZE) + blockSize = INIT_BLOCK_SIZE; + else + blockSize *= 2; + tem = (BLOCK *)pool->mem->malloc_fcn(offsetof(BLOCK, s) + + blockSize * sizeof(XML_Char)); + if (!tem) + return XML_FALSE; + tem->size = blockSize; + tem->next = pool->blocks; + pool->blocks = tem; + if (pool->ptr != pool->start) + memcpy(tem->s, pool->start, + (pool->ptr - pool->start) * sizeof(XML_Char)); + pool->ptr = tem->s + (pool->ptr - pool->start); + pool->start = tem->s; + pool->end = tem->s + blockSize; + } + return XML_TRUE; +} + +static int FASTCALL +nextScaffoldPart(XML_Parser parser) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + CONTENT_SCAFFOLD * me; + int next; + + if (!dtd->scaffIndex) { + dtd->scaffIndex = (int *)MALLOC(groupSize * sizeof(int)); + if (!dtd->scaffIndex) + return -1; + dtd->scaffIndex[0] = 0; + } + + if (dtd->scaffCount >= dtd->scaffSize) { + CONTENT_SCAFFOLD *temp; + if (dtd->scaffold) { + temp = (CONTENT_SCAFFOLD *) + REALLOC(dtd->scaffold, dtd->scaffSize * 2 * sizeof(CONTENT_SCAFFOLD)); + if (temp == NULL) + return -1; + dtd->scaffSize *= 2; + } + else { + temp = (CONTENT_SCAFFOLD *)MALLOC(INIT_SCAFFOLD_ELEMENTS + * sizeof(CONTENT_SCAFFOLD)); + if (temp == NULL) + return -1; + dtd->scaffSize = INIT_SCAFFOLD_ELEMENTS; + } + dtd->scaffold = temp; + } + next = dtd->scaffCount++; + me = &dtd->scaffold[next]; + if (dtd->scaffLevel) { + CONTENT_SCAFFOLD *parent = &dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel-1]]; + if (parent->lastchild) { + dtd->scaffold[parent->lastchild].nextsib = next; + } + if (!parent->childcnt) + parent->firstchild = next; + parent->lastchild = next; + parent->childcnt++; + } + me->firstchild = me->lastchild = me->childcnt = me->nextsib = 0; + return next; +} + +static void +build_node(XML_Parser parser, + int src_node, + XML_Content *dest, + XML_Content **contpos, + XML_Char **strpos) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + dest->type = dtd->scaffold[src_node].type; + dest->quant = dtd->scaffold[src_node].quant; + if (dest->type == XML_CTYPE_NAME) { + const XML_Char *src; + dest->name = *strpos; + src = dtd->scaffold[src_node].name; + for (;;) { + *(*strpos)++ = *src; + if (!*src) + break; + src++; + } + dest->numchildren = 0; + dest->children = NULL; + } + else { + unsigned int i; + int cn; + dest->numchildren = dtd->scaffold[src_node].childcnt; + dest->children = *contpos; + *contpos += dest->numchildren; + for (i = 0, cn = dtd->scaffold[src_node].firstchild; + i < dest->numchildren; + i++, cn = dtd->scaffold[cn].nextsib) { + build_node(parser, cn, &(dest->children[i]), contpos, strpos); + } + dest->name = NULL; + } +} + +static XML_Content * +build_model (XML_Parser parser) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + XML_Content *ret; + XML_Content *cpos; + XML_Char * str; + int allocsize = (dtd->scaffCount * sizeof(XML_Content) + + (dtd->contentStringLen * sizeof(XML_Char))); + + ret = (XML_Content *)MALLOC(allocsize); + if (!ret) + return NULL; + + str = (XML_Char *) (&ret[dtd->scaffCount]); + cpos = &ret[1]; + + build_node(parser, 0, ret, &cpos, &str); + return ret; +} + +static ELEMENT_TYPE * +getElementType(XML_Parser parser, + const ENCODING *enc, + const char *ptr, + const char *end) +{ + DTD * const dtd = _dtd; /* save one level of indirection */ + const XML_Char *name = poolStoreString(&dtd->pool, enc, ptr, end); + ELEMENT_TYPE *ret; + + if (!name) + return NULL; + ret = (ELEMENT_TYPE *) lookup(parser, &dtd->elementTypes, name, sizeof(ELEMENT_TYPE)); + if (!ret) + return NULL; + if (ret->name != name) + poolDiscard(&dtd->pool); + else { + poolFinish(&dtd->pool); + if (!setElementTypePrefix(parser, ret)) + return NULL; + } + return ret; +} diff --git a/xml/details/expat/xmlrole.c b/xml/details/expat/xmlrole.c new file mode 100644 index 0000000..c36db3a --- /dev/null +++ b/xml/details/expat/xmlrole.c @@ -0,0 +1,1470 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#include <stddef.h> + +#include <xml/details/expat/config.h> + +#include <xml/details/expat/expat_external.h> +#include <xml/details/expat/internal.h> +#include <xml/details/expat/xmlrole.h> +#include <xml/details/expat/ascii.h> + +/* Doesn't check: + + that ,| are not mixed in a model group + content of literals + +*/ + +static const char KW_ANY[] = { + ASCII_A, ASCII_N, ASCII_Y, '\0' }; +static const char KW_ATTLIST[] = { + ASCII_A, ASCII_T, ASCII_T, ASCII_L, ASCII_I, ASCII_S, ASCII_T, '\0' }; +static const char KW_CDATA[] = { + ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' }; +static const char KW_DOCTYPE[] = { + ASCII_D, ASCII_O, ASCII_C, ASCII_T, ASCII_Y, ASCII_P, ASCII_E, '\0' }; +static const char KW_ELEMENT[] = { + ASCII_E, ASCII_L, ASCII_E, ASCII_M, ASCII_E, ASCII_N, ASCII_T, '\0' }; +static const char KW_EMPTY[] = { + ASCII_E, ASCII_M, ASCII_P, ASCII_T, ASCII_Y, '\0' }; +static const char KW_ENTITIES[] = { + ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_I, ASCII_E, ASCII_S, + '\0' }; +static const char KW_ENTITY[] = { + ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_Y, '\0' }; +static const char KW_FIXED[] = { + ASCII_F, ASCII_I, ASCII_X, ASCII_E, ASCII_D, '\0' }; +static const char KW_ID[] = { + ASCII_I, ASCII_D, '\0' }; +static const char KW_IDREF[] = { + ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, '\0' }; +static const char KW_IDREFS[] = { + ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, ASCII_S, '\0' }; +#ifdef XML_DTD +static const char KW_IGNORE[] = { + ASCII_I, ASCII_G, ASCII_N, ASCII_O, ASCII_R, ASCII_E, '\0' }; +#endif +static const char KW_IMPLIED[] = { + ASCII_I, ASCII_M, ASCII_P, ASCII_L, ASCII_I, ASCII_E, ASCII_D, '\0' }; +#ifdef XML_DTD +static const char KW_INCLUDE[] = { + ASCII_I, ASCII_N, ASCII_C, ASCII_L, ASCII_U, ASCII_D, ASCII_E, '\0' }; +#endif +static const char KW_NDATA[] = { + ASCII_N, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' }; +static const char KW_NMTOKEN[] = { + ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, '\0' }; +static const char KW_NMTOKENS[] = { + ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, ASCII_S, + '\0' }; +static const char KW_NOTATION[] = + { ASCII_N, ASCII_O, ASCII_T, ASCII_A, ASCII_T, ASCII_I, ASCII_O, ASCII_N, + '\0' }; +static const char KW_PCDATA[] = { + ASCII_P, ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' }; +static const char KW_PUBLIC[] = { + ASCII_P, ASCII_U, ASCII_B, ASCII_L, ASCII_I, ASCII_C, '\0' }; +static const char KW_REQUIRED[] = { + ASCII_R, ASCII_E, ASCII_Q, ASCII_U, ASCII_I, ASCII_R, ASCII_E, ASCII_D, + '\0' }; +static const char KW_SYSTEM[] = { + ASCII_S, ASCII_Y, ASCII_S, ASCII_T, ASCII_E, ASCII_M, '\0' }; + +#ifndef MIN_BYTES_PER_CHAR +#define MIN_BYTES_PER_CHAR(enc) ((enc)->minBytesPerChar) +#endif + +#ifdef XML_DTD +#define setTopLevel(state) \ + ((state)->handler = ((state)->documentEntity \ + ? internalSubset \ + : externalSubset1)) +#else /* not XML_DTD */ +#define setTopLevel(state) ((state)->handler = internalSubset) +#endif /* not XML_DTD */ + +typedef int PTRCALL PROLOG_HANDLER(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc); + +static PROLOG_HANDLER + prolog0, prolog1, prolog2, + doctype0, doctype1, doctype2, doctype3, doctype4, doctype5, + internalSubset, + entity0, entity1, entity2, entity3, entity4, entity5, entity6, + entity7, entity8, entity9, entity10, + notation0, notation1, notation2, notation3, notation4, + attlist0, attlist1, attlist2, attlist3, attlist4, attlist5, attlist6, + attlist7, attlist8, attlist9, + element0, element1, element2, element3, element4, element5, element6, + element7, +#ifdef XML_DTD + externalSubset0, externalSubset1, + condSect0, condSect1, condSect2, +#endif /* XML_DTD */ + declClose, + error; + +static int FASTCALL common(PROLOG_STATE *state, int tok); + +static int PTRCALL +prolog0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + state->handler = prolog1; + return XML_ROLE_NONE; + case XML_TOK_XML_DECL: + state->handler = prolog1; + return XML_ROLE_XML_DECL; + case XML_TOK_PI: + state->handler = prolog1; + return XML_ROLE_PI; + case XML_TOK_COMMENT: + state->handler = prolog1; + return XML_ROLE_COMMENT; + case XML_TOK_BOM: + return XML_ROLE_NONE; + case XML_TOK_DECL_OPEN: + if (!XmlNameMatchesAscii(enc, + ptr + 2 * MIN_BYTES_PER_CHAR(enc), + end, + KW_DOCTYPE)) + break; + state->handler = doctype0; + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_INSTANCE_START: + state->handler = error; + return XML_ROLE_INSTANCE_START; + } + return common(state, tok); +} + +static int PTRCALL +prolog1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NONE; + case XML_TOK_PI: + return XML_ROLE_PI; + case XML_TOK_COMMENT: + return XML_ROLE_COMMENT; + case XML_TOK_BOM: + return XML_ROLE_NONE; + case XML_TOK_DECL_OPEN: + if (!XmlNameMatchesAscii(enc, + ptr + 2 * MIN_BYTES_PER_CHAR(enc), + end, + KW_DOCTYPE)) + break; + state->handler = doctype0; + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_INSTANCE_START: + state->handler = error; + return XML_ROLE_INSTANCE_START; + } + return common(state, tok); +} + +static int PTRCALL +prolog2(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NONE; + case XML_TOK_PI: + return XML_ROLE_PI; + case XML_TOK_COMMENT: + return XML_ROLE_COMMENT; + case XML_TOK_INSTANCE_START: + state->handler = error; + return XML_ROLE_INSTANCE_START; + } + return common(state, tok); +} + +static int PTRCALL +doctype0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = doctype1; + return XML_ROLE_DOCTYPE_NAME; + } + return common(state, tok); +} + +static int PTRCALL +doctype1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_OPEN_BRACKET: + state->handler = internalSubset; + return XML_ROLE_DOCTYPE_INTERNAL_SUBSET; + case XML_TOK_DECL_CLOSE: + state->handler = prolog2; + return XML_ROLE_DOCTYPE_CLOSE; + case XML_TOK_NAME: + if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) { + state->handler = doctype3; + return XML_ROLE_DOCTYPE_NONE; + } + if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) { + state->handler = doctype2; + return XML_ROLE_DOCTYPE_NONE; + } + break; + } + return common(state, tok); +} + +static int PTRCALL +doctype2(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_LITERAL: + state->handler = doctype3; + return XML_ROLE_DOCTYPE_PUBLIC_ID; + } + return common(state, tok); +} + +static int PTRCALL +doctype3(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_LITERAL: + state->handler = doctype4; + return XML_ROLE_DOCTYPE_SYSTEM_ID; + } + return common(state, tok); +} + +static int PTRCALL +doctype4(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_OPEN_BRACKET: + state->handler = internalSubset; + return XML_ROLE_DOCTYPE_INTERNAL_SUBSET; + case XML_TOK_DECL_CLOSE: + state->handler = prolog2; + return XML_ROLE_DOCTYPE_CLOSE; + } + return common(state, tok); +} + +static int PTRCALL +doctype5(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_DECL_CLOSE: + state->handler = prolog2; + return XML_ROLE_DOCTYPE_CLOSE; + } + return common(state, tok); +} + +static int PTRCALL +internalSubset(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NONE; + case XML_TOK_DECL_OPEN: + if (XmlNameMatchesAscii(enc, + ptr + 2 * MIN_BYTES_PER_CHAR(enc), + end, + KW_ENTITY)) { + state->handler = entity0; + return XML_ROLE_ENTITY_NONE; + } + if (XmlNameMatchesAscii(enc, + ptr + 2 * MIN_BYTES_PER_CHAR(enc), + end, + KW_ATTLIST)) { + state->handler = attlist0; + return XML_ROLE_ATTLIST_NONE; + } + if (XmlNameMatchesAscii(enc, + ptr + 2 * MIN_BYTES_PER_CHAR(enc), + end, + KW_ELEMENT)) { + state->handler = element0; + return XML_ROLE_ELEMENT_NONE; + } + if (XmlNameMatchesAscii(enc, + ptr + 2 * MIN_BYTES_PER_CHAR(enc), + end, + KW_NOTATION)) { + state->handler = notation0; + return XML_ROLE_NOTATION_NONE; + } + break; + case XML_TOK_PI: + return XML_ROLE_PI; + case XML_TOK_COMMENT: + return XML_ROLE_COMMENT; + case XML_TOK_PARAM_ENTITY_REF: + return XML_ROLE_PARAM_ENTITY_REF; + case XML_TOK_CLOSE_BRACKET: + state->handler = doctype5; + return XML_ROLE_DOCTYPE_NONE; + case XML_TOK_NONE: + return XML_ROLE_NONE; + } + return common(state, tok); +} + +#ifdef XML_DTD + +static int PTRCALL +externalSubset0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + state->handler = externalSubset1; + if (tok == XML_TOK_XML_DECL) + return XML_ROLE_TEXT_DECL; + return externalSubset1(state, tok, ptr, end, enc); +} + +static int PTRCALL +externalSubset1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_COND_SECT_OPEN: + state->handler = condSect0; + return XML_ROLE_NONE; + case XML_TOK_COND_SECT_CLOSE: + if (state->includeLevel == 0) + break; + state->includeLevel -= 1; + return XML_ROLE_NONE; + case XML_TOK_PROLOG_S: + return XML_ROLE_NONE; + case XML_TOK_CLOSE_BRACKET: + break; + case XML_TOK_NONE: + if (state->includeLevel) + break; + return XML_ROLE_NONE; + default: + return internalSubset(state, tok, ptr, end, enc); + } + return common(state, tok); +} + +#endif /* XML_DTD */ + +static int PTRCALL +entity0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_PERCENT: + state->handler = entity1; + return XML_ROLE_ENTITY_NONE; + case XML_TOK_NAME: + state->handler = entity2; + return XML_ROLE_GENERAL_ENTITY_NAME; + } + return common(state, tok); +} + +static int PTRCALL +entity1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_NAME: + state->handler = entity7; + return XML_ROLE_PARAM_ENTITY_NAME; + } + return common(state, tok); +} + +static int PTRCALL +entity2(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_NAME: + if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) { + state->handler = entity4; + return XML_ROLE_ENTITY_NONE; + } + if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) { + state->handler = entity3; + return XML_ROLE_ENTITY_NONE; + } + break; + case XML_TOK_LITERAL: + state->handler = declClose; + state->role_none = XML_ROLE_ENTITY_NONE; + return XML_ROLE_ENTITY_VALUE; + } + return common(state, tok); +} + +static int PTRCALL +entity3(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_LITERAL: + state->handler = entity4; + return XML_ROLE_ENTITY_PUBLIC_ID; + } + return common(state, tok); +} + +static int PTRCALL +entity4(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_LITERAL: + state->handler = entity5; + return XML_ROLE_ENTITY_SYSTEM_ID; + } + return common(state, tok); +} + +static int PTRCALL +entity5(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_DECL_CLOSE: + setTopLevel(state); + return XML_ROLE_ENTITY_COMPLETE; + case XML_TOK_NAME: + if (XmlNameMatchesAscii(enc, ptr, end, KW_NDATA)) { + state->handler = entity6; + return XML_ROLE_ENTITY_NONE; + } + break; + } + return common(state, tok); +} + +static int PTRCALL +entity6(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_NAME: + state->handler = declClose; + state->role_none = XML_ROLE_ENTITY_NONE; + return XML_ROLE_ENTITY_NOTATION_NAME; + } + return common(state, tok); +} + +static int PTRCALL +entity7(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_NAME: + if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) { + state->handler = entity9; + return XML_ROLE_ENTITY_NONE; + } + if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) { + state->handler = entity8; + return XML_ROLE_ENTITY_NONE; + } + break; + case XML_TOK_LITERAL: + state->handler = declClose; + state->role_none = XML_ROLE_ENTITY_NONE; + return XML_ROLE_ENTITY_VALUE; + } + return common(state, tok); +} + +static int PTRCALL +entity8(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_LITERAL: + state->handler = entity9; + return XML_ROLE_ENTITY_PUBLIC_ID; + } + return common(state, tok); +} + +static int PTRCALL +entity9(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_LITERAL: + state->handler = entity10; + return XML_ROLE_ENTITY_SYSTEM_ID; + } + return common(state, tok); +} + +static int PTRCALL +entity10(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ENTITY_NONE; + case XML_TOK_DECL_CLOSE: + setTopLevel(state); + return XML_ROLE_ENTITY_COMPLETE; + } + return common(state, tok); +} + +static int PTRCALL +notation0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NOTATION_NONE; + case XML_TOK_NAME: + state->handler = notation1; + return XML_ROLE_NOTATION_NAME; + } + return common(state, tok); +} + +static int PTRCALL +notation1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NOTATION_NONE; + case XML_TOK_NAME: + if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) { + state->handler = notation3; + return XML_ROLE_NOTATION_NONE; + } + if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) { + state->handler = notation2; + return XML_ROLE_NOTATION_NONE; + } + break; + } + return common(state, tok); +} + +static int PTRCALL +notation2(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NOTATION_NONE; + case XML_TOK_LITERAL: + state->handler = notation4; + return XML_ROLE_NOTATION_PUBLIC_ID; + } + return common(state, tok); +} + +static int PTRCALL +notation3(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NOTATION_NONE; + case XML_TOK_LITERAL: + state->handler = declClose; + state->role_none = XML_ROLE_NOTATION_NONE; + return XML_ROLE_NOTATION_SYSTEM_ID; + } + return common(state, tok); +} + +static int PTRCALL +notation4(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NOTATION_NONE; + case XML_TOK_LITERAL: + state->handler = declClose; + state->role_none = XML_ROLE_NOTATION_NONE; + return XML_ROLE_NOTATION_SYSTEM_ID; + case XML_TOK_DECL_CLOSE: + setTopLevel(state); + return XML_ROLE_NOTATION_NO_SYSTEM_ID; + } + return common(state, tok); +} + +static int PTRCALL +attlist0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = attlist1; + return XML_ROLE_ATTLIST_ELEMENT_NAME; + } + return common(state, tok); +} + +static int PTRCALL +attlist1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_DECL_CLOSE: + setTopLevel(state); + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = attlist2; + return XML_ROLE_ATTRIBUTE_NAME; + } + return common(state, tok); +} + +static int PTRCALL +attlist2(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_NAME: + { + static const char * const types[] = { + KW_CDATA, + KW_ID, + KW_IDREF, + KW_IDREFS, + KW_ENTITY, + KW_ENTITIES, + KW_NMTOKEN, + KW_NMTOKENS, + }; + int i; + for (i = 0; i < (int)(sizeof(types)/sizeof(types[0])); i++) + if (XmlNameMatchesAscii(enc, ptr, end, types[i])) { + state->handler = attlist8; + return XML_ROLE_ATTRIBUTE_TYPE_CDATA + i; + } + } + if (XmlNameMatchesAscii(enc, ptr, end, KW_NOTATION)) { + state->handler = attlist5; + return XML_ROLE_ATTLIST_NONE; + } + break; + case XML_TOK_OPEN_PAREN: + state->handler = attlist3; + return XML_ROLE_ATTLIST_NONE; + } + return common(state, tok); +} + +static int PTRCALL +attlist3(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_NMTOKEN: + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = attlist4; + return XML_ROLE_ATTRIBUTE_ENUM_VALUE; + } + return common(state, tok); +} + +static int PTRCALL +attlist4(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_CLOSE_PAREN: + state->handler = attlist8; + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_OR: + state->handler = attlist3; + return XML_ROLE_ATTLIST_NONE; + } + return common(state, tok); +} + +static int PTRCALL +attlist5(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_OPEN_PAREN: + state->handler = attlist6; + return XML_ROLE_ATTLIST_NONE; + } + return common(state, tok); +} + +static int PTRCALL +attlist6(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_NAME: + state->handler = attlist7; + return XML_ROLE_ATTRIBUTE_NOTATION_VALUE; + } + return common(state, tok); +} + +static int PTRCALL +attlist7(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_CLOSE_PAREN: + state->handler = attlist8; + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_OR: + state->handler = attlist6; + return XML_ROLE_ATTLIST_NONE; + } + return common(state, tok); +} + +/* default value */ +static int PTRCALL +attlist8(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_POUND_NAME: + if (XmlNameMatchesAscii(enc, + ptr + MIN_BYTES_PER_CHAR(enc), + end, + KW_IMPLIED)) { + state->handler = attlist1; + return XML_ROLE_IMPLIED_ATTRIBUTE_VALUE; + } + if (XmlNameMatchesAscii(enc, + ptr + MIN_BYTES_PER_CHAR(enc), + end, + KW_REQUIRED)) { + state->handler = attlist1; + return XML_ROLE_REQUIRED_ATTRIBUTE_VALUE; + } + if (XmlNameMatchesAscii(enc, + ptr + MIN_BYTES_PER_CHAR(enc), + end, + KW_FIXED)) { + state->handler = attlist9; + return XML_ROLE_ATTLIST_NONE; + } + break; + case XML_TOK_LITERAL: + state->handler = attlist1; + return XML_ROLE_DEFAULT_ATTRIBUTE_VALUE; + } + return common(state, tok); +} + +static int PTRCALL +attlist9(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ATTLIST_NONE; + case XML_TOK_LITERAL: + state->handler = attlist1; + return XML_ROLE_FIXED_ATTRIBUTE_VALUE; + } + return common(state, tok); +} + +static int PTRCALL +element0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = element1; + return XML_ROLE_ELEMENT_NAME; + } + return common(state, tok); +} + +static int PTRCALL +element1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_NAME: + if (XmlNameMatchesAscii(enc, ptr, end, KW_EMPTY)) { + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + return XML_ROLE_CONTENT_EMPTY; + } + if (XmlNameMatchesAscii(enc, ptr, end, KW_ANY)) { + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + return XML_ROLE_CONTENT_ANY; + } + break; + case XML_TOK_OPEN_PAREN: + state->handler = element2; + state->level = 1; + return XML_ROLE_GROUP_OPEN; + } + return common(state, tok); +} + +static int PTRCALL +element2(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_POUND_NAME: + if (XmlNameMatchesAscii(enc, + ptr + MIN_BYTES_PER_CHAR(enc), + end, + KW_PCDATA)) { + state->handler = element3; + return XML_ROLE_CONTENT_PCDATA; + } + break; + case XML_TOK_OPEN_PAREN: + state->level = 2; + state->handler = element6; + return XML_ROLE_GROUP_OPEN; + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT; + case XML_TOK_NAME_QUESTION: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT_OPT; + case XML_TOK_NAME_ASTERISK: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT_REP; + case XML_TOK_NAME_PLUS: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT_PLUS; + } + return common(state, tok); +} + +static int PTRCALL +element3(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_CLOSE_PAREN: + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + return XML_ROLE_GROUP_CLOSE; + case XML_TOK_CLOSE_PAREN_ASTERISK: + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + return XML_ROLE_GROUP_CLOSE_REP; + case XML_TOK_OR: + state->handler = element4; + return XML_ROLE_ELEMENT_NONE; + } + return common(state, tok); +} + +static int PTRCALL +element4(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = element5; + return XML_ROLE_CONTENT_ELEMENT; + } + return common(state, tok); +} + +static int PTRCALL +element5(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_CLOSE_PAREN_ASTERISK: + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + return XML_ROLE_GROUP_CLOSE_REP; + case XML_TOK_OR: + state->handler = element4; + return XML_ROLE_ELEMENT_NONE; + } + return common(state, tok); +} + +static int PTRCALL +element6(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_OPEN_PAREN: + state->level += 1; + return XML_ROLE_GROUP_OPEN; + case XML_TOK_NAME: + case XML_TOK_PREFIXED_NAME: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT; + case XML_TOK_NAME_QUESTION: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT_OPT; + case XML_TOK_NAME_ASTERISK: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT_REP; + case XML_TOK_NAME_PLUS: + state->handler = element7; + return XML_ROLE_CONTENT_ELEMENT_PLUS; + } + return common(state, tok); +} + +static int PTRCALL +element7(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_ELEMENT_NONE; + case XML_TOK_CLOSE_PAREN: + state->level -= 1; + if (state->level == 0) { + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + } + return XML_ROLE_GROUP_CLOSE; + case XML_TOK_CLOSE_PAREN_ASTERISK: + state->level -= 1; + if (state->level == 0) { + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + } + return XML_ROLE_GROUP_CLOSE_REP; + case XML_TOK_CLOSE_PAREN_QUESTION: + state->level -= 1; + if (state->level == 0) { + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + } + return XML_ROLE_GROUP_CLOSE_OPT; + case XML_TOK_CLOSE_PAREN_PLUS: + state->level -= 1; + if (state->level == 0) { + state->handler = declClose; + state->role_none = XML_ROLE_ELEMENT_NONE; + } + return XML_ROLE_GROUP_CLOSE_PLUS; + case XML_TOK_COMMA: + state->handler = element6; + return XML_ROLE_GROUP_SEQUENCE; + case XML_TOK_OR: + state->handler = element6; + return XML_ROLE_GROUP_CHOICE; + } + return common(state, tok); +} + +#ifdef XML_DTD + +static int PTRCALL +condSect0(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NONE; + case XML_TOK_NAME: + if (XmlNameMatchesAscii(enc, ptr, end, KW_INCLUDE)) { + state->handler = condSect1; + return XML_ROLE_NONE; + } + if (XmlNameMatchesAscii(enc, ptr, end, KW_IGNORE)) { + state->handler = condSect2; + return XML_ROLE_NONE; + } + break; + } + return common(state, tok); +} + +static int PTRCALL +condSect1(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NONE; + case XML_TOK_OPEN_BRACKET: + state->handler = externalSubset1; + state->includeLevel += 1; + return XML_ROLE_NONE; + } + return common(state, tok); +} + +static int PTRCALL +condSect2(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return XML_ROLE_NONE; + case XML_TOK_OPEN_BRACKET: + state->handler = externalSubset1; + return XML_ROLE_IGNORE_SECT; + } + return common(state, tok); +} + +#endif /* XML_DTD */ + +static int PTRCALL +declClose(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + switch (tok) { + case XML_TOK_PROLOG_S: + return state->role_none; + case XML_TOK_DECL_CLOSE: + setTopLevel(state); + return state->role_none; + } + return common(state, tok); +} + +static int PTRCALL +error(PROLOG_STATE *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc) +{ + UNUSED(state); + UNUSED(tok); + UNUSED(ptr); + UNUSED(end); + UNUSED(enc); + + return XML_ROLE_NONE; +} + +static int FASTCALL +common(PROLOG_STATE *state, int tok) +{ +#ifdef XML_DTD + if (!state->documentEntity && tok == XML_TOK_PARAM_ENTITY_REF) + return XML_ROLE_INNER_PARAM_ENTITY_REF; +#endif + state->handler = error; + return XML_ROLE_ERROR; +} + +void +XmlPrologStateInit(PROLOG_STATE *state) +{ + state->handler = prolog0; +#ifdef XML_DTD + state->documentEntity = 1; + state->includeLevel = 0; + state->inEntityValue = 0; +#endif /* XML_DTD */ +} + +#ifdef XML_DTD + +void +XmlPrologStateInitExternalEntity(PROLOG_STATE *state) +{ + state->handler = externalSubset0; + state->documentEntity = 0; + state->includeLevel = 0; +} + +#endif /* XML_DTD */ diff --git a/xml/details/expat/xmlrole.h b/xml/details/expat/xmlrole.h new file mode 100644 index 0000000..0a0d6b8 --- /dev/null +++ b/xml/details/expat/xmlrole.h @@ -0,0 +1,114 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#ifndef XmlRole_INCLUDED +#define XmlRole_INCLUDED 1 + +#ifdef __VMS +/* 0 1 2 3 0 1 2 3 + 1234567890123456789012345678901 1234567890123456789012345678901 */ +#define XmlPrologStateInitExternalEntity XmlPrologStateInitExternalEnt +#endif + +#include <xml/details/expat/xmltok.h> + +#ifdef __cplusplus +extern "C" { +#endif + +enum { + XML_ROLE_ERROR = -1, + XML_ROLE_NONE = 0, + XML_ROLE_XML_DECL, + XML_ROLE_INSTANCE_START, + XML_ROLE_DOCTYPE_NONE, + XML_ROLE_DOCTYPE_NAME, + XML_ROLE_DOCTYPE_SYSTEM_ID, + XML_ROLE_DOCTYPE_PUBLIC_ID, + XML_ROLE_DOCTYPE_INTERNAL_SUBSET, + XML_ROLE_DOCTYPE_CLOSE, + XML_ROLE_GENERAL_ENTITY_NAME, + XML_ROLE_PARAM_ENTITY_NAME, + XML_ROLE_ENTITY_NONE, + XML_ROLE_ENTITY_VALUE, + XML_ROLE_ENTITY_SYSTEM_ID, + XML_ROLE_ENTITY_PUBLIC_ID, + XML_ROLE_ENTITY_COMPLETE, + XML_ROLE_ENTITY_NOTATION_NAME, + XML_ROLE_NOTATION_NONE, + XML_ROLE_NOTATION_NAME, + XML_ROLE_NOTATION_SYSTEM_ID, + XML_ROLE_NOTATION_NO_SYSTEM_ID, + XML_ROLE_NOTATION_PUBLIC_ID, + XML_ROLE_ATTRIBUTE_NAME, + XML_ROLE_ATTRIBUTE_TYPE_CDATA, + XML_ROLE_ATTRIBUTE_TYPE_ID, + XML_ROLE_ATTRIBUTE_TYPE_IDREF, + XML_ROLE_ATTRIBUTE_TYPE_IDREFS, + XML_ROLE_ATTRIBUTE_TYPE_ENTITY, + XML_ROLE_ATTRIBUTE_TYPE_ENTITIES, + XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN, + XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS, + XML_ROLE_ATTRIBUTE_ENUM_VALUE, + XML_ROLE_ATTRIBUTE_NOTATION_VALUE, + XML_ROLE_ATTLIST_NONE, + XML_ROLE_ATTLIST_ELEMENT_NAME, + XML_ROLE_IMPLIED_ATTRIBUTE_VALUE, + XML_ROLE_REQUIRED_ATTRIBUTE_VALUE, + XML_ROLE_DEFAULT_ATTRIBUTE_VALUE, + XML_ROLE_FIXED_ATTRIBUTE_VALUE, + XML_ROLE_ELEMENT_NONE, + XML_ROLE_ELEMENT_NAME, + XML_ROLE_CONTENT_ANY, + XML_ROLE_CONTENT_EMPTY, + XML_ROLE_CONTENT_PCDATA, + XML_ROLE_GROUP_OPEN, + XML_ROLE_GROUP_CLOSE, + XML_ROLE_GROUP_CLOSE_REP, + XML_ROLE_GROUP_CLOSE_OPT, + XML_ROLE_GROUP_CLOSE_PLUS, + XML_ROLE_GROUP_CHOICE, + XML_ROLE_GROUP_SEQUENCE, + XML_ROLE_CONTENT_ELEMENT, + XML_ROLE_CONTENT_ELEMENT_REP, + XML_ROLE_CONTENT_ELEMENT_OPT, + XML_ROLE_CONTENT_ELEMENT_PLUS, + XML_ROLE_PI, + XML_ROLE_COMMENT, +#ifdef XML_DTD + XML_ROLE_TEXT_DECL, + XML_ROLE_IGNORE_SECT, + XML_ROLE_INNER_PARAM_ENTITY_REF, +#endif /* XML_DTD */ + XML_ROLE_PARAM_ENTITY_REF +}; + +typedef struct prolog_state { + int (PTRCALL *handler) (struct prolog_state *state, + int tok, + const char *ptr, + const char *end, + const ENCODING *enc); + unsigned level; + int role_none; +#ifdef XML_DTD + unsigned includeLevel; + int documentEntity; + int inEntityValue; +#endif /* XML_DTD */ +} PROLOG_STATE; + +void XmlPrologStateInit(PROLOG_STATE *); +#ifdef XML_DTD +void XmlPrologStateInitExternalEntity(PROLOG_STATE *); +#endif /* XML_DTD */ + +#define XmlTokenRole(state, tok, ptr, end, enc) \ + (((state)->handler)(state, tok, ptr, end, enc)) + +#ifdef __cplusplus +} +#endif + +#endif /* not XmlRole_INCLUDED */ diff --git a/xml/details/expat/xmltok.c b/xml/details/expat/xmltok.c new file mode 100644 index 0000000..b15fcc8 --- /dev/null +++ b/xml/details/expat/xmltok.c @@ -0,0 +1,1681 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#include <stddef.h> + +#include <xml/details/expat/config.h> + +#include <xml/details/expat/expat_external.h> +#include <xml/details/expat/internal.h> +#include <xml/details/expat/xmltok.h> +#include <xml/details/expat/nametab.h> + +#ifdef XML_DTD +#define IGNORE_SECTION_TOK_VTABLE , PREFIX(ignoreSectionTok) +#else +#define IGNORE_SECTION_TOK_VTABLE /* as nothing */ +#endif + +#define VTABLE1 \ + { PREFIX(prologTok), PREFIX(contentTok), \ + PREFIX(cdataSectionTok) IGNORE_SECTION_TOK_VTABLE }, \ + { PREFIX(attributeValueTok), PREFIX(entityValueTok) }, \ + PREFIX(sameName), \ + PREFIX(nameMatchesAscii), \ + PREFIX(nameLength), \ + PREFIX(skipS), \ + PREFIX(getAtts), \ + PREFIX(charRefNumber), \ + PREFIX(predefinedEntityName), \ + PREFIX(updatePosition), \ + PREFIX(isPublicId) + +#define VTABLE VTABLE1, PREFIX(toUtf8), PREFIX(toUtf16) + +#define UCS2_GET_NAMING(pages, hi, lo) \ + (namingBitmap[(pages[hi] << 3) + ((lo) >> 5)] & (1 << ((lo) & 0x1F))) + +/* A 2 byte UTF-8 representation splits the characters 11 bits between + the bottom 5 and 6 bits of the bytes. We need 8 bits to index into + pages, 3 bits to add to that index and 5 bits to generate the mask. +*/ +#define UTF8_GET_NAMING2(pages, byte) \ + (namingBitmap[((pages)[(((byte)[0]) >> 2) & 7] << 3) \ + + ((((byte)[0]) & 3) << 1) \ + + ((((byte)[1]) >> 5) & 1)] \ + & (1 << (((byte)[1]) & 0x1F))) + +/* A 3 byte UTF-8 representation splits the characters 16 bits between + the bottom 4, 6 and 6 bits of the bytes. We need 8 bits to index + into pages, 3 bits to add to that index and 5 bits to generate the + mask. +*/ +#define UTF8_GET_NAMING3(pages, byte) \ + (namingBitmap[((pages)[((((byte)[0]) & 0xF) << 4) \ + + ((((byte)[1]) >> 2) & 0xF)] \ + << 3) \ + + ((((byte)[1]) & 3) << 1) \ + + ((((byte)[2]) >> 5) & 1)] \ + & (1 << (((byte)[2]) & 0x1F))) + +#define UTF8_GET_NAMING(pages, p, n) \ + ((n) == 2 \ + ? UTF8_GET_NAMING2(pages, (const unsigned char *)(p)) \ + : ((n) == 3 \ + ? UTF8_GET_NAMING3(pages, (const unsigned char *)(p)) \ + : 0)) + +/* Detection of invalid UTF-8 sequences is based on Table 3.1B + of Unicode 3.2: http://www.unicode.org/unicode/reports/tr28/ + with the additional restriction of not allowing the Unicode + code points 0xFFFF and 0xFFFE (sequences EF,BF,BF and EF,BF,BE). + Implementation details: + (A & 0x80) == 0 means A < 0x80 + and + (A & 0xC0) == 0xC0 means A > 0xBF +*/ + +#define UTF8_INVALID2(p) \ + ((*p) < 0xC2 || ((p)[1] & 0x80) == 0 || ((p)[1] & 0xC0) == 0xC0) + +#define UTF8_INVALID3(p) \ + (((p)[2] & 0x80) == 0 \ + || \ + ((*p) == 0xEF && (p)[1] == 0xBF \ + ? \ + (p)[2] > 0xBD \ + : \ + ((p)[2] & 0xC0) == 0xC0) \ + || \ + ((*p) == 0xE0 \ + ? \ + (p)[1] < 0xA0 || ((p)[1] & 0xC0) == 0xC0 \ + : \ + ((p)[1] & 0x80) == 0 \ + || \ + ((*p) == 0xED ? (p)[1] > 0x9F : ((p)[1] & 0xC0) == 0xC0))) + +#define UTF8_INVALID4(p) \ + (((p)[3] & 0x80) == 0 || ((p)[3] & 0xC0) == 0xC0 \ + || \ + ((p)[2] & 0x80) == 0 || ((p)[2] & 0xC0) == 0xC0 \ + || \ + ((*p) == 0xF0 \ + ? \ + (p)[1] < 0x90 || ((p)[1] & 0xC0) == 0xC0 \ + : \ + ((p)[1] & 0x80) == 0 \ + || \ + ((*p) == 0xF4 ? (p)[1] > 0x8F : ((p)[1] & 0xC0) == 0xC0))) + +static int PTRFASTCALL +isNever(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + UNUSED(p); + + return 0; +} + +static int PTRFASTCALL +utf8_isName2(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + + return UTF8_GET_NAMING2(namePages, (const unsigned char *)p); +} + +static int PTRFASTCALL +utf8_isName3(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + + return UTF8_GET_NAMING3(namePages, (const unsigned char *)p); +} + +#define utf8_isName4 isNever + +static int PTRFASTCALL +utf8_isNmstrt2(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + + return UTF8_GET_NAMING2(nmstrtPages, (const unsigned char *)p); +} + +static int PTRFASTCALL +utf8_isNmstrt3(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + + return UTF8_GET_NAMING3(nmstrtPages, (const unsigned char *)p); +} + +#define utf8_isNmstrt4 isNever + +static int PTRFASTCALL +utf8_isInvalid2(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + + return UTF8_INVALID2((const unsigned char *)p); +} + +static int PTRFASTCALL +utf8_isInvalid3(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + + return UTF8_INVALID3((const unsigned char *)p); +} + +static int PTRFASTCALL +utf8_isInvalid4(const ENCODING *enc, const char *p) +{ + UNUSED(enc); + + return UTF8_INVALID4((const unsigned char *)p); +} + +struct normal_encoding { + ENCODING enc; + unsigned char type[256]; +#ifdef XML_MIN_SIZE + int (PTRFASTCALL *byteType)(const ENCODING *, const char *); + int (PTRFASTCALL *isNameMin)(const ENCODING *, const char *); + int (PTRFASTCALL *isNmstrtMin)(const ENCODING *, const char *); + int (PTRFASTCALL *byteToAscii)(const ENCODING *, const char *); + int (PTRCALL *charMatches)(const ENCODING *, const char *, int); +#endif /* XML_MIN_SIZE */ + int (PTRFASTCALL *isName2)(const ENCODING *, const char *); + int (PTRFASTCALL *isName3)(const ENCODING *, const char *); + int (PTRFASTCALL *isName4)(const ENCODING *, const char *); + int (PTRFASTCALL *isNmstrt2)(const ENCODING *, const char *); + int (PTRFASTCALL *isNmstrt3)(const ENCODING *, const char *); + int (PTRFASTCALL *isNmstrt4)(const ENCODING *, const char *); + int (PTRFASTCALL *isInvalid2)(const ENCODING *, const char *); + int (PTRFASTCALL *isInvalid3)(const ENCODING *, const char *); + int (PTRFASTCALL *isInvalid4)(const ENCODING *, const char *); +}; + +#define AS_NORMAL_ENCODING(enc) ((const struct normal_encoding *) (enc)) + +#ifdef XML_MIN_SIZE + +#define STANDARD_VTABLE(E) \ + E ## byteType, \ + E ## isNameMin, \ + E ## isNmstrtMin, \ + E ## byteToAscii, \ + E ## charMatches, + +#define ZERO_VTABLE /* as nothing */ + +#else + +#define STANDARD_VTABLE(E) /* as nothing */ + +#define ZERO_VTABLE \ + 0, \ + 0, \ + 0, \ + 0, \ + 0, \ + 0, \ + 0, \ + 0, \ + 0 + +#endif + +#define NORMAL_VTABLE(E) \ + E ## isName2, \ + E ## isName3, \ + E ## isName4, \ + E ## isNmstrt2, \ + E ## isNmstrt3, \ + E ## isNmstrt4, \ + E ## isInvalid2, \ + E ## isInvalid3, \ + E ## isInvalid4 + +static int FASTCALL checkCharRefNumber(int); + +#include <xml/details/expat/xmltok_impl.h> +#include <xml/details/expat/ascii.h> + +#ifdef XML_MIN_SIZE +#define sb_isNameMin isNever +#define sb_isNmstrtMin isNever +#endif + +#ifdef XML_MIN_SIZE +#define MINBPC(enc) ((enc)->minBytesPerChar) +#else +/* minimum bytes per character */ +#define MINBPC(enc) 1 +#endif + +#define SB_BYTE_TYPE(enc, p) \ + (((struct normal_encoding *)(enc))->type[(unsigned char)*(p)]) + +#ifdef XML_MIN_SIZE +static int PTRFASTCALL +sb_byteType(const ENCODING *enc, const char *p) +{ + return SB_BYTE_TYPE(enc, p); +} +#define BYTE_TYPE(enc, p) \ + (AS_NORMAL_ENCODING(enc)->byteType(enc, p)) +#else +#define BYTE_TYPE(enc, p) SB_BYTE_TYPE(enc, p) +#endif + +#ifdef XML_MIN_SIZE +#define BYTE_TO_ASCII(enc, p) \ + (AS_NORMAL_ENCODING(enc)->byteToAscii(enc, p)) +static int PTRFASTCALL +sb_byteToAscii(const ENCODING *enc, const char *p) +{ + return *p; +} +#else +#define BYTE_TO_ASCII(enc, p) (*(p)) +#endif + +#define IS_NAME_CHAR(enc, p, n) \ + (AS_NORMAL_ENCODING(enc)->isName ## n(enc, p)) +#define IS_NMSTRT_CHAR(enc, p, n) \ + (AS_NORMAL_ENCODING(enc)->isNmstrt ## n(enc, p)) +#define IS_INVALID_CHAR(enc, p, n) \ + (AS_NORMAL_ENCODING(enc)->isInvalid ## n(enc, p)) + +#ifdef XML_MIN_SIZE +#define IS_NAME_CHAR_MINBPC(enc, p) \ + (AS_NORMAL_ENCODING(enc)->isNameMin(enc, p)) +#define IS_NMSTRT_CHAR_MINBPC(enc, p) \ + (AS_NORMAL_ENCODING(enc)->isNmstrtMin(enc, p)) +#else +#define IS_NAME_CHAR_MINBPC(enc, p) (0) +#define IS_NMSTRT_CHAR_MINBPC(enc, p) (0) +#endif + +#ifdef XML_MIN_SIZE +#define CHAR_MATCHES(enc, p, c) \ + (AS_NORMAL_ENCODING(enc)->charMatches(enc, p, c)) +static int PTRCALL +sb_charMatches(const ENCODING *enc, const char *p, int c) +{ + return *p == c; +} +#else +/* c is an ASCII character */ +#define CHAR_MATCHES(enc, p, c) (*(p) == c) +#endif + +#define PREFIX(ident) normal_ ## ident +#define XML_TOK_IMPL_C +#include <xml/details/expat/xmltok_impl.c> +#undef XML_TOK_IMPL_C + +#undef MINBPC +#undef BYTE_TYPE +#undef BYTE_TO_ASCII +#undef CHAR_MATCHES +#undef IS_NAME_CHAR +#undef IS_NAME_CHAR_MINBPC +#undef IS_NMSTRT_CHAR +#undef IS_NMSTRT_CHAR_MINBPC +#undef IS_INVALID_CHAR + +enum { /* UTF8_cvalN is value of masked first byte of N byte sequence */ + UTF8_cval1 = 0x00, + UTF8_cval2 = 0xc0, + UTF8_cval3 = 0xe0, + UTF8_cval4 = 0xf0 +}; + +static void PTRCALL +utf8_toUtf8(const ENCODING *enc, + const char **fromP, const char *fromLim, + char **toP, const char *toLim) +{ + char *to; + const char *from; + + UNUSED(enc); + + if (fromLim - *fromP > toLim - *toP) { + /* Avoid copying partial characters. */ + for (fromLim = *fromP + (toLim - *toP); fromLim > *fromP; fromLim--) + if (((unsigned char)fromLim[-1] & 0xc0) != 0x80) + break; + } + for (to = *toP, from = *fromP; from != fromLim; from++, to++) + *to = *from; + *fromP = from; + *toP = to; +} + +static void PTRCALL +utf8_toUtf16(const ENCODING *enc, + const char **fromP, const char *fromLim, + unsigned short **toP, const unsigned short *toLim) +{ + unsigned short *to = *toP; + const char *from = *fromP; + while (from != fromLim && to != toLim) { + switch (((struct normal_encoding *)enc)->type[(unsigned char)*from]) { + case BT_LEAD2: + *to++ = (unsigned short)(((from[0] & 0x1f) << 6) | (from[1] & 0x3f)); + from += 2; + break; + case BT_LEAD3: + *to++ = (unsigned short)(((from[0] & 0xf) << 12) + | ((from[1] & 0x3f) << 6) | (from[2] & 0x3f)); + from += 3; + break; + case BT_LEAD4: + { + unsigned long n; + if (to + 1 == toLim) + goto after; + n = ((from[0] & 0x7) << 18) | ((from[1] & 0x3f) << 12) + | ((from[2] & 0x3f) << 6) | (from[3] & 0x3f); + n -= 0x10000; + to[0] = (unsigned short)((n >> 10) | 0xD800); + to[1] = (unsigned short)((n & 0x3FF) | 0xDC00); + to += 2; + from += 4; + } + break; + default: + *to++ = *from++; + break; + } + } +after: + *fromP = from; + *toP = to; +} + +#ifdef XML_NS +static const struct normal_encoding utf8_encoding_ns = { + { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 }, + { +#include <xml/details/expat/asciitab.h> +#include <xml/details/expat/utf8tab.h> + }, + STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_) +}; +#endif + +static const struct normal_encoding utf8_encoding = { + { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/asciitab.h> +#undef BT_COLON +#include <xml/details/expat/utf8tab.h> + }, + STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_) +}; + +#ifdef XML_NS + +static const struct normal_encoding internal_utf8_encoding_ns = { + { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 }, + { +#include <xml/details/expat/iasciitab.h> +#include <xml/details/expat/utf8tab.h> + }, + STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_) +}; + +#endif + +static const struct normal_encoding internal_utf8_encoding = { + { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/iasciitab.h> +#undef BT_COLON +#include <xml/details/expat/utf8tab.h> + }, + STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_) +}; + +static void PTRCALL +latin1_toUtf8(const ENCODING *enc, + const char **fromP, const char *fromLim, + char **toP, const char *toLim) +{ + UNUSED(enc); + + for (;;) { + unsigned char c; + if (*fromP == fromLim) + break; + c = (unsigned char)**fromP; + if (c & 0x80) { + if (toLim - *toP < 2) + break; + *(*toP)++ = (char)((c >> 6) | UTF8_cval2); + *(*toP)++ = (char)((c & 0x3f) | 0x80); + (*fromP)++; + } + else { + if (*toP == toLim) + break; + *(*toP)++ = *(*fromP)++; + } + } +} + +static void PTRCALL +latin1_toUtf16(const ENCODING *enc, + const char **fromP, const char *fromLim, + unsigned short **toP, const unsigned short *toLim) +{ + UNUSED(enc); + + while (*fromP != fromLim && *toP != toLim) + *(*toP)++ = (unsigned char)*(*fromP)++; +} + +#ifdef XML_NS + +static const struct normal_encoding latin1_encoding_ns = { + { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 }, + { +#include <xml/details/expat/asciitab.h> +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(sb_) ZERO_VTABLE +}; + +#endif + +static const struct normal_encoding latin1_encoding = { + { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/asciitab.h> +#undef BT_COLON +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(sb_) ZERO_VTABLE +}; + +static void PTRCALL +ascii_toUtf8(const ENCODING *enc, + const char **fromP, const char *fromLim, + char **toP, const char *toLim) +{ + UNUSED(enc); + + while (*fromP != fromLim && *toP != toLim) + *(*toP)++ = *(*fromP)++; +} + +#ifdef XML_NS + +static const struct normal_encoding ascii_encoding_ns = { + { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 }, + { +#include <xml/details/expat/asciitab.h> +/* BT_NONXML == 0 */ + }, + STANDARD_VTABLE(sb_) ZERO_VTABLE +}; + +#endif + +static const struct normal_encoding ascii_encoding = { + { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/asciitab.h> +#undef BT_COLON +/* BT_NONXML == 0 */ + }, + STANDARD_VTABLE(sb_) ZERO_VTABLE +}; + +static int PTRFASTCALL +unicode_byte_type(char hi, char lo) +{ + switch ((unsigned char)hi) { + case 0xD8: case 0xD9: case 0xDA: case 0xDB: + return BT_LEAD4; + case 0xDC: case 0xDD: case 0xDE: case 0xDF: + return BT_TRAIL; + case 0xFF: + switch ((unsigned char)lo) { + case 0xFF: + case 0xFE: + return BT_NONXML; + } + break; + } + return BT_NONASCII; +} + +#define DEFINE_UTF16_TO_UTF8(E) \ +static void PTRCALL \ +E ## toUtf8(const ENCODING *enc, \ + const char **fromP, const char *fromLim, \ + char **toP, const char *toLim) \ +{ \ + const char *from; \ + UNUSED(enc); \ + for (from = *fromP; from != fromLim; from += 2) { \ + int plane; \ + unsigned char lo2; \ + unsigned char lo = GET_LO(from); \ + unsigned char hi = GET_HI(from); \ + switch (hi) { \ + case 0: \ + if (lo < 0x80) { \ + if (*toP == toLim) { \ + *fromP = from; \ + return; \ + } \ + *(*toP)++ = lo; \ + break; \ + } \ + /* fall through */ \ + case 0x1: case 0x2: case 0x3: \ + case 0x4: case 0x5: case 0x6: case 0x7: \ + if (toLim - *toP < 2) { \ + *fromP = from; \ + return; \ + } \ + *(*toP)++ = ((lo >> 6) | (hi << 2) | UTF8_cval2); \ + *(*toP)++ = ((lo & 0x3f) | 0x80); \ + break; \ + default: \ + if (toLim - *toP < 3) { \ + *fromP = from; \ + return; \ + } \ + /* 16 bits divided 4, 6, 6 amongst 3 bytes */ \ + *(*toP)++ = ((hi >> 4) | UTF8_cval3); \ + *(*toP)++ = (((hi & 0xf) << 2) | (lo >> 6) | 0x80); \ + *(*toP)++ = ((lo & 0x3f) | 0x80); \ + break; \ + case 0xD8: case 0xD9: case 0xDA: case 0xDB: \ + if (toLim - *toP < 4) { \ + *fromP = from; \ + return; \ + } \ + plane = (((hi & 0x3) << 2) | ((lo >> 6) & 0x3)) + 1; \ + *(*toP)++ = ((plane >> 2) | UTF8_cval4); \ + *(*toP)++ = (((lo >> 2) & 0xF) | ((plane & 0x3) << 4) | 0x80); \ + from += 2; \ + lo2 = GET_LO(from); \ + *(*toP)++ = (((lo & 0x3) << 4) \ + | ((GET_HI(from) & 0x3) << 2) \ + | (lo2 >> 6) \ + | 0x80); \ + *(*toP)++ = ((lo2 & 0x3f) | 0x80); \ + break; \ + } \ + } \ + *fromP = from; \ +} + +#define DEFINE_UTF16_TO_UTF16(E) \ +static void PTRCALL \ +E ## toUtf16(const ENCODING *enc, \ + const char **fromP, const char *fromLim, \ + unsigned short **toP, const unsigned short *toLim) \ +{ \ + UNUSED(enc); \ + /* Avoid copying first half only of surrogate */ \ + if (fromLim - *fromP > ((toLim - *toP) << 1) \ + && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) \ + fromLim -= 2; \ + for (; *fromP != fromLim && *toP != toLim; *fromP += 2) \ + *(*toP)++ = (GET_HI(*fromP) << 8) | GET_LO(*fromP); \ +} + +#define SET2(ptr, ch) \ + (((ptr)[0] = ((ch) & 0xff)), ((ptr)[1] = ((ch) >> 8))) +#define GET_LO(ptr) ((unsigned char)(ptr)[0]) +#define GET_HI(ptr) ((unsigned char)(ptr)[1]) + +DEFINE_UTF16_TO_UTF8(little2_) +DEFINE_UTF16_TO_UTF16(little2_) + +#undef SET2 +#undef GET_LO +#undef GET_HI + +#define SET2(ptr, ch) \ + (((ptr)[0] = ((ch) >> 8)), ((ptr)[1] = ((ch) & 0xFF))) +#define GET_LO(ptr) ((unsigned char)(ptr)[1]) +#define GET_HI(ptr) ((unsigned char)(ptr)[0]) + +DEFINE_UTF16_TO_UTF8(big2_) +DEFINE_UTF16_TO_UTF16(big2_) + +#undef SET2 +#undef GET_LO +#undef GET_HI + +#define LITTLE2_BYTE_TYPE(enc, p) \ + ((p)[1] == 0 \ + ? ((struct normal_encoding *)(enc))->type[(unsigned char)*(p)] \ + : unicode_byte_type((p)[1], (p)[0])) +#define LITTLE2_BYTE_TO_ASCII(enc, p) ((p)[1] == 0 ? (p)[0] : -1) +#define LITTLE2_CHAR_MATCHES(enc, p, c) ((p)[1] == 0 && (p)[0] == c) +#define LITTLE2_IS_NAME_CHAR_MINBPC(enc, p) \ + UCS2_GET_NAMING(namePages, (unsigned char)p[1], (unsigned char)p[0]) +#define LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p) \ + UCS2_GET_NAMING(nmstrtPages, (unsigned char)p[1], (unsigned char)p[0]) + +#ifdef XML_MIN_SIZE + +static int PTRFASTCALL +little2_byteType(const ENCODING *enc, const char *p) +{ + return LITTLE2_BYTE_TYPE(enc, p); +} + +static int PTRFASTCALL +little2_byteToAscii(const ENCODING *enc, const char *p) +{ + return LITTLE2_BYTE_TO_ASCII(enc, p); +} + +static int PTRCALL +little2_charMatches(const ENCODING *enc, const char *p, int c) +{ + return LITTLE2_CHAR_MATCHES(enc, p, c); +} + +static int PTRFASTCALL +little2_isNameMin(const ENCODING *enc, const char *p) +{ + return LITTLE2_IS_NAME_CHAR_MINBPC(enc, p); +} + +static int PTRFASTCALL +little2_isNmstrtMin(const ENCODING *enc, const char *p) +{ + return LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p); +} + +#undef VTABLE +#define VTABLE VTABLE1, little2_toUtf8, little2_toUtf16 + +#else /* not XML_MIN_SIZE */ + +#undef PREFIX +#define PREFIX(ident) little2_ ## ident +#define MINBPC(enc) 2 +/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */ +#define BYTE_TYPE(enc, p) LITTLE2_BYTE_TYPE(enc, p) +#define BYTE_TO_ASCII(enc, p) LITTLE2_BYTE_TO_ASCII(enc, p) +#define CHAR_MATCHES(enc, p, c) LITTLE2_CHAR_MATCHES(enc, p, c) +#define IS_NAME_CHAR(enc, p, n) 0 +#define IS_NAME_CHAR_MINBPC(enc, p) LITTLE2_IS_NAME_CHAR_MINBPC(enc, p) +#define IS_NMSTRT_CHAR(enc, p, n) (0) +#define IS_NMSTRT_CHAR_MINBPC(enc, p) LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p) + +#define XML_TOK_IMPL_C +#include <xml/details/expat/xmltok_impl.c> +#undef XML_TOK_IMPL_C + +#undef MINBPC +#undef BYTE_TYPE +#undef BYTE_TO_ASCII +#undef CHAR_MATCHES +#undef IS_NAME_CHAR +#undef IS_NAME_CHAR_MINBPC +#undef IS_NMSTRT_CHAR +#undef IS_NMSTRT_CHAR_MINBPC +#undef IS_INVALID_CHAR + +#endif /* not XML_MIN_SIZE */ + +#ifdef XML_NS + +static const struct normal_encoding little2_encoding_ns = { + { VTABLE, 2, 0, +#if BYTEORDER == 1234 + 1 +#else + 0 +#endif + }, + { +#include <xml/details/expat/asciitab.h> +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(little2_) ZERO_VTABLE +}; + +#endif + +static const struct normal_encoding little2_encoding = { + { VTABLE, 2, 0, +#if BYTEORDER == 1234 + 1 +#else + 0 +#endif + }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/asciitab.h> +#undef BT_COLON +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(little2_) ZERO_VTABLE +}; + +#if BYTEORDER != 4321 + +#ifdef XML_NS + +static const struct normal_encoding internal_little2_encoding_ns = { + { VTABLE, 2, 0, 1 }, + { +#include <xml/details/expat/iasciitab.h> +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(little2_) ZERO_VTABLE +}; + +#endif + +static const struct normal_encoding internal_little2_encoding = { + { VTABLE, 2, 0, 1 }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/iasciitab.h> +#undef BT_COLON +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(little2_) ZERO_VTABLE +}; + +#endif + + +#define BIG2_BYTE_TYPE(enc, p) \ + ((p)[0] == 0 \ + ? ((struct normal_encoding *)(enc))->type[(unsigned char)(p)[1]] \ + : unicode_byte_type((p)[0], (p)[1])) +#define BIG2_BYTE_TO_ASCII(enc, p) ((p)[0] == 0 ? (p)[1] : -1) +#define BIG2_CHAR_MATCHES(enc, p, c) ((p)[0] == 0 && (p)[1] == c) +#define BIG2_IS_NAME_CHAR_MINBPC(enc, p) \ + UCS2_GET_NAMING(namePages, (unsigned char)p[0], (unsigned char)p[1]) +#define BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p) \ + UCS2_GET_NAMING(nmstrtPages, (unsigned char)p[0], (unsigned char)p[1]) + +#ifdef XML_MIN_SIZE + +static int PTRFASTCALL +big2_byteType(const ENCODING *enc, const char *p) +{ + return BIG2_BYTE_TYPE(enc, p); +} + +static int PTRFASTCALL +big2_byteToAscii(const ENCODING *enc, const char *p) +{ + return BIG2_BYTE_TO_ASCII(enc, p); +} + +static int PTRCALL +big2_charMatches(const ENCODING *enc, const char *p, int c) +{ + return BIG2_CHAR_MATCHES(enc, p, c); +} + +static int PTRFASTCALL +big2_isNameMin(const ENCODING *enc, const char *p) +{ + return BIG2_IS_NAME_CHAR_MINBPC(enc, p); +} + +static int PTRFASTCALL +big2_isNmstrtMin(const ENCODING *enc, const char *p) +{ + return BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p); +} + +#undef VTABLE +#define VTABLE VTABLE1, big2_toUtf8, big2_toUtf16 + +#else /* not XML_MIN_SIZE */ + +#undef PREFIX +#define PREFIX(ident) big2_ ## ident +#define MINBPC(enc) 2 +/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */ +#define BYTE_TYPE(enc, p) BIG2_BYTE_TYPE(enc, p) +#define BYTE_TO_ASCII(enc, p) BIG2_BYTE_TO_ASCII(enc, p) +#define CHAR_MATCHES(enc, p, c) BIG2_CHAR_MATCHES(enc, p, c) +#define IS_NAME_CHAR(enc, p, n) 0 +#define IS_NAME_CHAR_MINBPC(enc, p) BIG2_IS_NAME_CHAR_MINBPC(enc, p) +#define IS_NMSTRT_CHAR(enc, p, n) (0) +#define IS_NMSTRT_CHAR_MINBPC(enc, p) BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p) + +#define XML_TOK_IMPL_C +#include <xml/details/expat/xmltok_impl.c> +#undef XML_TOK_IMPL_C + +#undef MINBPC +#undef BYTE_TYPE +#undef BYTE_TO_ASCII +#undef CHAR_MATCHES +#undef IS_NAME_CHAR +#undef IS_NAME_CHAR_MINBPC +#undef IS_NMSTRT_CHAR +#undef IS_NMSTRT_CHAR_MINBPC +#undef IS_INVALID_CHAR + +#endif /* not XML_MIN_SIZE */ + +#ifdef XML_NS + +static const struct normal_encoding big2_encoding_ns = { + { VTABLE, 2, 0, +#if BYTEORDER == 4321 + 1 +#else + 0 +#endif + }, + { +#include <xml/details/expat/asciitab.h> +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(big2_) ZERO_VTABLE +}; + +#endif + +static const struct normal_encoding big2_encoding = { + { VTABLE, 2, 0, +#if BYTEORDER == 4321 + 1 +#else + 0 +#endif + }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/asciitab.h> +#undef BT_COLON +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(big2_) ZERO_VTABLE +}; + +#if BYTEORDER != 1234 + +#ifdef XML_NS + +static const struct normal_encoding internal_big2_encoding_ns = { + { VTABLE, 2, 0, 1 }, + { +#include <xml/details/expat/iasciitab.h> +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(big2_) ZERO_VTABLE +}; + +#endif + +static const struct normal_encoding internal_big2_encoding = { + { VTABLE, 2, 0, 1 }, + { +#define BT_COLON BT_NMSTRT +#include <xml/details/expat/iasciitab.h> +#undef BT_COLON +#include <xml/details/expat/latin1tab.h> + }, + STANDARD_VTABLE(big2_) ZERO_VTABLE +}; + +#endif + +#undef PREFIX + +static int FASTCALL +streqci(const char *s1, const char *s2) +{ + for (;;) { + char c1 = *s1++; + char c2 = *s2++; + if (ASCII_a <= c1 && c1 <= ASCII_z) + c1 += ASCII_A - ASCII_a; + if (ASCII_a <= c2 && c2 <= ASCII_z) + c2 += ASCII_A - ASCII_a; + if (c1 != c2) + return 0; + if (!c1) + break; + } + return 1; +} + +static void PTRCALL +initUpdatePosition(const ENCODING *enc, const char *ptr, + const char *end, POSITION *pos) +{ + UNUSED(enc); + normal_updatePosition(&utf8_encoding.enc, ptr, end, pos); +} + +static int +toAscii(const ENCODING *enc, const char *ptr, const char *end) +{ + char buf[1]; + char *p = buf; + XmlUtf8Convert(enc, &ptr, end, &p, p + 1); + if (p == buf) + return -1; + else + return buf[0]; +} + +static int FASTCALL +isSpace(int c) +{ + switch (c) { + case 0x20: + case 0xD: + case 0xA: + case 0x9: + return 1; + } + return 0; +} + +/* Return 1 if there's just optional white space or there's an S + followed by name=val. +*/ +static int +parsePseudoAttribute(const ENCODING *enc, + const char *ptr, + const char *end, + const char **namePtr, + const char **nameEndPtr, + const char **valPtr, + const char **nextTokPtr) +{ + int c; + char open; + if (ptr == end) { + *namePtr = NULL; + return 1; + } + if (!isSpace(toAscii(enc, ptr, end))) { + *nextTokPtr = ptr; + return 0; + } + do { + ptr += enc->minBytesPerChar; + } while (isSpace(toAscii(enc, ptr, end))); + if (ptr == end) { + *namePtr = NULL; + return 1; + } + *namePtr = ptr; + for (;;) { + c = toAscii(enc, ptr, end); + if (c == -1) { + *nextTokPtr = ptr; + return 0; + } + if (c == ASCII_EQUALS) { + *nameEndPtr = ptr; + break; + } + if (isSpace(c)) { + *nameEndPtr = ptr; + do { + ptr += enc->minBytesPerChar; + } while (isSpace(c = toAscii(enc, ptr, end))); + if (c != ASCII_EQUALS) { + *nextTokPtr = ptr; + return 0; + } + break; + } + ptr += enc->minBytesPerChar; + } + if (ptr == *namePtr) { + *nextTokPtr = ptr; + return 0; + } + ptr += enc->minBytesPerChar; + c = toAscii(enc, ptr, end); + while (isSpace(c)) { + ptr += enc->minBytesPerChar; + c = toAscii(enc, ptr, end); + } + if (c != ASCII_QUOT && c != ASCII_APOS) { + *nextTokPtr = ptr; + return 0; + } + open = (char)c; + ptr += enc->minBytesPerChar; + *valPtr = ptr; + for (;; ptr += enc->minBytesPerChar) { + c = toAscii(enc, ptr, end); + if (c == open) + break; + if (!(ASCII_a <= c && c <= ASCII_z) + && !(ASCII_A <= c && c <= ASCII_Z) + && !(ASCII_0 <= c && c <= ASCII_9) + && c != ASCII_PERIOD + && c != ASCII_MINUS + && c != ASCII_UNDERSCORE) { + *nextTokPtr = ptr; + return 0; + } + } + *nextTokPtr = ptr + enc->minBytesPerChar; + return 1; +} + +static const char KW_version[] = { + ASCII_v, ASCII_e, ASCII_r, ASCII_s, ASCII_i, ASCII_o, ASCII_n, '\0' +}; + +static const char KW_encoding[] = { + ASCII_e, ASCII_n, ASCII_c, ASCII_o, ASCII_d, ASCII_i, ASCII_n, ASCII_g, '\0' +}; + +static const char KW_standalone[] = { + ASCII_s, ASCII_t, ASCII_a, ASCII_n, ASCII_d, ASCII_a, ASCII_l, ASCII_o, + ASCII_n, ASCII_e, '\0' +}; + +static const char KW_yes[] = { + ASCII_y, ASCII_e, ASCII_s, '\0' +}; + +static const char KW_no[] = { + ASCII_n, ASCII_o, '\0' +}; + +static int +doParseXmlDecl(const ENCODING *(*encodingFinder)(const ENCODING *, + const char *, + const char *), + int isGeneralTextEntity, + const ENCODING *enc, + const char *ptr, + const char *end, + const char **badPtr, + const char **versionPtr, + const char **versionEndPtr, + const char **encodingName, + const ENCODING **encoding, + int *standalone) +{ + const char *val = NULL; + const char *name = NULL; + const char *nameEnd = NULL; + ptr += 5 * enc->minBytesPerChar; + end -= 2 * enc->minBytesPerChar; + if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr) + || !name) { + *badPtr = ptr; + return 0; + } + if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_version)) { + if (!isGeneralTextEntity) { + *badPtr = name; + return 0; + } + } + else { + if (versionPtr) + *versionPtr = val; + if (versionEndPtr) + *versionEndPtr = ptr; + if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)) { + *badPtr = ptr; + return 0; + } + if (!name) { + if (isGeneralTextEntity) { + /* a TextDecl must have an EncodingDecl */ + *badPtr = ptr; + return 0; + } + return 1; + } + } + if (XmlNameMatchesAscii(enc, name, nameEnd, KW_encoding)) { + int c = toAscii(enc, val, end); + if (!(ASCII_a <= c && c <= ASCII_z) && !(ASCII_A <= c && c <= ASCII_Z)) { + *badPtr = val; + return 0; + } + if (encodingName) + *encodingName = val; + if (encoding) + *encoding = encodingFinder(enc, val, ptr - enc->minBytesPerChar); + if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)) { + *badPtr = ptr; + return 0; + } + if (!name) + return 1; + } + if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_standalone) + || isGeneralTextEntity) { + *badPtr = name; + return 0; + } + if (XmlNameMatchesAscii(enc, val, ptr - enc->minBytesPerChar, KW_yes)) { + if (standalone) + *standalone = 1; + } + else if (XmlNameMatchesAscii(enc, val, ptr - enc->minBytesPerChar, KW_no)) { + if (standalone) + *standalone = 0; + } + else { + *badPtr = val; + return 0; + } + while (isSpace(toAscii(enc, ptr, end))) + ptr += enc->minBytesPerChar; + if (ptr != end) { + *badPtr = ptr; + return 0; + } + return 1; +} + +static int FASTCALL +checkCharRefNumber(int result) +{ + switch (result >> 8) { + case 0xD8: case 0xD9: case 0xDA: case 0xDB: + case 0xDC: case 0xDD: case 0xDE: case 0xDF: + return -1; + case 0: + if (latin1_encoding.type[result] == BT_NONXML) + return -1; + break; + case 0xFF: + if (result == 0xFFFE || result == 0xFFFF) + return -1; + break; + } + return result; +} + +int FASTCALL +XmlUtf8Encode(int c, char *buf) +{ + enum { + /* minN is minimum legal resulting value for N byte sequence */ + min2 = 0x80, + min3 = 0x800, + min4 = 0x10000 + }; + + if (c < 0) + return 0; + if (c < min2) { + buf[0] = (char)(c | UTF8_cval1); + return 1; + } + if (c < min3) { + buf[0] = (char)((c >> 6) | UTF8_cval2); + buf[1] = (char)((c & 0x3f) | 0x80); + return 2; + } + if (c < min4) { + buf[0] = (char)((c >> 12) | UTF8_cval3); + buf[1] = (char)(((c >> 6) & 0x3f) | 0x80); + buf[2] = (char)((c & 0x3f) | 0x80); + return 3; + } + if (c < 0x110000) { + buf[0] = (char)((c >> 18) | UTF8_cval4); + buf[1] = (char)(((c >> 12) & 0x3f) | 0x80); + buf[2] = (char)(((c >> 6) & 0x3f) | 0x80); + buf[3] = (char)((c & 0x3f) | 0x80); + return 4; + } + return 0; +} + +int FASTCALL +XmlUtf16Encode(int charNum, unsigned short *buf) +{ + if (charNum < 0) + return 0; + if (charNum < 0x10000) { + buf[0] = (unsigned short)charNum; + return 1; + } + if (charNum < 0x110000) { + charNum -= 0x10000; + buf[0] = (unsigned short)((charNum >> 10) + 0xD800); + buf[1] = (unsigned short)((charNum & 0x3FF) + 0xDC00); + return 2; + } + return 0; +} + +struct unknown_encoding { + struct normal_encoding normal; + CONVERTER convert; + void *userData; + unsigned short utf16[256]; + char utf8[256][4]; +}; + +#define AS_UNKNOWN_ENCODING(enc) ((const struct unknown_encoding *) (enc)) + +int +XmlSizeOfUnknownEncoding(void) +{ + return sizeof(struct unknown_encoding); +} + +static int PTRFASTCALL +unknown_isName(const ENCODING *enc, const char *p) +{ + const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc); + int c = uenc->convert(uenc->userData, p); + if (c & ~0xFFFF) + return 0; + return UCS2_GET_NAMING(namePages, c >> 8, c & 0xFF); +} + +static int PTRFASTCALL +unknown_isNmstrt(const ENCODING *enc, const char *p) +{ + const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc); + int c = uenc->convert(uenc->userData, p); + if (c & ~0xFFFF) + return 0; + return UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xFF); +} + +static int PTRFASTCALL +unknown_isInvalid(const ENCODING *enc, const char *p) +{ + const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc); + int c = uenc->convert(uenc->userData, p); + return (c & ~0xFFFF) || checkCharRefNumber(c) < 0; +} + +static void PTRCALL +unknown_toUtf8(const ENCODING *enc, + const char **fromP, const char *fromLim, + char **toP, const char *toLim) +{ + const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc); + char buf[XML_UTF8_ENCODE_MAX]; + for (;;) { + const char *utf8; + int n; + if (*fromP == fromLim) + break; + utf8 = uenc->utf8[(unsigned char)**fromP]; + n = *utf8++; + if (n == 0) { + int c = uenc->convert(uenc->userData, *fromP); + n = XmlUtf8Encode(c, buf); + if (n > toLim - *toP) + break; + utf8 = buf; + *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP] + - (BT_LEAD2 - 2)); + } + else { + if (n > toLim - *toP) + break; + (*fromP)++; + } + do { + *(*toP)++ = *utf8++; + } while (--n != 0); + } +} + +static void PTRCALL +unknown_toUtf16(const ENCODING *enc, + const char **fromP, const char *fromLim, + unsigned short **toP, const unsigned short *toLim) +{ + const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc); + while (*fromP != fromLim && *toP != toLim) { + unsigned short c = uenc->utf16[(unsigned char)**fromP]; + if (c == 0) { + c = (unsigned short) + uenc->convert(uenc->userData, *fromP); + *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP] + - (BT_LEAD2 - 2)); + } + else + (*fromP)++; + *(*toP)++ = c; + } +} + +ENCODING * +XmlInitUnknownEncoding(void *mem, + int *table, + CONVERTER convert, + void *userData) +{ + int i; + struct unknown_encoding *e = (struct unknown_encoding *)mem; + for (i = 0; i < (int)sizeof(struct normal_encoding); i++) + ((char *)mem)[i] = ((char *)&latin1_encoding)[i]; + for (i = 0; i < 128; i++) + if (latin1_encoding.type[i] != BT_OTHER + && latin1_encoding.type[i] != BT_NONXML + && table[i] != i) + return 0; + for (i = 0; i < 256; i++) { + int c = table[i]; + if (c == -1) { + e->normal.type[i] = BT_MALFORM; + /* This shouldn't really get used. */ + e->utf16[i] = 0xFFFF; + e->utf8[i][0] = 1; + e->utf8[i][1] = 0; + } + else if (c < 0) { + if (c < -4) + return 0; + e->normal.type[i] = (unsigned char)(BT_LEAD2 - (c + 2)); + e->utf8[i][0] = 0; + e->utf16[i] = 0; + } + else if (c < 0x80) { + if (latin1_encoding.type[c] != BT_OTHER + && latin1_encoding.type[c] != BT_NONXML + && c != i) + return 0; + e->normal.type[i] = latin1_encoding.type[c]; + e->utf8[i][0] = 1; + e->utf8[i][1] = (char)c; + e->utf16[i] = (unsigned short)(c == 0 ? 0xFFFF : c); + } + else if (checkCharRefNumber(c) < 0) { + e->normal.type[i] = BT_NONXML; + /* This shouldn't really get used. */ + e->utf16[i] = 0xFFFF; + e->utf8[i][0] = 1; + e->utf8[i][1] = 0; + } + else { + if (c > 0xFFFF) + return 0; + if (UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xff)) + e->normal.type[i] = BT_NMSTRT; + else if (UCS2_GET_NAMING(namePages, c >> 8, c & 0xff)) + e->normal.type[i] = BT_NAME; + else + e->normal.type[i] = BT_OTHER; + e->utf8[i][0] = (char)XmlUtf8Encode(c, e->utf8[i] + 1); + e->utf16[i] = (unsigned short)c; + } + } + e->userData = userData; + e->convert = convert; + if (convert) { + e->normal.isName2 = unknown_isName; + e->normal.isName3 = unknown_isName; + e->normal.isName4 = unknown_isName; + e->normal.isNmstrt2 = unknown_isNmstrt; + e->normal.isNmstrt3 = unknown_isNmstrt; + e->normal.isNmstrt4 = unknown_isNmstrt; + e->normal.isInvalid2 = unknown_isInvalid; + e->normal.isInvalid3 = unknown_isInvalid; + e->normal.isInvalid4 = unknown_isInvalid; + } + e->normal.enc.utf8Convert = unknown_toUtf8; + e->normal.enc.utf16Convert = unknown_toUtf16; + return &(e->normal.enc); +} + +/* If this enumeration is changed, getEncodingIndex and encodings +must also be changed. */ +enum { + UNKNOWN_ENC = -1, + ISO_8859_1_ENC = 0, + US_ASCII_ENC, + UTF_8_ENC, + UTF_16_ENC, + UTF_16BE_ENC, + UTF_16LE_ENC, + /* must match encodingNames up to here */ + NO_ENC +}; + +static const char KW_ISO_8859_1[] = { + ASCII_I, ASCII_S, ASCII_O, ASCII_MINUS, ASCII_8, ASCII_8, ASCII_5, ASCII_9, + ASCII_MINUS, ASCII_1, '\0' +}; +static const char KW_US_ASCII[] = { + ASCII_U, ASCII_S, ASCII_MINUS, ASCII_A, ASCII_S, ASCII_C, ASCII_I, ASCII_I, + '\0' +}; +static const char KW_UTF_8[] = { + ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_8, '\0' +}; +static const char KW_UTF_16[] = { + ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, '\0' +}; +static const char KW_UTF_16BE[] = { + ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_B, ASCII_E, + '\0' +}; +static const char KW_UTF_16LE[] = { + ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_L, ASCII_E, + '\0' +}; + +static int FASTCALL +getEncodingIndex(const char *name) +{ + static const char * const encodingNames[] = { + KW_ISO_8859_1, + KW_US_ASCII, + KW_UTF_8, + KW_UTF_16, + KW_UTF_16BE, + KW_UTF_16LE, + }; + int i; + if (name == NULL) + return NO_ENC; + for (i = 0; i < (int)(sizeof(encodingNames)/sizeof(encodingNames[0])); i++) + if (streqci(name, encodingNames[i])) + return i; + return UNKNOWN_ENC; +} + +/* For binary compatibility, we store the index of the encoding + specified at initialization in the isUtf16 member. +*/ + +#define INIT_ENC_INDEX(enc) ((int)(enc)->initEnc.isUtf16) +#define SET_INIT_ENC_INDEX(enc, i) ((enc)->initEnc.isUtf16 = (char)i) + +/* This is what detects the encoding. encodingTable maps from + encoding indices to encodings; INIT_ENC_INDEX(enc) is the index of + the external (protocol) specified encoding; state is + XML_CONTENT_STATE if we're parsing an external text entity, and + XML_PROLOG_STATE otherwise. +*/ + + +static int +initScan(const ENCODING * const *encodingTable, + const INIT_ENCODING *enc, + int state, + const char *ptr, + const char *end, + const char **nextTokPtr) +{ + const ENCODING **encPtr; + + if (ptr == end) + return XML_TOK_NONE; + encPtr = enc->encPtr; + if (ptr + 1 == end) { + /* only a single byte available for auto-detection */ +#ifndef XML_DTD /* FIXME */ + /* a well-formed document entity must have more than one byte */ + if (state != XML_CONTENT_STATE) + return XML_TOK_PARTIAL; +#endif + /* so we're parsing an external text entity... */ + /* if UTF-16 was externally specified, then we need at least 2 bytes */ + switch (INIT_ENC_INDEX(enc)) { + case UTF_16_ENC: + case UTF_16LE_ENC: + case UTF_16BE_ENC: + return XML_TOK_PARTIAL; + } + switch ((unsigned char)*ptr) { + case 0xFE: + case 0xFF: + case 0xEF: /* possibly first byte of UTF-8 BOM */ + if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC + && state == XML_CONTENT_STATE) + break; + /* fall through */ + case 0x00: + case 0x3C: + return XML_TOK_PARTIAL; + } + } + else { + switch (((unsigned char)ptr[0] << 8) | (unsigned char)ptr[1]) { + case 0xFEFF: + if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC + && state == XML_CONTENT_STATE) + break; + *nextTokPtr = ptr + 2; + *encPtr = encodingTable[UTF_16BE_ENC]; + return XML_TOK_BOM; + /* 00 3C is handled in the default case */ + case 0x3C00: + if ((INIT_ENC_INDEX(enc) == UTF_16BE_ENC + || INIT_ENC_INDEX(enc) == UTF_16_ENC) + && state == XML_CONTENT_STATE) + break; + *encPtr = encodingTable[UTF_16LE_ENC]; + return XmlTok(*encPtr, state, ptr, end, nextTokPtr); + case 0xFFFE: + if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC + && state == XML_CONTENT_STATE) + break; + *nextTokPtr = ptr + 2; + *encPtr = encodingTable[UTF_16LE_ENC]; + return XML_TOK_BOM; + case 0xEFBB: + /* Maybe a UTF-8 BOM (EF BB BF) */ + /* If there's an explicitly specified (external) encoding + of ISO-8859-1 or some flavour of UTF-16 + and this is an external text entity, + don't look for the BOM, + because it might be a legal data. + */ + if (state == XML_CONTENT_STATE) { + int e = INIT_ENC_INDEX(enc); + if (e == ISO_8859_1_ENC || e == UTF_16BE_ENC + || e == UTF_16LE_ENC || e == UTF_16_ENC) + break; + } + if (ptr + 2 == end) + return XML_TOK_PARTIAL; + if ((unsigned char)ptr[2] == 0xBF) { + *nextTokPtr = ptr + 3; + *encPtr = encodingTable[UTF_8_ENC]; + return XML_TOK_BOM; + } + break; + default: + if (ptr[0] == '\0') { + /* 0 isn't a legal data character. Furthermore a document + entity can only start with ASCII characters. So the only + way this can fail to be big-endian UTF-16 if it it's an + external parsed general entity that's labelled as + UTF-16LE. + */ + if (state == XML_CONTENT_STATE && INIT_ENC_INDEX(enc) == UTF_16LE_ENC) + break; + *encPtr = encodingTable[UTF_16BE_ENC]; + return XmlTok(*encPtr, state, ptr, end, nextTokPtr); + } + else if (ptr[1] == '\0') { + /* We could recover here in the case: + - parsing an external entity + - second byte is 0 + - no externally specified encoding + - no encoding declaration + by assuming UTF-16LE. But we don't, because this would mean when + presented just with a single byte, we couldn't reliably determine + whether we needed further bytes. + */ + if (state == XML_CONTENT_STATE) + break; + *encPtr = encodingTable[UTF_16LE_ENC]; + return XmlTok(*encPtr, state, ptr, end, nextTokPtr); + } + break; + } + } + *encPtr = encodingTable[INIT_ENC_INDEX(enc)]; + return XmlTok(*encPtr, state, ptr, end, nextTokPtr); +} + + +#define NS(x) x +#define ns(x) x +#define XML_TOK_NS_C +#include <xml/details/expat/xmltok_ns.c> +#undef XML_TOK_NS_C +#undef NS +#undef ns + +#ifdef XML_NS + +#define NS(x) x ## NS +#define ns(x) x ## _ns + +#define XML_TOK_NS_C +#include <xml/details/expat/xmltok_ns.c> +#undef XML_TOK_NS_C + +#undef NS +#undef ns + +ENCODING * +XmlInitUnknownEncodingNS(void *mem, + int *table, + CONVERTER convert, + void *userData) +{ + ENCODING *enc = XmlInitUnknownEncoding(mem, table, convert, userData); + if (enc) + ((struct normal_encoding *)enc)->type[ASCII_COLON] = BT_COLON; + return enc; +} + +#endif /* XML_NS */ diff --git a/xml/details/expat/xmltok.h b/xml/details/expat/xmltok.h new file mode 100644 index 0000000..ca867aa --- /dev/null +++ b/xml/details/expat/xmltok.h @@ -0,0 +1,316 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +#ifndef XmlTok_INCLUDED +#define XmlTok_INCLUDED 1 + +#ifdef __cplusplus +extern "C" { +#endif + +/* The following token may be returned by XmlContentTok */ +#define XML_TOK_TRAILING_RSQB -5 /* ] or ]] at the end of the scan; might be + start of illegal ]]> sequence */ +/* The following tokens may be returned by both XmlPrologTok and + XmlContentTok. +*/ +#define XML_TOK_NONE -4 /* The string to be scanned is empty */ +#define XML_TOK_TRAILING_CR -3 /* A CR at the end of the scan; + might be part of CRLF sequence */ +#define XML_TOK_PARTIAL_CHAR -2 /* only part of a multibyte sequence */ +#define XML_TOK_PARTIAL -1 /* only part of a token */ +#define XML_TOK_INVALID 0 + +/* The following tokens are returned by XmlContentTok; some are also + returned by XmlAttributeValueTok, XmlEntityTok, XmlCdataSectionTok. +*/ +#define XML_TOK_START_TAG_WITH_ATTS 1 +#define XML_TOK_START_TAG_NO_ATTS 2 +#define XML_TOK_EMPTY_ELEMENT_WITH_ATTS 3 /* empty element tag <e/> */ +#define XML_TOK_EMPTY_ELEMENT_NO_ATTS 4 +#define XML_TOK_END_TAG 5 +#define XML_TOK_DATA_CHARS 6 +#define XML_TOK_DATA_NEWLINE 7 +#define XML_TOK_CDATA_SECT_OPEN 8 +#define XML_TOK_ENTITY_REF 9 +#define XML_TOK_CHAR_REF 10 /* numeric character reference */ + +/* The following tokens may be returned by both XmlPrologTok and + XmlContentTok. +*/ +#define XML_TOK_PI 11 /* processing instruction */ +#define XML_TOK_XML_DECL 12 /* XML decl or text decl */ +#define XML_TOK_COMMENT 13 +#define XML_TOK_BOM 14 /* Byte order mark */ + +/* The following tokens are returned only by XmlPrologTok */ +#define XML_TOK_PROLOG_S 15 +#define XML_TOK_DECL_OPEN 16 /* <!foo */ +#define XML_TOK_DECL_CLOSE 17 /* > */ +#define XML_TOK_NAME 18 +#define XML_TOK_NMTOKEN 19 +#define XML_TOK_POUND_NAME 20 /* #name */ +#define XML_TOK_OR 21 /* | */ +#define XML_TOK_PERCENT 22 +#define XML_TOK_OPEN_PAREN 23 +#define XML_TOK_CLOSE_PAREN 24 +#define XML_TOK_OPEN_BRACKET 25 +#define XML_TOK_CLOSE_BRACKET 26 +#define XML_TOK_LITERAL 27 +#define XML_TOK_PARAM_ENTITY_REF 28 +#define XML_TOK_INSTANCE_START 29 + +/* The following occur only in element type declarations */ +#define XML_TOK_NAME_QUESTION 30 /* name? */ +#define XML_TOK_NAME_ASTERISK 31 /* name* */ +#define XML_TOK_NAME_PLUS 32 /* name+ */ +#define XML_TOK_COND_SECT_OPEN 33 /* <![ */ +#define XML_TOK_COND_SECT_CLOSE 34 /* ]]> */ +#define XML_TOK_CLOSE_PAREN_QUESTION 35 /* )? */ +#define XML_TOK_CLOSE_PAREN_ASTERISK 36 /* )* */ +#define XML_TOK_CLOSE_PAREN_PLUS 37 /* )+ */ +#define XML_TOK_COMMA 38 + +/* The following token is returned only by XmlAttributeValueTok */ +#define XML_TOK_ATTRIBUTE_VALUE_S 39 + +/* The following token is returned only by XmlCdataSectionTok */ +#define XML_TOK_CDATA_SECT_CLOSE 40 + +/* With namespace processing this is returned by XmlPrologTok for a + name with a colon. +*/ +#define XML_TOK_PREFIXED_NAME 41 + +#ifdef XML_DTD +#define XML_TOK_IGNORE_SECT 42 +#endif /* XML_DTD */ + +#ifdef XML_DTD +#define XML_N_STATES 4 +#else /* not XML_DTD */ +#define XML_N_STATES 3 +#endif /* not XML_DTD */ + +#define XML_PROLOG_STATE 0 +#define XML_CONTENT_STATE 1 +#define XML_CDATA_SECTION_STATE 2 +#ifdef XML_DTD +#define XML_IGNORE_SECTION_STATE 3 +#endif /* XML_DTD */ + +#define XML_N_LITERAL_TYPES 2 +#define XML_ATTRIBUTE_VALUE_LITERAL 0 +#define XML_ENTITY_VALUE_LITERAL 1 + +/* The size of the buffer passed to XmlUtf8Encode must be at least this. */ +#define XML_UTF8_ENCODE_MAX 4 +/* The size of the buffer passed to XmlUtf16Encode must be at least this. */ +#define XML_UTF16_ENCODE_MAX 2 + +typedef struct position { + /* first line and first column are 0 not 1 */ + XML_Size lineNumber; + XML_Size columnNumber; +} POSITION; + +typedef struct { + const char *name; + const char *valuePtr; + const char *valueEnd; + char normalized; +} ATTRIBUTE; + +struct encoding; +typedef struct encoding ENCODING; + +typedef int (PTRCALL *SCANNER)(const ENCODING *, + const char *, + const char *, + const char **); + +struct encoding { + SCANNER scanners[XML_N_STATES]; + SCANNER literalScanners[XML_N_LITERAL_TYPES]; + int (PTRCALL *sameName)(const ENCODING *, + const char *, + const char *); + int (PTRCALL *nameMatchesAscii)(const ENCODING *, + const char *, + const char *, + const char *); + int (PTRFASTCALL *nameLength)(const ENCODING *, const char *); + const char *(PTRFASTCALL *skipS)(const ENCODING *, const char *); + int (PTRCALL *getAtts)(const ENCODING *enc, + const char *ptr, + int attsMax, + ATTRIBUTE *atts); + int (PTRFASTCALL *charRefNumber)(const ENCODING *enc, const char *ptr); + int (PTRCALL *predefinedEntityName)(const ENCODING *, + const char *, + const char *); + void (PTRCALL *updatePosition)(const ENCODING *, + const char *ptr, + const char *end, + POSITION *); + int (PTRCALL *isPublicId)(const ENCODING *enc, + const char *ptr, + const char *end, + const char **badPtr); + void (PTRCALL *utf8Convert)(const ENCODING *enc, + const char **fromP, + const char *fromLim, + char **toP, + const char *toLim); + void (PTRCALL *utf16Convert)(const ENCODING *enc, + const char **fromP, + const char *fromLim, + unsigned short **toP, + const unsigned short *toLim); + int minBytesPerChar; + char isUtf8; + char isUtf16; +}; + +/* Scan the string starting at ptr until the end of the next complete + token, but do not scan past eptr. Return an integer giving the + type of token. + + Return XML_TOK_NONE when ptr == eptr; nextTokPtr will not be set. + + Return XML_TOK_PARTIAL when the string does not contain a complete + token; nextTokPtr will not be set. + + Return XML_TOK_INVALID when the string does not start a valid + token; nextTokPtr will be set to point to the character which made + the token invalid. + + Otherwise the string starts with a valid token; nextTokPtr will be + set to point to the character following the end of that token. + + Each data character counts as a single token, but adjacent data + characters may be returned together. Similarly for characters in + the prolog outside literals, comments and processing instructions. +*/ + + +#define XmlTok(enc, state, ptr, end, nextTokPtr) \ + (((enc)->scanners[state])(enc, ptr, end, nextTokPtr)) + +#define XmlPrologTok(enc, ptr, end, nextTokPtr) \ + XmlTok(enc, XML_PROLOG_STATE, ptr, end, nextTokPtr) + +#define XmlContentTok(enc, ptr, end, nextTokPtr) \ + XmlTok(enc, XML_CONTENT_STATE, ptr, end, nextTokPtr) + +#define XmlCdataSectionTok(enc, ptr, end, nextTokPtr) \ + XmlTok(enc, XML_CDATA_SECTION_STATE, ptr, end, nextTokPtr) + +#ifdef XML_DTD + +#define XmlIgnoreSectionTok(enc, ptr, end, nextTokPtr) \ + XmlTok(enc, XML_IGNORE_SECTION_STATE, ptr, end, nextTokPtr) + +#endif /* XML_DTD */ + +/* This is used for performing a 2nd-level tokenization on the content + of a literal that has already been returned by XmlTok. +*/ +#define XmlLiteralTok(enc, literalType, ptr, end, nextTokPtr) \ + (((enc)->literalScanners[literalType])(enc, ptr, end, nextTokPtr)) + +#define XmlAttributeValueTok(enc, ptr, end, nextTokPtr) \ + XmlLiteralTok(enc, XML_ATTRIBUTE_VALUE_LITERAL, ptr, end, nextTokPtr) + +#define XmlEntityValueTok(enc, ptr, end, nextTokPtr) \ + XmlLiteralTok(enc, XML_ENTITY_VALUE_LITERAL, ptr, end, nextTokPtr) + +#define XmlSameName(enc, ptr1, ptr2) (((enc)->sameName)(enc, ptr1, ptr2)) + +#define XmlNameMatchesAscii(enc, ptr1, end1, ptr2) \ + (((enc)->nameMatchesAscii)(enc, ptr1, end1, ptr2)) + +#define XmlNameLength(enc, ptr) \ + (((enc)->nameLength)(enc, ptr)) + +#define XmlSkipS(enc, ptr) \ + (((enc)->skipS)(enc, ptr)) + +#define XmlGetAttributes(enc, ptr, attsMax, atts) \ + (((enc)->getAtts)(enc, ptr, attsMax, atts)) + +#define XmlCharRefNumber(enc, ptr) \ + (((enc)->charRefNumber)(enc, ptr)) + +#define XmlPredefinedEntityName(enc, ptr, end) \ + (((enc)->predefinedEntityName)(enc, ptr, end)) + +#define XmlUpdatePosition(enc, ptr, end, pos) \ + (((enc)->updatePosition)(enc, ptr, end, pos)) + +#define XmlIsPublicId(enc, ptr, end, badPtr) \ + (((enc)->isPublicId)(enc, ptr, end, badPtr)) + +#define XmlUtf8Convert(enc, fromP, fromLim, toP, toLim) \ + (((enc)->utf8Convert)(enc, fromP, fromLim, toP, toLim)) + +#define XmlUtf16Convert(enc, fromP, fromLim, toP, toLim) \ + (((enc)->utf16Convert)(enc, fromP, fromLim, toP, toLim)) + +typedef struct { + ENCODING initEnc; + const ENCODING **encPtr; +} INIT_ENCODING; + +int XmlParseXmlDecl(int isGeneralTextEntity, + const ENCODING *enc, + const char *ptr, + const char *end, + const char **badPtr, + const char **versionPtr, + const char **versionEndPtr, + const char **encodingNamePtr, + const ENCODING **namedEncodingPtr, + int *standalonePtr); + +int XmlInitEncoding(INIT_ENCODING *, const ENCODING **, const char *name); +const ENCODING *XmlGetUtf8InternalEncoding(void); +const ENCODING *XmlGetUtf16InternalEncoding(void); +int FASTCALL XmlUtf8Encode(int charNumber, char *buf); +int FASTCALL XmlUtf16Encode(int charNumber, unsigned short *buf); +int XmlSizeOfUnknownEncoding(void); + + +typedef int (XMLCALL *CONVERTER) (void *userData, const char *p); + +ENCODING * +XmlInitUnknownEncoding(void *mem, + int *table, + CONVERTER convert, + void *userData); + +int XmlParseXmlDeclNS(int isGeneralTextEntity, + const ENCODING *enc, + const char *ptr, + const char *end, + const char **badPtr, + const char **versionPtr, + const char **versionEndPtr, + const char **encodingNamePtr, + const ENCODING **namedEncodingPtr, + int *standalonePtr); + +int XmlInitEncodingNS(INIT_ENCODING *, const ENCODING **, const char *name); +const ENCODING *XmlGetUtf8InternalEncodingNS(void); +const ENCODING *XmlGetUtf16InternalEncodingNS(void); +ENCODING * +XmlInitUnknownEncodingNS(void *mem, + int *table, + CONVERTER convert, + void *userData); +#ifdef __cplusplus +} +#endif + +#endif /* not XmlTok_INCLUDED */ diff --git a/xml/details/expat/xmltok_impl.c b/xml/details/expat/xmltok_impl.c new file mode 100644 index 0000000..de4d21a --- /dev/null +++ b/xml/details/expat/xmltok_impl.c @@ -0,0 +1,1796 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +/* This file is included! */ +#ifdef XML_TOK_IMPL_C + +#ifndef IS_INVALID_CHAR +#define IS_INVALID_CHAR(enc, ptr, n) (0) +#endif + +#define INVALID_LEAD_CASE(n, ptr, nextTokPtr) \ + case BT_LEAD ## n: \ + if (end - ptr < n) \ + return XML_TOK_PARTIAL_CHAR; \ + if (IS_INVALID_CHAR(enc, ptr, n)) { \ + *(nextTokPtr) = (ptr); \ + return XML_TOK_INVALID; \ + } \ + ptr += n; \ + break; + +#define INVALID_CASES(ptr, nextTokPtr) \ + INVALID_LEAD_CASE(2, ptr, nextTokPtr) \ + INVALID_LEAD_CASE(3, ptr, nextTokPtr) \ + INVALID_LEAD_CASE(4, ptr, nextTokPtr) \ + case BT_NONXML: \ + case BT_MALFORM: \ + case BT_TRAIL: \ + *(nextTokPtr) = (ptr); \ + return XML_TOK_INVALID; + +#define CHECK_NAME_CASE(n, enc, ptr, end, nextTokPtr) \ + case BT_LEAD ## n: \ + if (end - ptr < n) \ + return XML_TOK_PARTIAL_CHAR; \ + if (!IS_NAME_CHAR(enc, ptr, n)) { \ + *nextTokPtr = ptr; \ + return XML_TOK_INVALID; \ + } \ + ptr += n; \ + break; + +#define CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) \ + case BT_NONASCII: \ + if (!IS_NAME_CHAR_MINBPC(enc, ptr)) { \ + *nextTokPtr = ptr; \ + return XML_TOK_INVALID; \ + } \ + case BT_NMSTRT: \ + case BT_HEX: \ + case BT_DIGIT: \ + case BT_NAME: \ + case BT_MINUS: \ + ptr += MINBPC(enc); \ + break; \ + CHECK_NAME_CASE(2, enc, ptr, end, nextTokPtr) \ + CHECK_NAME_CASE(3, enc, ptr, end, nextTokPtr) \ + CHECK_NAME_CASE(4, enc, ptr, end, nextTokPtr) + +#define CHECK_NMSTRT_CASE(n, enc, ptr, end, nextTokPtr) \ + case BT_LEAD ## n: \ + if (end - ptr < n) \ + return XML_TOK_PARTIAL_CHAR; \ + if (!IS_NMSTRT_CHAR(enc, ptr, n)) { \ + *nextTokPtr = ptr; \ + return XML_TOK_INVALID; \ + } \ + ptr += n; \ + break; + +#define CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) \ + case BT_NONASCII: \ + if (!IS_NMSTRT_CHAR_MINBPC(enc, ptr)) { \ + *nextTokPtr = ptr; \ + return XML_TOK_INVALID; \ + } \ + case BT_NMSTRT: \ + case BT_HEX: \ + ptr += MINBPC(enc); \ + break; \ + CHECK_NMSTRT_CASE(2, enc, ptr, end, nextTokPtr) \ + CHECK_NMSTRT_CASE(3, enc, ptr, end, nextTokPtr) \ + CHECK_NMSTRT_CASE(4, enc, ptr, end, nextTokPtr) + +#ifndef PREFIX +#define PREFIX(ident) ident +#endif + +/* ptr points to character following "<!-" */ + +static int PTRCALL +PREFIX(scanComment)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + if (ptr != end) { + if (!CHAR_MATCHES(enc, ptr, ASCII_MINUS)) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + ptr += MINBPC(enc); + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + INVALID_CASES(ptr, nextTokPtr) + case BT_MINUS: + if ((ptr += MINBPC(enc)) == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr, ASCII_MINUS)) { + if ((ptr += MINBPC(enc)) == end) + return XML_TOK_PARTIAL; + if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_COMMENT; + } + break; + default: + ptr += MINBPC(enc); + break; + } + } + } + return XML_TOK_PARTIAL; +} + +/* ptr points to character following "<!" */ + +static int PTRCALL +PREFIX(scanDecl)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + case BT_MINUS: + return PREFIX(scanComment)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_LSQB: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_COND_SECT_OPEN; + case BT_NMSTRT: + case BT_HEX: + ptr += MINBPC(enc); + break; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + case BT_PERCNT: + if (ptr + MINBPC(enc) == end) + return XML_TOK_PARTIAL; + /* don't allow <!ENTITY% foo "whatever"> */ + switch (BYTE_TYPE(enc, ptr + MINBPC(enc))) { + case BT_S: case BT_CR: case BT_LF: case BT_PERCNT: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + /* fall through */ + case BT_S: case BT_CR: case BT_LF: + *nextTokPtr = ptr; + return XML_TOK_DECL_OPEN; + case BT_NMSTRT: + case BT_HEX: + ptr += MINBPC(enc); + break; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +} + +static int PTRCALL +PREFIX(checkPiTarget)(const ENCODING *enc, const char *ptr, + const char *end, int *tokPtr) +{ + int upper = 0; + + UNUSED(enc); + + *tokPtr = XML_TOK_PI; + if (end - ptr != MINBPC(enc)*3) + return 1; + switch (BYTE_TO_ASCII(enc, ptr)) { + case ASCII_x: + break; + case ASCII_X: + upper = 1; + break; + default: + return 1; + } + ptr += MINBPC(enc); + switch (BYTE_TO_ASCII(enc, ptr)) { + case ASCII_m: + break; + case ASCII_M: + upper = 1; + break; + default: + return 1; + } + ptr += MINBPC(enc); + switch (BYTE_TO_ASCII(enc, ptr)) { + case ASCII_l: + break; + case ASCII_L: + upper = 1; + break; + default: + return 1; + } + if (upper) + return 0; + *tokPtr = XML_TOK_XML_DECL; + return 1; +} + +/* ptr points to character following "<?" */ + +static int PTRCALL +PREFIX(scanPi)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + int tok; + const char *target = ptr; + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) + case BT_S: case BT_CR: case BT_LF: + if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + ptr += MINBPC(enc); + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + INVALID_CASES(ptr, nextTokPtr) + case BT_QUEST: + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr, ASCII_GT)) { + *nextTokPtr = ptr + MINBPC(enc); + return tok; + } + break; + default: + ptr += MINBPC(enc); + break; + } + } + return XML_TOK_PARTIAL; + case BT_QUEST: + if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr, ASCII_GT)) { + *nextTokPtr = ptr + MINBPC(enc); + return tok; + } + /* fall through */ + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +} + +static int PTRCALL +PREFIX(scanCdataSection)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + static const char CDATA_LSQB[] = { ASCII_C, ASCII_D, ASCII_A, + ASCII_T, ASCII_A, ASCII_LSQB }; + int i; + + UNUSED(enc); + + /* CDATA[ */ + if (end - ptr < 6 * MINBPC(enc)) + return XML_TOK_PARTIAL; + for (i = 0; i < 6; i++, ptr += MINBPC(enc)) { + if (!CHAR_MATCHES(enc, ptr, CDATA_LSQB[i])) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + *nextTokPtr = ptr; + return XML_TOK_CDATA_SECT_OPEN; +} + +static int PTRCALL +PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + if (ptr == end) + return XML_TOK_NONE; + if (MINBPC(enc) > 1) { + size_t n = end - ptr; + if (n & (MINBPC(enc) - 1)) { + n &= ~(MINBPC(enc) - 1); + if (n == 0) + return XML_TOK_PARTIAL; + end = ptr + n; + } + } + switch (BYTE_TYPE(enc, ptr)) { + case BT_RSQB: + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB)) + break; + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) { + ptr -= MINBPC(enc); + break; + } + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_CDATA_SECT_CLOSE; + case BT_CR: + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + if (BYTE_TYPE(enc, ptr) == BT_LF) + ptr += MINBPC(enc); + *nextTokPtr = ptr; + return XML_TOK_DATA_NEWLINE; + case BT_LF: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_DATA_NEWLINE; + INVALID_CASES(ptr, nextTokPtr) + default: + ptr += MINBPC(enc); + break; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { +#define LEAD_CASE(n) \ + case BT_LEAD ## n: \ + if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \ + *nextTokPtr = ptr; \ + return XML_TOK_DATA_CHARS; \ + } \ + ptr += n; \ + break; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_NONXML: + case BT_MALFORM: + case BT_TRAIL: + case BT_CR: + case BT_LF: + case BT_RSQB: + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + default: + ptr += MINBPC(enc); + break; + } + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; +} + +/* ptr points to character following "</" */ + +static int PTRCALL +PREFIX(scanEndTag)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) + case BT_S: case BT_CR: case BT_LF: + for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) { + switch (BYTE_TYPE(enc, ptr)) { + case BT_S: case BT_CR: case BT_LF: + break; + case BT_GT: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_END_TAG; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +#ifdef XML_NS + case BT_COLON: + /* no need to check qname syntax here, + since end-tag must match exactly */ + ptr += MINBPC(enc); + break; +#endif + case BT_GT: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_END_TAG; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +} + +/* ptr points to character following "&#X" */ + +static int PTRCALL +PREFIX(scanHexCharRef)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + if (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + case BT_DIGIT: + case BT_HEX: + break; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) { + switch (BYTE_TYPE(enc, ptr)) { + case BT_DIGIT: + case BT_HEX: + break; + case BT_SEMI: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_CHAR_REF; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + } + return XML_TOK_PARTIAL; +} + +/* ptr points to character following "&#" */ + +static int PTRCALL +PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + if (ptr != end) { + if (CHAR_MATCHES(enc, ptr, ASCII_x)) + return PREFIX(scanHexCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr); + switch (BYTE_TYPE(enc, ptr)) { + case BT_DIGIT: + break; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) { + switch (BYTE_TYPE(enc, ptr)) { + case BT_DIGIT: + break; + case BT_SEMI: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_CHAR_REF; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + } + return XML_TOK_PARTIAL; +} + +/* ptr points to character following "&" */ + +static int PTRCALL +PREFIX(scanRef)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + case BT_NUM: + return PREFIX(scanCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr); + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) + case BT_SEMI: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_ENTITY_REF; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +} + +/* ptr points to character following first character of attribute name */ + +static int PTRCALL +PREFIX(scanAtts)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ +#ifdef XML_NS + int hadColon = 0; +#endif + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) +#ifdef XML_NS + case BT_COLON: + if (hadColon) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + hadColon = 1; + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + break; +#endif + case BT_S: case BT_CR: case BT_LF: + for (;;) { + int t; + + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + t = BYTE_TYPE(enc, ptr); + if (t == BT_EQUALS) + break; + switch (t) { + case BT_S: + case BT_LF: + case BT_CR: + break; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + /* fall through */ + case BT_EQUALS: + { + int open; +#ifdef XML_NS + hadColon = 0; +#endif + for (;;) { + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + open = BYTE_TYPE(enc, ptr); + if (open == BT_QUOT || open == BT_APOS) + break; + switch (open) { + case BT_S: + case BT_LF: + case BT_CR: + break; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + ptr += MINBPC(enc); + /* in attribute value */ + for (;;) { + int t; + if (ptr == end) + return XML_TOK_PARTIAL; + t = BYTE_TYPE(enc, ptr); + if (t == open) + break; + switch (t) { + INVALID_CASES(ptr, nextTokPtr) + case BT_AMP: + { + int tok = PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, &ptr); + if (tok <= 0) { + if (tok == XML_TOK_INVALID) + *nextTokPtr = ptr; + return tok; + } + break; + } + case BT_LT: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + default: + ptr += MINBPC(enc); + break; + } + } + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + case BT_S: + case BT_CR: + case BT_LF: + break; + case BT_SOL: + goto sol; + case BT_GT: + goto gt; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + /* ptr points to closing quote */ + for (;;) { + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + case BT_S: case BT_CR: case BT_LF: + continue; + case BT_GT: + gt: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_START_TAG_WITH_ATTS; + case BT_SOL: + sol: + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_EMPTY_ELEMENT_WITH_ATTS; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + break; + } + break; + } + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +} + +/* ptr points to character following "<" */ + +static int PTRCALL +PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ +#ifdef XML_NS + int hadColon; +#endif + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + case BT_EXCL: + if ((ptr += MINBPC(enc)) == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + case BT_MINUS: + return PREFIX(scanComment)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_LSQB: + return PREFIX(scanCdataSection)(enc, ptr + MINBPC(enc), + end, nextTokPtr); + } + *nextTokPtr = ptr; + return XML_TOK_INVALID; + case BT_QUEST: + return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_SOL: + return PREFIX(scanEndTag)(enc, ptr + MINBPC(enc), end, nextTokPtr); + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } +#ifdef XML_NS + hadColon = 0; +#endif + /* we have a start-tag */ + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) +#ifdef XML_NS + case BT_COLON: + if (hadColon) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + hadColon = 1; + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + break; +#endif + case BT_S: case BT_CR: case BT_LF: + { + ptr += MINBPC(enc); + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + case BT_GT: + goto gt; + case BT_SOL: + goto sol; + case BT_S: case BT_CR: case BT_LF: + ptr += MINBPC(enc); + continue; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + return PREFIX(scanAtts)(enc, ptr, end, nextTokPtr); + } + return XML_TOK_PARTIAL; + } + case BT_GT: + gt: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_START_TAG_NO_ATTS; + case BT_SOL: + sol: + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_EMPTY_ELEMENT_NO_ATTS; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +} + +static int PTRCALL +PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ + if (ptr == end) + return XML_TOK_NONE; + if (MINBPC(enc) > 1) { + size_t n = end - ptr; + if (n & (MINBPC(enc) - 1)) { + n &= ~(MINBPC(enc) - 1); + if (n == 0) + return XML_TOK_PARTIAL; + end = ptr + n; + } + } + switch (BYTE_TYPE(enc, ptr)) { + case BT_LT: + return PREFIX(scanLt)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_AMP: + return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_CR: + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_TRAILING_CR; + if (BYTE_TYPE(enc, ptr) == BT_LF) + ptr += MINBPC(enc); + *nextTokPtr = ptr; + return XML_TOK_DATA_NEWLINE; + case BT_LF: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_DATA_NEWLINE; + case BT_RSQB: + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_TRAILING_RSQB; + if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB)) + break; + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_TRAILING_RSQB; + if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) { + ptr -= MINBPC(enc); + break; + } + *nextTokPtr = ptr; + return XML_TOK_INVALID; + INVALID_CASES(ptr, nextTokPtr) + default: + ptr += MINBPC(enc); + break; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { +#define LEAD_CASE(n) \ + case BT_LEAD ## n: \ + if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \ + *nextTokPtr = ptr; \ + return XML_TOK_DATA_CHARS; \ + } \ + ptr += n; \ + break; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_RSQB: + if (ptr + MINBPC(enc) != end) { + if (!CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_RSQB)) { + ptr += MINBPC(enc); + break; + } + if (ptr + 2*MINBPC(enc) != end) { + if (!CHAR_MATCHES(enc, ptr + 2*MINBPC(enc), ASCII_GT)) { + ptr += MINBPC(enc); + break; + } + *nextTokPtr = ptr + 2*MINBPC(enc); + return XML_TOK_INVALID; + } + } + /* fall through */ + case BT_AMP: + case BT_LT: + case BT_NONXML: + case BT_MALFORM: + case BT_TRAIL: + case BT_CR: + case BT_LF: + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + default: + ptr += MINBPC(enc); + break; + } + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; +} + +/* ptr points to character following "%" */ + +static int PTRCALL +PREFIX(scanPercent)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + case BT_S: case BT_LF: case BT_CR: case BT_PERCNT: + *nextTokPtr = ptr; + return XML_TOK_PERCENT; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) + case BT_SEMI: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_PARAM_ENTITY_REF; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return XML_TOK_PARTIAL; +} + +static int PTRCALL +PREFIX(scanPoundName)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) + case BT_CR: case BT_LF: case BT_S: + case BT_RPAR: case BT_GT: case BT_PERCNT: case BT_VERBAR: + *nextTokPtr = ptr; + return XML_TOK_POUND_NAME; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return -XML_TOK_POUND_NAME; +} + +static int PTRCALL +PREFIX(scanLit)(int open, const ENCODING *enc, + const char *ptr, const char *end, + const char **nextTokPtr) +{ + while (ptr != end) { + int t = BYTE_TYPE(enc, ptr); + switch (t) { + INVALID_CASES(ptr, nextTokPtr) + case BT_QUOT: + case BT_APOS: + ptr += MINBPC(enc); + if (t != open) + break; + if (ptr == end) + return -XML_TOK_LITERAL; + *nextTokPtr = ptr; + switch (BYTE_TYPE(enc, ptr)) { + case BT_S: case BT_CR: case BT_LF: + case BT_GT: case BT_PERCNT: case BT_LSQB: + return XML_TOK_LITERAL; + default: + return XML_TOK_INVALID; + } + default: + ptr += MINBPC(enc); + break; + } + } + return XML_TOK_PARTIAL; +} + +static int PTRCALL +PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ + int tok; + if (ptr == end) + return XML_TOK_NONE; + if (MINBPC(enc) > 1) { + size_t n = end - ptr; + if (n & (MINBPC(enc) - 1)) { + n &= ~(MINBPC(enc) - 1); + if (n == 0) + return XML_TOK_PARTIAL; + end = ptr + n; + } + } + switch (BYTE_TYPE(enc, ptr)) { + case BT_QUOT: + return PREFIX(scanLit)(BT_QUOT, enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_APOS: + return PREFIX(scanLit)(BT_APOS, enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_LT: + { + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_PARTIAL; + switch (BYTE_TYPE(enc, ptr)) { + case BT_EXCL: + return PREFIX(scanDecl)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_QUEST: + return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_NMSTRT: + case BT_HEX: + case BT_NONASCII: + case BT_LEAD2: + case BT_LEAD3: + case BT_LEAD4: + *nextTokPtr = ptr - MINBPC(enc); + return XML_TOK_INSTANCE_START; + } + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + case BT_CR: + if (ptr + MINBPC(enc) == end) { + *nextTokPtr = end; + /* indicate that this might be part of a CR/LF pair */ + return -XML_TOK_PROLOG_S; + } + /* fall through */ + case BT_S: case BT_LF: + for (;;) { + ptr += MINBPC(enc); + if (ptr == end) + break; + switch (BYTE_TYPE(enc, ptr)) { + case BT_S: case BT_LF: + break; + case BT_CR: + /* don't split CR/LF pair */ + if (ptr + MINBPC(enc) != end) + break; + /* fall through */ + default: + *nextTokPtr = ptr; + return XML_TOK_PROLOG_S; + } + } + *nextTokPtr = ptr; + return XML_TOK_PROLOG_S; + case BT_PERCNT: + return PREFIX(scanPercent)(enc, ptr + MINBPC(enc), end, nextTokPtr); + case BT_COMMA: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_COMMA; + case BT_LSQB: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_OPEN_BRACKET; + case BT_RSQB: + ptr += MINBPC(enc); + if (ptr == end) + return -XML_TOK_CLOSE_BRACKET; + if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) { + if (ptr + MINBPC(enc) == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_GT)) { + *nextTokPtr = ptr + 2*MINBPC(enc); + return XML_TOK_COND_SECT_CLOSE; + } + } + *nextTokPtr = ptr; + return XML_TOK_CLOSE_BRACKET; + case BT_LPAR: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_OPEN_PAREN; + case BT_RPAR: + ptr += MINBPC(enc); + if (ptr == end) + return -XML_TOK_CLOSE_PAREN; + switch (BYTE_TYPE(enc, ptr)) { + case BT_AST: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_CLOSE_PAREN_ASTERISK; + case BT_QUEST: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_CLOSE_PAREN_QUESTION; + case BT_PLUS: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_CLOSE_PAREN_PLUS; + case BT_CR: case BT_LF: case BT_S: + case BT_GT: case BT_COMMA: case BT_VERBAR: + case BT_RPAR: + *nextTokPtr = ptr; + return XML_TOK_CLOSE_PAREN; + } + *nextTokPtr = ptr; + return XML_TOK_INVALID; + case BT_VERBAR: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_OR; + case BT_GT: + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_DECL_CLOSE; + case BT_NUM: + return PREFIX(scanPoundName)(enc, ptr + MINBPC(enc), end, nextTokPtr); +#define LEAD_CASE(n) \ + case BT_LEAD ## n: \ + if (end - ptr < n) \ + return XML_TOK_PARTIAL_CHAR; \ + if (IS_NMSTRT_CHAR(enc, ptr, n)) { \ + ptr += n; \ + tok = XML_TOK_NAME; \ + break; \ + } \ + if (IS_NAME_CHAR(enc, ptr, n)) { \ + ptr += n; \ + tok = XML_TOK_NMTOKEN; \ + break; \ + } \ + *nextTokPtr = ptr; \ + return XML_TOK_INVALID; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_NMSTRT: + case BT_HEX: + tok = XML_TOK_NAME; + ptr += MINBPC(enc); + break; + case BT_DIGIT: + case BT_NAME: + case BT_MINUS: +#ifdef XML_NS + case BT_COLON: +#endif + tok = XML_TOK_NMTOKEN; + ptr += MINBPC(enc); + break; + case BT_NONASCII: + if (IS_NMSTRT_CHAR_MINBPC(enc, ptr)) { + ptr += MINBPC(enc); + tok = XML_TOK_NAME; + break; + } + if (IS_NAME_CHAR_MINBPC(enc, ptr)) { + ptr += MINBPC(enc); + tok = XML_TOK_NMTOKEN; + break; + } + /* fall through */ + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) + case BT_GT: case BT_RPAR: case BT_COMMA: + case BT_VERBAR: case BT_LSQB: case BT_PERCNT: + case BT_S: case BT_CR: case BT_LF: + *nextTokPtr = ptr; + return tok; +#ifdef XML_NS + case BT_COLON: + ptr += MINBPC(enc); + switch (tok) { + case XML_TOK_NAME: + if (ptr == end) + return XML_TOK_PARTIAL; + tok = XML_TOK_PREFIXED_NAME; + switch (BYTE_TYPE(enc, ptr)) { + CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) + default: + tok = XML_TOK_NMTOKEN; + break; + } + break; + case XML_TOK_PREFIXED_NAME: + tok = XML_TOK_NMTOKEN; + break; + } + break; +#endif + case BT_PLUS: + if (tok == XML_TOK_NMTOKEN) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_NAME_PLUS; + case BT_AST: + if (tok == XML_TOK_NMTOKEN) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_NAME_ASTERISK; + case BT_QUEST: + if (tok == XML_TOK_NMTOKEN) { + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_NAME_QUESTION; + default: + *nextTokPtr = ptr; + return XML_TOK_INVALID; + } + } + return -tok; +} + +static int PTRCALL +PREFIX(attributeValueTok)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + const char *start; + if (ptr == end) + return XML_TOK_NONE; + start = ptr; + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { +#define LEAD_CASE(n) \ + case BT_LEAD ## n: ptr += n; break; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_AMP: + if (ptr == start) + return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr); + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + case BT_LT: + /* this is for inside entity references */ + *nextTokPtr = ptr; + return XML_TOK_INVALID; + case BT_LF: + if (ptr == start) { + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_DATA_NEWLINE; + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + case BT_CR: + if (ptr == start) { + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_TRAILING_CR; + if (BYTE_TYPE(enc, ptr) == BT_LF) + ptr += MINBPC(enc); + *nextTokPtr = ptr; + return XML_TOK_DATA_NEWLINE; + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + case BT_S: + if (ptr == start) { + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_ATTRIBUTE_VALUE_S; + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + default: + ptr += MINBPC(enc); + break; + } + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; +} + +static int PTRCALL +PREFIX(entityValueTok)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + const char *start; + if (ptr == end) + return XML_TOK_NONE; + start = ptr; + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { +#define LEAD_CASE(n) \ + case BT_LEAD ## n: ptr += n; break; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_AMP: + if (ptr == start) + return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr); + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + case BT_PERCNT: + if (ptr == start) { + int tok = PREFIX(scanPercent)(enc, ptr + MINBPC(enc), + end, nextTokPtr); + return (tok == XML_TOK_PERCENT) ? XML_TOK_INVALID : tok; + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + case BT_LF: + if (ptr == start) { + *nextTokPtr = ptr + MINBPC(enc); + return XML_TOK_DATA_NEWLINE; + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + case BT_CR: + if (ptr == start) { + ptr += MINBPC(enc); + if (ptr == end) + return XML_TOK_TRAILING_CR; + if (BYTE_TYPE(enc, ptr) == BT_LF) + ptr += MINBPC(enc); + *nextTokPtr = ptr; + return XML_TOK_DATA_NEWLINE; + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; + default: + ptr += MINBPC(enc); + break; + } + } + *nextTokPtr = ptr; + return XML_TOK_DATA_CHARS; +} + +#ifdef XML_DTD + +static int PTRCALL +PREFIX(ignoreSectionTok)(const ENCODING *enc, const char *ptr, + const char *end, const char **nextTokPtr) +{ + int level = 0; + if (MINBPC(enc) > 1) { + size_t n = end - ptr; + if (n & (MINBPC(enc) - 1)) { + n &= ~(MINBPC(enc) - 1); + end = ptr + n; + } + } + while (ptr != end) { + switch (BYTE_TYPE(enc, ptr)) { + INVALID_CASES(ptr, nextTokPtr) + case BT_LT: + if ((ptr += MINBPC(enc)) == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr, ASCII_EXCL)) { + if ((ptr += MINBPC(enc)) == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr, ASCII_LSQB)) { + ++level; + ptr += MINBPC(enc); + } + } + break; + case BT_RSQB: + if ((ptr += MINBPC(enc)) == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) { + if ((ptr += MINBPC(enc)) == end) + return XML_TOK_PARTIAL; + if (CHAR_MATCHES(enc, ptr, ASCII_GT)) { + ptr += MINBPC(enc); + if (level == 0) { + *nextTokPtr = ptr; + return XML_TOK_IGNORE_SECT; + } + --level; + } + } + break; + default: + ptr += MINBPC(enc); + break; + } + } + return XML_TOK_PARTIAL; +} + +#endif /* XML_DTD */ + +static int PTRCALL +PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end, + const char **badPtr) +{ + ptr += MINBPC(enc); + end -= MINBPC(enc); + for (; ptr != end; ptr += MINBPC(enc)) { + switch (BYTE_TYPE(enc, ptr)) { + case BT_DIGIT: + case BT_HEX: + case BT_MINUS: + case BT_APOS: + case BT_LPAR: + case BT_RPAR: + case BT_PLUS: + case BT_COMMA: + case BT_SOL: + case BT_EQUALS: + case BT_QUEST: + case BT_CR: + case BT_LF: + case BT_SEMI: + case BT_EXCL: + case BT_AST: + case BT_PERCNT: + case BT_NUM: +#ifdef XML_NS + case BT_COLON: +#endif + break; + case BT_S: + if (CHAR_MATCHES(enc, ptr, ASCII_TAB)) { + *badPtr = ptr; + return 0; + } + break; + case BT_NAME: + case BT_NMSTRT: + if (!(BYTE_TO_ASCII(enc, ptr) & ~0x7f)) + break; + default: + switch (BYTE_TO_ASCII(enc, ptr)) { + case 0x24: /* $ */ + case 0x40: /* @ */ + break; + default: + *badPtr = ptr; + return 0; + } + break; + } + } + return 1; +} + +/* This must only be called for a well-formed start-tag or empty + element tag. Returns the number of attributes. Pointers to the + first attsMax attributes are stored in atts. +*/ + +static int PTRCALL +PREFIX(getAtts)(const ENCODING *enc, const char *ptr, + int attsMax, ATTRIBUTE *atts) +{ + enum { other, inName, inValue } state = inName; + int nAtts = 0; + int open = 0; /* defined when state == inValue; + initialization just to shut up compilers */ + + for (ptr += MINBPC(enc);; ptr += MINBPC(enc)) { + switch (BYTE_TYPE(enc, ptr)) { +#define START_NAME \ + if (state == other) { \ + if (nAtts < attsMax) { \ + atts[nAtts].name = ptr; \ + atts[nAtts].normalized = 1; \ + } \ + state = inName; \ + } +#define LEAD_CASE(n) \ + case BT_LEAD ## n: START_NAME ptr += (n - MINBPC(enc)); break; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_NONASCII: + case BT_NMSTRT: + case BT_HEX: + START_NAME + break; +#undef START_NAME + case BT_QUOT: + if (state != inValue) { + if (nAtts < attsMax) + atts[nAtts].valuePtr = ptr + MINBPC(enc); + state = inValue; + open = BT_QUOT; + } + else if (open == BT_QUOT) { + state = other; + if (nAtts < attsMax) + atts[nAtts].valueEnd = ptr; + nAtts++; + } + break; + case BT_APOS: + if (state != inValue) { + if (nAtts < attsMax) + atts[nAtts].valuePtr = ptr + MINBPC(enc); + state = inValue; + open = BT_APOS; + } + else if (open == BT_APOS) { + state = other; + if (nAtts < attsMax) + atts[nAtts].valueEnd = ptr; + nAtts++; + } + break; + case BT_AMP: + if (nAtts < attsMax) + atts[nAtts].normalized = 0; + break; + case BT_S: + if (state == inName) + state = other; + else if (state == inValue + && nAtts < attsMax + && atts[nAtts].normalized + && (ptr == atts[nAtts].valuePtr + || BYTE_TO_ASCII(enc, ptr) != ASCII_SPACE + || BYTE_TO_ASCII(enc, ptr + MINBPC(enc)) == ASCII_SPACE + || BYTE_TYPE(enc, ptr + MINBPC(enc)) == open)) + atts[nAtts].normalized = 0; + break; + case BT_CR: case BT_LF: + /* This case ensures that the first attribute name is counted + Apart from that we could just change state on the quote. */ + if (state == inName) + state = other; + else if (state == inValue && nAtts < attsMax) + atts[nAtts].normalized = 0; + break; + case BT_GT: + case BT_SOL: + if (state != inValue) + return nAtts; + break; + default: + break; + } + } + /* not reached */ +} + +static int PTRFASTCALL +PREFIX(charRefNumber)(const ENCODING *enc, const char *ptr) +{ + int result = 0; + + UNUSED(enc); + + /* skip &# */ + ptr += 2*MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_x)) { + for (ptr += MINBPC(enc); + !CHAR_MATCHES(enc, ptr, ASCII_SEMI); + ptr += MINBPC(enc)) { + int c = BYTE_TO_ASCII(enc, ptr); + switch (c) { + case ASCII_0: case ASCII_1: case ASCII_2: case ASCII_3: case ASCII_4: + case ASCII_5: case ASCII_6: case ASCII_7: case ASCII_8: case ASCII_9: + result <<= 4; + result |= (c - ASCII_0); + break; + case ASCII_A: case ASCII_B: case ASCII_C: + case ASCII_D: case ASCII_E: case ASCII_F: + result <<= 4; + result += 10 + (c - ASCII_A); + break; + case ASCII_a: case ASCII_b: case ASCII_c: + case ASCII_d: case ASCII_e: case ASCII_f: + result <<= 4; + result += 10 + (c - ASCII_a); + break; + } + if (result >= 0x110000) + return -1; + } + } + else { + for (; !CHAR_MATCHES(enc, ptr, ASCII_SEMI); ptr += MINBPC(enc)) { + int c = BYTE_TO_ASCII(enc, ptr); + result *= 10; + result += (c - ASCII_0); + if (result >= 0x110000) + return -1; + } + } + return checkCharRefNumber(result); +} + +static int PTRCALL +PREFIX(predefinedEntityName)(const ENCODING *enc, const char *ptr, + const char *end) +{ + UNUSED(enc); + + switch ((end - ptr)/MINBPC(enc)) { + case 2: + if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_t)) { + switch (BYTE_TO_ASCII(enc, ptr)) { + case ASCII_l: + return ASCII_LT; + case ASCII_g: + return ASCII_GT; + } + } + break; + case 3: + if (CHAR_MATCHES(enc, ptr, ASCII_a)) { + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_m)) { + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_p)) + return ASCII_AMP; + } + } + break; + case 4: + switch (BYTE_TO_ASCII(enc, ptr)) { + case ASCII_q: + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_u)) { + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_o)) { + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_t)) + return ASCII_QUOT; + } + } + break; + case ASCII_a: + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_p)) { + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_o)) { + ptr += MINBPC(enc); + if (CHAR_MATCHES(enc, ptr, ASCII_s)) + return ASCII_APOS; + } + } + break; + } + } + return 0; +} + +static int PTRCALL +PREFIX(sameName)(const ENCODING *enc, const char *ptr1, const char *ptr2) +{ + for (;;) { + switch (BYTE_TYPE(enc, ptr1)) { +#define LEAD_CASE(n) \ + case BT_LEAD ## n: \ + if (*ptr1++ != *ptr2++) \ + return 0; + LEAD_CASE(4) LEAD_CASE(3) LEAD_CASE(2) +#undef LEAD_CASE + /* fall through */ + if (*ptr1++ != *ptr2++) + return 0; + break; + case BT_NONASCII: + case BT_NMSTRT: +#ifdef XML_NS + case BT_COLON: +#endif + case BT_HEX: + case BT_DIGIT: + case BT_NAME: + case BT_MINUS: + if (*ptr2++ != *ptr1++) + return 0; + if (MINBPC(enc) > 1) { + if (*ptr2++ != *ptr1++) + return 0; + if (MINBPC(enc) > 2) { + if (*ptr2++ != *ptr1++) + return 0; + if (MINBPC(enc) > 3) { + if (*ptr2++ != *ptr1++) + return 0; + } + } + } + break; + default: + if (MINBPC(enc) == 1 && *ptr1 == *ptr2) + return 1; + switch (BYTE_TYPE(enc, ptr2)) { + case BT_LEAD2: + case BT_LEAD3: + case BT_LEAD4: + case BT_NONASCII: + case BT_NMSTRT: +#ifdef XML_NS + case BT_COLON: +#endif + case BT_HEX: + case BT_DIGIT: + case BT_NAME: + case BT_MINUS: + return 0; + default: + return 1; + } + } + } + /* not reached */ +} + +static int PTRCALL +PREFIX(nameMatchesAscii)(const ENCODING *enc, const char *ptr1, + const char *end1, const char *ptr2) +{ + UNUSED(enc); + + for (; *ptr2; ptr1 += MINBPC(enc), ptr2++) { + if (ptr1 == end1) + return 0; + if (!CHAR_MATCHES(enc, ptr1, *ptr2)) + return 0; + } + return ptr1 == end1; +} + +static int PTRFASTCALL +PREFIX(nameLength)(const ENCODING *enc, const char *ptr) +{ + const char *start = ptr; + for (;;) { + switch (BYTE_TYPE(enc, ptr)) { +#define LEAD_CASE(n) \ + case BT_LEAD ## n: ptr += n; break; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_NONASCII: + case BT_NMSTRT: +#ifdef XML_NS + case BT_COLON: +#endif + case BT_HEX: + case BT_DIGIT: + case BT_NAME: + case BT_MINUS: + ptr += MINBPC(enc); + break; + default: + return (int)(ptr - start); + } + } +} + +static const char * PTRFASTCALL +PREFIX(skipS)(const ENCODING *enc, const char *ptr) +{ + for (;;) { + switch (BYTE_TYPE(enc, ptr)) { + case BT_LF: + case BT_CR: + case BT_S: + ptr += MINBPC(enc); + break; + default: + return ptr; + } + } +} + +static void PTRCALL +PREFIX(updatePosition)(const ENCODING *enc, + const char *ptr, + const char *end, + POSITION *pos) +{ + while (ptr < end) { + switch (BYTE_TYPE(enc, ptr)) { +#define LEAD_CASE(n) \ + case BT_LEAD ## n: \ + ptr += n; \ + break; + LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4) +#undef LEAD_CASE + case BT_LF: + pos->columnNumber = (XML_Size)-1; + pos->lineNumber++; + ptr += MINBPC(enc); + break; + case BT_CR: + pos->lineNumber++; + ptr += MINBPC(enc); + if (ptr != end && BYTE_TYPE(enc, ptr) == BT_LF) + ptr += MINBPC(enc); + pos->columnNumber = (XML_Size)-1; + break; + default: + ptr += MINBPC(enc); + break; + } + pos->columnNumber++; + } +} + +#undef DO_LEAD_CASE +#undef MULTIBYTE_CASES +#undef INVALID_CASES +#undef CHECK_NAME_CASE +#undef CHECK_NAME_CASES +#undef CHECK_NMSTRT_CASE +#undef CHECK_NMSTRT_CASES + +#endif /* XML_TOK_IMPL_C */ diff --git a/xml/details/expat/xmltok_impl.h b/xml/details/expat/xmltok_impl.h new file mode 100644 index 0000000..da0ea60 --- /dev/null +++ b/xml/details/expat/xmltok_impl.h @@ -0,0 +1,46 @@ +/* +Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd +See the file COPYING for copying permission. +*/ + +enum { + BT_NONXML, + BT_MALFORM, + BT_LT, + BT_AMP, + BT_RSQB, + BT_LEAD2, + BT_LEAD3, + BT_LEAD4, + BT_TRAIL, + BT_CR, + BT_LF, + BT_GT, + BT_QUOT, + BT_APOS, + BT_EQUALS, + BT_QUEST, + BT_EXCL, + BT_SOL, + BT_SEMI, + BT_NUM, + BT_LSQB, + BT_S, + BT_NMSTRT, + BT_COLON, + BT_HEX, + BT_DIGIT, + BT_NAME, + BT_MINUS, + BT_OTHER, /* known not to be a name or name start character */ + BT_NONASCII, /* might be a name or name start character */ + BT_PERCNT, + BT_LPAR, + BT_RPAR, + BT_AST, + BT_PLUS, + BT_COMMA, + BT_VERBAR +}; + +#include <stddef.h> diff --git a/xml/details/expat/xmltok_ns.c b/xml/details/expat/xmltok_ns.c new file mode 100644 index 0000000..c3b88fd --- /dev/null +++ b/xml/details/expat/xmltok_ns.c @@ -0,0 +1,115 @@ +/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd + See the file COPYING for copying permission. +*/ + +/* This file is included! */ +#ifdef XML_TOK_NS_C + +const ENCODING * +NS(XmlGetUtf8InternalEncoding)(void) +{ + return &ns(internal_utf8_encoding).enc; +} + +const ENCODING * +NS(XmlGetUtf16InternalEncoding)(void) +{ +#if BYTEORDER == 1234 + return &ns(internal_little2_encoding).enc; +#elif BYTEORDER == 4321 + return &ns(internal_big2_encoding).enc; +#else + const short n = 1; + return (*(const char *)&n + ? &ns(internal_little2_encoding).enc + : &ns(internal_big2_encoding).enc); +#endif +} + +static const ENCODING * const NS(encodings)[] = { + &ns(latin1_encoding).enc, + &ns(ascii_encoding).enc, + &ns(utf8_encoding).enc, + &ns(big2_encoding).enc, + &ns(big2_encoding).enc, + &ns(little2_encoding).enc, + &ns(utf8_encoding).enc /* NO_ENC */ +}; + +static int PTRCALL +NS(initScanProlog)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ + return initScan(NS(encodings), (const INIT_ENCODING *)enc, + XML_PROLOG_STATE, ptr, end, nextTokPtr); +} + +static int PTRCALL +NS(initScanContent)(const ENCODING *enc, const char *ptr, const char *end, + const char **nextTokPtr) +{ + return initScan(NS(encodings), (const INIT_ENCODING *)enc, + XML_CONTENT_STATE, ptr, end, nextTokPtr); +} + +int +NS(XmlInitEncoding)(INIT_ENCODING *p, const ENCODING **encPtr, + const char *name) +{ + int i = getEncodingIndex(name); + if (i == UNKNOWN_ENC) + return 0; + SET_INIT_ENC_INDEX(p, i); + p->initEnc.scanners[XML_PROLOG_STATE] = NS(initScanProlog); + p->initEnc.scanners[XML_CONTENT_STATE] = NS(initScanContent); + p->initEnc.updatePosition = initUpdatePosition; + p->encPtr = encPtr; + *encPtr = &(p->initEnc); + return 1; +} + +static const ENCODING * +NS(findEncoding)(const ENCODING *enc, const char *ptr, const char *end) +{ +#define ENCODING_MAX 128 + char buf[ENCODING_MAX]; + char *p = buf; + int i; + XmlUtf8Convert(enc, &ptr, end, &p, p + ENCODING_MAX - 1); + if (ptr != end) + return 0; + *p = 0; + if (streqci(buf, KW_UTF_16) && enc->minBytesPerChar == 2) + return enc; + i = getEncodingIndex(buf); + if (i == UNKNOWN_ENC) + return 0; + return NS(encodings)[i]; +} + +int +NS(XmlParseXmlDecl)(int isGeneralTextEntity, + const ENCODING *enc, + const char *ptr, + const char *end, + const char **badPtr, + const char **versionPtr, + const char **versionEndPtr, + const char **encodingName, + const ENCODING **encoding, + int *standalone) +{ + return doParseXmlDecl(NS(findEncoding), + isGeneralTextEntity, + enc, + ptr, + end, + badPtr, + versionPtr, + versionEndPtr, + encodingName, + encoding, + standalone); +} + +#endif /* XML_TOK_NS_C */ diff --git a/xml/details/export.hxx b/xml/details/export.hxx new file mode 100644 index 0000000..fb6d3b6 --- /dev/null +++ b/xml/details/export.hxx @@ -0,0 +1,36 @@ +// file : xml/details/export.hxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#ifndef XML_DETAILS_EXPORT_HXX +#define XML_DETAILS_EXPORT_HXX + +#include <xml/details/config.hxx> + +#ifdef LIBSTUDXML_STATIC_LIB +# define LIBSTUDXML_EXPORT +#else +# ifdef _WIN32 +# ifdef _MSC_VER +# ifdef LIBSTUDXML_DYNAMIC_LIB +# define LIBSTUDXML_EXPORT __declspec(dllexport) +# else +# define LIBSTUDXML_EXPORT __declspec(dllimport) +# endif +# else +# ifdef LIBSTUDXML_DYNAMIC_LIB +# ifdef DLL_EXPORT +# define LIBSTUDXML_EXPORT __declspec(dllexport) +# else +# define LIBSTUDXML_EXPORT +# endif +# else +# define LIBSTUDXML_EXPORT __declspec(dllimport) +# endif +# endif +# else +# define LIBSTUDXML_EXPORT +# endif +#endif + +#endif // XML_DETAILS_EXPORT_HXX diff --git a/xml/details/genx/LICENSE b/xml/details/genx/LICENSE new file mode 100644 index 0000000..0248bdb --- /dev/null +++ b/xml/details/genx/LICENSE @@ -0,0 +1,22 @@ +Copyright (c) 2007-2013 Code Synthesis Tools CC. +Copyright (c) Tim Bray and Sun Microsystems, 2004. + +Permission is hereby granted, free of charge, to any person obtaining +a copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to +permit persons to whom the Software is furnished to do so, subject to +the following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + diff --git a/xml/details/genx/README b/xml/details/genx/README new file mode 100644 index 0000000..10a7238 --- /dev/null +++ b/xml/details/genx/README @@ -0,0 +1,2 @@ +This directory contains a customized version of the Genx XML generator +library for internal use by libstudxml. diff --git a/xml/details/genx/char-props.c b/xml/details/genx/char-props.c new file mode 100644 index 0000000..5f9c152 --- /dev/null +++ b/xml/details/genx/char-props.c @@ -0,0 +1,395 @@ +/* + * Copyright (c) 2007-2013 Code Synthesis Tools CC. + * Copyright (c) 2004 by Tim Bray and Sun Microsystems. + * + * For copying permission, see the accompanying COPYING file. + */ + +/* + * Construct character-properties tables for genx. + * Quite likely there's a better way. + * This version is generated semi-automatically from the source code of the + * XML specification via emacs global replace and keyboard macros + */ +#include <xml/details/genx/genx.h> + +static void charProp(char * p, int c, int prop) +{ + p[c] |= prop; +} + +static void rangeProp(char * p, int start, int end, int prop) +{ + int i; + for (i = start; i <= end; i++) + p[i] |= prop; +} + +void genxSetCharProps(char * p) +{ + int i; + + for (i = 0; i < GENX_CHAR_TABLE_SIZE; i++) + p[i] = 0; + + /* per XML 1.0 */ + charProp(p, 0x9, GENX_XML_CHAR); + charProp(p, 0xa, GENX_XML_CHAR); + charProp(p, 0xd, GENX_XML_CHAR); + rangeProp(p, 0x20, 0xff, GENX_XML_CHAR); + +#if GENX_CHAR_TABLE_SIZE == 0x10000 + rangeProp(p, 0x0100, 0xd7ff, GENX_XML_CHAR); + rangeProp(p, 0xe000, 0xfffd, GENX_XML_CHAR); +#endif + + /* Letter ::= BaseChar | Ideographic */ + rangeProp(p, 0x0041, 0x005A, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0061, 0x007A, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x00C0, 0x00D6, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x00D8, 0x00F6, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x00F8, 0x00FF, GENX_LETTER|GENX_NAMECHAR); + +#if GENX_CHAR_TABLE_SIZE == 0x10000 + + rangeProp(p, 0x0100, 0x0131, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0134, 0x013E, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0141, 0x0148, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x014A, 0x017E, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0180, 0x01C3, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x01CD, 0x01F0, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x01F4, 0x01F5, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x01FA, 0x0217, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0250, 0x02A8, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x02BB, 0x02C1, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0386, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0388, 0x038A, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x038C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x038E, 0x03A1, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x03A3, 0x03CE, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x03D0, 0x03D6, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x03DA, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x03DC, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x03DE, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x03E0, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x03E2, 0x03F3, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0401, 0x040C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x040E, 0x044F, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0451, 0x045C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x045E, 0x0481, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0490, 0x04C4, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x04C7, 0x04C8, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x04CB, 0x04CC, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x04D0, 0x04EB, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x04EE, 0x04F5, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x04F8, 0x04F9, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0531, 0x0556, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0559, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0561, 0x0586, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x05D0, 0x05EA, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x05F0, 0x05F2, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0621, 0x063A, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0641, 0x064A, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0671, 0x06B7, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x06BA, 0x06BE, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x06C0, 0x06CE, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x06D0, 0x06D3, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x06D5, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x06E5, 0x06E6, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0905, 0x0939, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x093D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0958, 0x0961, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0985, 0x098C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x098F, 0x0990, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0993, 0x09A8, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x09AA, 0x09B0, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x09B2, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x09B6, 0x09B9, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x09DC, 0x09DD, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x09DF, 0x09E1, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x09F0, 0x09F1, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A05, 0x0A0A, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A0F, 0x0A10, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A13, 0x0A28, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A2A, 0x0A30, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A32, 0x0A33, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A35, 0x0A36, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A38, 0x0A39, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A59, 0x0A5C, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0A5E, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A72, 0x0A74, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A85, 0x0A8B, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0A8D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A8F, 0x0A91, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A93, 0x0AA8, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0AAA, 0x0AB0, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0AB2, 0x0AB3, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0AB5, 0x0AB9, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0ABD, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0AE0, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B05, 0x0B0C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B0F, 0x0B10, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B13, 0x0B28, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B2A, 0x0B30, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B32, 0x0B33, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B36, 0x0B39, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0B3D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B5C, 0x0B5D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B5F, 0x0B61, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B85, 0x0B8A, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B8E, 0x0B90, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B92, 0x0B95, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B99, 0x0B9A, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0B9C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B9E, 0x0B9F, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0BA3, 0x0BA4, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0BA8, 0x0BAA, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0BAE, 0x0BB5, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0BB7, 0x0BB9, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C05, 0x0C0C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C0E, 0x0C10, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C12, 0x0C28, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C2A, 0x0C33, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C35, 0x0C39, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C60, 0x0C61, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C85, 0x0C8C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C8E, 0x0C90, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C92, 0x0CA8, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0CAA, 0x0CB3, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0CB5, 0x0CB9, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0CDE, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0CE0, 0x0CE1, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0D05, 0x0D0C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0D0E, 0x0D10, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0D12, 0x0D28, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0D2A, 0x0D39, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0D60, 0x0D61, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E01, 0x0E2E, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0E30, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E32, 0x0E33, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E40, 0x0E45, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E81, 0x0E82, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0E84, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E87, 0x0E88, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0E8A, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0E8D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E94, 0x0E97, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E99, 0x0E9F, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0EA1, 0x0EA3, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0EA5, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0EA7, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0EAA, 0x0EAB, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0EAD, 0x0EAE, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0EB0, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0EB2, 0x0EB3, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0EBD, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0EC0, 0x0EC4, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0F40, 0x0F47, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0F49, 0x0F69, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x10A0, 0x10C5, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x10D0, 0x10F6, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1100, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1102, 0x1103, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1105, 0x1107, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1109, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x110B, 0x110C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x110E, 0x1112, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x113C, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x113E, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1140, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x114C, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x114E, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1150, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1154, 0x1155, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1159, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x115F, 0x1161, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1163, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1165, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1167, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1169, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x116D, 0x116E, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1172, 0x1173, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1175, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x119E, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x11A8, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x11AB, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x11AE, 0x11AF, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x11B7, 0x11B8, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x11BA, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x11BC, 0x11C2, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x11EB, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x11F0, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x11F9, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1E00, 0x1E9B, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1EA0, 0x1EF9, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1F00, 0x1F15, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1F18, 0x1F1D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1F20, 0x1F45, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1F48, 0x1F4D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1F50, 0x1F57, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1F59, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1F5B, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1F5D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1F5F, 0x1F7D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1F80, 0x1FB4, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FB6, 0x1FBC, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x1FBE, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FC2, 0x1FC4, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FC6, 0x1FCC, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FD0, 0x1FD3, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FD6, 0x1FDB, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FE0, 0x1FEC, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FF2, 0x1FF4, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x1FF6, 0x1FFC, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x2126, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x212A, 0x212B, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x212E, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x2180, 0x2182, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x3041, 0x3094, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x30A1, 0x30FA, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x3105, 0x312C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0xAC00, 0xD7A3, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x4E00, 0x9FA5, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x3007, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x3021, 0x3029, GENX_LETTER|GENX_NAMECHAR); + +#endif /* GENX_CHAR_TABLE_SIZE == 0x10000 */ + + /* + * NameChar ::= + * Letter | Digit | '.' | '-' | '_' | ':' | CombiningChar | Extender + */ + + charProp(p, '.', GENX_NAMECHAR); + charProp(p, '-', GENX_NAMECHAR); + charProp(p, '_', GENX_NAMECHAR); + + rangeProp(p, 0x0030, 0x0039, GENX_NAMECHAR); + charProp(p, 0x00B7, GENX_LETTER|GENX_NAMECHAR); + +#if GENX_CHAR_TABLE_SIZE == 0x10000 + + rangeProp(p, 0x0660, 0x0669, GENX_NAMECHAR); + rangeProp(p, 0x06F0, 0x06F9, GENX_NAMECHAR); + rangeProp(p, 0x0966, 0x096F, GENX_NAMECHAR); + rangeProp(p, 0x09E6, 0x09EF, GENX_NAMECHAR); + rangeProp(p, 0x0A66, 0x0A6F, GENX_NAMECHAR); + rangeProp(p, 0x0AE6, 0x0AEF, GENX_NAMECHAR); + rangeProp(p, 0x0B66, 0x0B6F, GENX_NAMECHAR); + rangeProp(p, 0x0BE7, 0x0BEF, GENX_NAMECHAR); + rangeProp(p, 0x0C66, 0x0C6F, GENX_NAMECHAR); + rangeProp(p, 0x0CE6, 0x0CEF, GENX_NAMECHAR); + rangeProp(p, 0x0D66, 0x0D6F, GENX_NAMECHAR); + rangeProp(p, 0x0E50, 0x0E59, GENX_NAMECHAR); + rangeProp(p, 0x0ED0, 0x0ED9, GENX_NAMECHAR); + rangeProp(p, 0x0F20, 0x0F29, GENX_NAMECHAR); + rangeProp(p, 0x0300, 0x0345, GENX_NAMECHAR); + rangeProp(p, 0x0360, 0x0361, GENX_NAMECHAR); + rangeProp(p, 0x0483, 0x0486, GENX_NAMECHAR); + rangeProp(p, 0x0591, 0x05A1, GENX_NAMECHAR); + rangeProp(p, 0x05A3, 0x05B9, GENX_NAMECHAR); + rangeProp(p, 0x05BB, 0x05BD, GENX_NAMECHAR); + charProp(p, 0x05BF, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x05C1, 0x05C2, GENX_NAMECHAR); + charProp(p, 0x05C4, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x064B, 0x0652, GENX_NAMECHAR); + charProp(p, 0x0670, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x06D6, 0x06DC, GENX_NAMECHAR); + rangeProp(p, 0x06DD, 0x06DF, GENX_NAMECHAR); + rangeProp(p, 0x06E0, 0x06E4, GENX_NAMECHAR); + rangeProp(p, 0x06E7, 0x06E8, GENX_NAMECHAR); + rangeProp(p, 0x06EA, 0x06ED, GENX_NAMECHAR); + rangeProp(p, 0x0901, 0x0903, GENX_NAMECHAR); + charProp(p, 0x093C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x093E, 0x094C, GENX_NAMECHAR); + charProp(p, 0x094D, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0951, 0x0954, GENX_NAMECHAR); + rangeProp(p, 0x0962, 0x0963, GENX_NAMECHAR); + rangeProp(p, 0x0981, 0x0983, GENX_NAMECHAR); + charProp(p, 0x09BC, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x09BE, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x09BF, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x09C0, 0x09C4, GENX_NAMECHAR); + rangeProp(p, 0x09C7, 0x09C8, GENX_NAMECHAR); + rangeProp(p, 0x09CB, 0x09CD, GENX_NAMECHAR); + charProp(p, 0x09D7, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x09E2, 0x09E3, GENX_NAMECHAR); + charProp(p, 0x0A02, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0A3C, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0A3E, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0A3F, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0A40, 0x0A42, GENX_NAMECHAR); + rangeProp(p, 0x0A47, 0x0A48, GENX_NAMECHAR); + rangeProp(p, 0x0A4B, 0x0A4D, GENX_NAMECHAR); + rangeProp(p, 0x0A70, 0x0A71, GENX_NAMECHAR); + rangeProp(p, 0x0A81, 0x0A83, GENX_NAMECHAR); + charProp(p, 0x0ABC, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0ABE, 0x0AC5, GENX_NAMECHAR); + rangeProp(p, 0x0AC7, 0x0AC9, GENX_NAMECHAR); + rangeProp(p, 0x0ACB, 0x0ACD, GENX_NAMECHAR); + rangeProp(p, 0x0B01, 0x0B03, GENX_NAMECHAR); + charProp(p, 0x0B3C, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0B3E, 0x0B43, GENX_NAMECHAR); + rangeProp(p, 0x0B47, 0x0B48, GENX_NAMECHAR); + rangeProp(p, 0x0B4B, 0x0B4D, GENX_NAMECHAR); + rangeProp(p, 0x0B56, 0x0B57, GENX_NAMECHAR); + rangeProp(p, 0x0B82, 0x0B83, GENX_NAMECHAR); + rangeProp(p, 0x0BBE, 0x0BC2, GENX_NAMECHAR); + rangeProp(p, 0x0BC6, 0x0BC8, GENX_NAMECHAR); + rangeProp(p, 0x0BCA, 0x0BCD, GENX_NAMECHAR); + charProp(p, 0x0BD7, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0C01, 0x0C03, GENX_NAMECHAR); + rangeProp(p, 0x0C3E, 0x0C44, GENX_NAMECHAR); + rangeProp(p, 0x0C46, 0x0C48, GENX_NAMECHAR); + rangeProp(p, 0x0C4A, 0x0C4D, GENX_NAMECHAR); + rangeProp(p, 0x0C55, 0x0C56, GENX_NAMECHAR); + rangeProp(p, 0x0C82, 0x0C83, GENX_NAMECHAR); + rangeProp(p, 0x0CBE, 0x0CC4, GENX_NAMECHAR); + rangeProp(p, 0x0CC6, 0x0CC8, GENX_NAMECHAR); + rangeProp(p, 0x0CCA, 0x0CCD, GENX_NAMECHAR); + rangeProp(p, 0x0CD5, 0x0CD6, GENX_NAMECHAR); + rangeProp(p, 0x0D02, 0x0D03, GENX_NAMECHAR); + rangeProp(p, 0x0D3E, 0x0D43, GENX_NAMECHAR); + rangeProp(p, 0x0D46, 0x0D48, GENX_NAMECHAR); + rangeProp(p, 0x0D4A, 0x0D4D, GENX_NAMECHAR); + charProp(p, 0x0D57, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0E31, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0E34, 0x0E3A, GENX_NAMECHAR); + rangeProp(p, 0x0E47, 0x0E4E, GENX_NAMECHAR); + charProp(p, 0x0EB1, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0EB4, 0x0EB9, GENX_NAMECHAR); + rangeProp(p, 0x0EBB, 0x0EBC, GENX_NAMECHAR); + rangeProp(p, 0x0EC8, 0x0ECD, GENX_NAMECHAR); + rangeProp(p, 0x0F18, 0x0F19, GENX_NAMECHAR); + charProp(p, 0x0F35, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0F37, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0F39, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0F3E, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0F3F, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0F71, 0x0F84, GENX_NAMECHAR); + rangeProp(p, 0x0F86, 0x0F8B, GENX_NAMECHAR); + rangeProp(p, 0x0F90, 0x0F95, GENX_NAMECHAR); + charProp(p, 0x0F97, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x0F99, 0x0FAD, GENX_NAMECHAR); + rangeProp(p, 0x0FB1, 0x0FB7, GENX_NAMECHAR); + charProp(p, 0x0FB9, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x20D0, 0x20DC, GENX_NAMECHAR); + charProp(p, 0x20E1, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x302A, 0x302F, GENX_NAMECHAR); + charProp(p, 0x3099, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x309A, GENX_LETTER|GENX_NAMECHAR); + + charProp(p, 0x02D0, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x02D1, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0387, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0640, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0E46, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x0EC6, GENX_LETTER|GENX_NAMECHAR); + charProp(p, 0x3005, GENX_LETTER|GENX_NAMECHAR); + rangeProp(p, 0x3031, 0x3035, GENX_NAMECHAR); + rangeProp(p, 0x309D, 0x309E, GENX_NAMECHAR); + rangeProp(p, 0x30FC, 0x30FE, GENX_NAMECHAR); + +#endif /* GENX_CHAR_TABLE_SIZE == 0x10000 */ +} + diff --git a/xml/details/genx/genx.c b/xml/details/genx/genx.c new file mode 100644 index 0000000..80c2448 --- /dev/null +++ b/xml/details/genx/genx.c @@ -0,0 +1,2324 @@ +/* + * Copyright (c) 2007-2013 Code Synthesis Tools CC. + * Copyright (c) 2004 by Tim Bray and Sun Microsystems. + * + * For copying permission, see the accompanying COPYING file. + */ + +#define GENX_VERSION "cs-1" + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include <xml/details/genx/genx.h> + +#define Boolean int +#define True 1 +#define False 0 +#define STRLEN_XMLNS_COLON 6 + + +/******************************* + * writer state + */ +typedef enum +{ + SEQUENCE_NO_DOC, + SEQUENCE_PRE_DOC, + SEQUENCE_POST_DOC, + SEQUENCE_START_TAG, + SEQUENCE_ATTRIBUTES, + SEQUENCE_START_ATTR, + SEQUENCE_CONTENT +} writerSequence; + +/******************************* + * generic pointer list + */ +typedef struct +{ + genxWriter writer; + int count; + int space; + void * * pointers; +} plist; + +/******************************* + * text collector, for attribute values + */ +typedef struct +{ + utf8 buf; + int used; + int space; +} collector; + +/******************************* + * Structs with opaquely-exposed handles + */ + +/* + * This one's tricky, to handle stacking namespaces + * 'declaration' is the current attribute which would be used to + * declare the currently-effective prefix + * 'defDeclaration' is a appropriate declaration when this is being + * used with the default prefix as passed to genxDeclareNamespace + * baroque is true if this namespace has been used with more than one + * prefix, or is the default namespace but has been unset + */ +struct genxNamespace_rec +{ + genxWriter writer; + utf8 name; + int declCount; + Boolean baroque; + genxAttribute declaration; + genxAttribute defaultDecl; +}; + +struct genxElement_rec +{ + genxWriter writer; + utf8 type; + genxNamespace ns; +}; + +typedef enum +{ + ATTR_NSDECL, + ATTR_NAKED, + ATTR_PREFIXED +} attrType; + +struct genxAttribute_rec +{ + genxWriter writer; + utf8 name; + genxNamespace ns; + collector value; + int provided; /* provided for current element? */ + attrType atype; + genxAttribute next; /* Attribute order chain if not canonical. */ +}; + +/******************************* + * genx's sandbox + */ +struct genxWriter_rec +{ + genxSender * sender; + genxStatus status; + writerSequence sequence; + char xmlChars[GENX_CHAR_TABLE_SIZE]; + void * userData; + int nextPrefix; + utf8 empty; + Boolean defaultNsDeclared; + genxAttribute xmlnsEquals; + genxElement nowStarting; + genxAttribute nowStartingAttr; + plist namespaces; + plist elements; + plist attributes; + plist prefixes; + plist stack; + struct genxAttribute_rec arec; /* Dummy attribute used for lookup. */ + char * etext[100]; + genxAlloc alloc; + genxDealloc dealloc; + + /* Pretty-printing state */ + int ppIndent; + int ppDepth; + Boolean ppSimple; + + /* Canonicalization. */ + Boolean canonical; + + /* Attrbute order when not canonical. */ + genxAttribute firstAttribute; + genxAttribute lastAttribute; +}; + +/******************************* + * Forward declarations + */ +static genxAttribute declareAttribute(genxWriter w, genxNamespace ns, + constUtf8 name, constUtf8 valuestr, + genxStatus * statusP); +static genxStatus addNamespace(genxNamespace ns, constUtf8 prefix); +static genxStatus unsetDefaultNamespace(genxWriter w); +static genxStatus addAttribute(genxAttribute a, constUtf8 valuestr); +void genxSetCharProps(char * p); + +/******************************* + * End of declarations + */ + +/******************************* + * private memory utilities + */ +static void * allocate(genxWriter w, int bytes) +{ + if (w->alloc) + return (void *) (*w->alloc)(w->userData, bytes); + else + return (void *) malloc(bytes); +} + +static void deallocate(genxWriter w, void * data) +{ + if (w->dealloc) + (*w->dealloc)(w->userData, data); + else if (w->alloc == NULL) + free(data); +} + +static utf8 copy(genxWriter w, constUtf8 from) +{ + utf8 temp; + + if ((temp = (utf8) allocate(w, strlen((const char *) from) + 1)) == NULL) + return NULL; + strcpy((char *) temp, (const char *) from); + return temp; +} + +static genxStatus initCollector(genxWriter w, collector * c) +{ + c->space = 100; + if ((c->buf = (utf8) allocate(w, c->space)) == NULL) + return GENX_ALLOC_FAILED; + c->used = 0; + return GENX_SUCCESS; +} + +static genxStatus growCollector(genxWriter w, collector * c, int size) +{ + utf8 newSpace; + + c->space = size * 2; + if ((newSpace = (utf8) allocate(w, c->space)) == NULL) + return GENX_ALLOC_FAILED; + + strncpy((char *) newSpace, (const char *) c->buf, c->used); + newSpace[c->used] = 0; + deallocate(w, c->buf); + c->buf = newSpace; + return GENX_SUCCESS; +} + +static void startCollect(collector * c) +{ + c->used = 0; +} +static void endCollect(collector * c) +{ + c->buf[c->used] = 0; +} + +static genxStatus collectString(genxWriter w, collector * c, constUtf8 string) +{ + int sl = strlen((const char *) string); + + if (sl >= c->space) + if ((w->status = growCollector(w, c, sl)) != GENX_SUCCESS) + return GENX_ALLOC_FAILED; + + strcpy((char *) c->buf, (const char *) string); + return GENX_SUCCESS; +} + +#define collectPiece(w,c,d,size) {if (((c)->used+(size))>=(c)->space){if (((w)->status=growCollector(w,c,(c)->used+(size)))!=GENX_SUCCESS) return (w)->status;}strncpy((char *)(c)->buf+(c)->used,d,size);(c)->used+=size;} + +/******************************* + * private list utilities + */ +static genxStatus initPlist(genxWriter w, plist * pl) +{ + pl->writer = w; + pl->count = 0; + pl->space = 10; + pl->pointers = (void * *) allocate(w, pl->space * sizeof(void *)); + if (pl->pointers == NULL) + return GENX_ALLOC_FAILED; + + return GENX_SUCCESS; +} + +/* + * make room in a plist + */ +static Boolean checkExpand(plist * pl) +{ + void * * newlist; + int i; + + if (pl->count < pl->space) + return True; + + pl->space *= 2; + newlist = (void * *) allocate(pl->writer, pl->space * sizeof(void *)); + if (newlist == NULL) + return False; + for (i = 0; i < pl->count; i++) + newlist[i] = pl->pointers[i]; + deallocate(pl->writer, pl->pointers); + pl->pointers = newlist; + + return True; +} + +/* + * stick something on the end of a plist + */ +static genxStatus listAppend(plist * pl, void * pointer) +{ + if (!checkExpand(pl)) + return GENX_ALLOC_FAILED; + + pl->pointers[pl->count++] = pointer; + return GENX_SUCCESS; +} + +/* + * insert in place, shuffling up + */ +static genxStatus listInsert(plist * pl, void * pointer, int at) +{ + int i; + + if (!checkExpand(pl)) + return GENX_ALLOC_FAILED; + + for (i = pl->count; i > at; i--) + pl->pointers[i] = pl->pointers[i - 1]; + pl->count++; + + pl->pointers[at] = pointer; + return GENX_SUCCESS; +} + +/******************************* + * list lookups + */ + +static genxNamespace findNamespace(plist * pl, constUtf8 uri) +{ + int i; + genxNamespace * nn = (genxNamespace *) pl->pointers; + + for (i = 0; i < pl->count; i++) + if (strcmp((char *) uri, (const char *) nn[i]->name) == 0) + return nn[i]; + + return NULL; +} + +static genxElement findElement(plist * pl, constUtf8 xmlns, constUtf8 type) +{ + int i; + genxElement * ee = (genxElement *) pl->pointers; + + for (i = 0; i < pl->count; i++) + { + if (xmlns == NULL) + { + if (ee[i]->ns == NULL && strcmp((const char *) type, + (const char *) ee[i]->type) == 0) + return ee[i]; + } + else + { + if (ee[i]->ns != NULL && + strcmp((const char *) xmlns, (const char *) ee[i]->ns->name) == 0 && + strcmp((const char *) type, (const char *) ee[i]->type) == 0) + return ee[i]; + } + } + + return NULL; +} + +/* + * store & intern a prefix, after giving it the + * "xmlns:" prefix. Don't allow storing the same one twice unless 'force' + * is set. + */ +static utf8 storePrefix(genxWriter w, constUtf8 prefix, Boolean force) +{ + int high, low; + utf8 * pp = (utf8 *) w->prefixes.pointers; + unsigned char buf[1024]; + + if (prefix[0] == 0) + prefix = (utf8) "xmlns"; + else + { + sprintf((char *) buf, "xmlns:%s", prefix); + prefix = buf; + } + + high = w->prefixes.count; low = -1; + while (high - low > 1) + { + int probe = (high + low) / 2; + if (strcmp((const char *) prefix, (const char *) pp[probe]) < 0) + high = probe; + else + low = probe; + } + + /* already there? */ + if (low != -1 && strcmp((const char *) prefix, (const char *) pp[low]) == 0) + { + if (force) + return pp[low]; + + w->status = GENX_DUPLICATE_PREFIX; + return NULL; + } + + /* copy & insert */ + if ((prefix = copy(w, prefix)) == NULL) + { + w->status = GENX_ALLOC_FAILED; + return NULL; + } + + w->status = listInsert(&w->prefixes, (void *) prefix, high); + if (w->status != GENX_SUCCESS) + return NULL; + + return (utf8) prefix; +} + +/******************************* + * UTF8 bit-banging + */ + +/* + * Retrieve the character pointed at, and advance the pointer; return -1 on + * error + */ +int genxNextUnicodeChar(constUtf8 * sp) +{ + utf8 s = (utf8) *sp; + int c; + + if (*s == 0) + return -1; + + if (*s < 0x80) + c = *s++; + + /* all this encoding sanity-checking taken from section 3.10 of Unicode 4 */ + else if (*s < 0xc2) + goto malformed; + + /* 2-byte encodings, first byte c2 .. df */ + else if (*s < 0xe0) + { + c = (*s++ & 0x1f) << 6; + + /* + * for this common idiom, if ((c & 0xc0) != 0x80) is slightly faster + * on MacOS (PPC) + */ + if (*s < 0x80 || *s > 0xbf) + goto malformed; + + c |= *s++ & 0x3f; + } + + /* 3-byte encodings, first byte e0 .. ef */ + else if (*s < 0xf0) + { + int b0 = *s; + c = (*s++ & 0x0f) << 12; + + if ((b0 == 0xe0 && (*s < 0xa0 || *s > 0xbf)) || + (b0 < 0xed && (*s < 0x80 || *s > 0xbf)) || + (b0 == 0xed && (*s < 0x80 || *s > 0x9f)) || + (b0 > 0xed && (*s < 0x80 || *s > 0xbf))) + goto malformed; + + c |= (*s++ & 0x3f) << 6; + + if (*s < 0x80 || *s > 0xbf) + goto malformed; + + c |= *s++ & 0x3f; + } + + /* 4-byte encodings, first byte f0 .. f4 */ + else if (*s < 0xf5) + { + int b0 = *s; + c = (*s++ & 0x07) << 18; + + if ((b0 == 0xf0 && (*s < 0x90 || *s > 0xbf)) || + (b0 < 0xf4 && (*s < 0x80 || *s > 0xbf)) || + (b0 >= 0xf4 && (*s < 0x80 || *s > 0x8f))) + goto malformed; + + c |= (*s++ & 0x3f) << 12; + + if (*s < 0x80 || *s > 0xbf) + goto malformed; + + c |= (*s++ & 0x3f) << 6; + + if (*s < 0x80 || *s > 0xbf) + goto malformed; + + c |= *s++ & 0x3f; + } + else + goto malformed; + + *sp = s; + return c; + + /* + * this is needed by scrubText, which wants to get the pointer moved + * past the problem area. + */ +malformed: + if (*s) + ++s; + *sp = s; + return -1; +} + +static Boolean isXMLChar(genxWriter w, int c) +{ + if (c < 0) + return False; + else if (c < GENX_CHAR_TABLE_SIZE) + return (int) w->xmlChars[c]; + else + return (c <= 0x10ffff); +} + +static Boolean isLetter(genxWriter w, int c) +{ + if (c < 0 || c > 0xffff) + return False; + else + { +#if GENX_CHAR_TABLE_SIZE == 0x10000 + return w->xmlChars[c] & GENX_LETTER; +#else + return c < GENX_CHAR_TABLE_SIZE ? (w->xmlChars[c] & GENX_LETTER) : True; +#endif + } +} + +static Boolean isNameChar(genxWriter w, int c) +{ + if (c < 0 || c > 0xffff) + return False; + else + { +#if GENX_CHAR_TABLE_SIZE == 0x10000 + return w->xmlChars[c] & GENX_NAMECHAR; +#else + return c < GENX_CHAR_TABLE_SIZE ? (w->xmlChars[c] & GENX_NAMECHAR) : True; +#endif + } +} + +/******************************* + * Constructors, setters/getters + */ + +/* + * Construct a new genxWriter + */ +genxWriter genxNew(genxAlloc alloc, genxDealloc dealloc, void * userData) +{ + genxWriter w; + genxNamespace xml; + + if (alloc) + w = (genxWriter) (*alloc)(userData, sizeof(struct genxWriter_rec)); + else + w = (genxWriter) malloc(sizeof(struct genxWriter_rec)); + + if (w == NULL) + return NULL; + + w->status = GENX_SUCCESS; + w->alloc = alloc; + w->dealloc = dealloc; + w->userData = userData; + w->sequence = SEQUENCE_NO_DOC; + + if (initPlist(w, &w->namespaces) != GENX_SUCCESS || + initPlist(w, &w->elements) != GENX_SUCCESS || + initPlist(w, &w->attributes) != GENX_SUCCESS || + initPlist(w, &w->prefixes) != GENX_SUCCESS || + initPlist(w, &w->stack) != GENX_SUCCESS) + return NULL; + + if ((w->status = initCollector(w, &w->arec.value)) != GENX_SUCCESS) + return NULL; + + if ((w->empty = copy(w, (utf8) "")) == NULL) + { + w->status = GENX_ALLOC_FAILED; + return NULL; + } + + w->xmlnsEquals = declareAttribute(w, NULL, (utf8) "xmlns", NULL, &w->status); + if (w->xmlnsEquals == NULL || w->status != GENX_SUCCESS) + return NULL; + w->defaultNsDeclared = False; + + w->nextPrefix = 1; + + genxSetCharProps(w->xmlChars); + + w->etext[GENX_SUCCESS] = "success"; + w->etext[GENX_BAD_UTF8] = "invalid UTF-8"; + w->etext[GENX_NON_XML_CHARACTER] = "non-XML character"; + w->etext[GENX_BAD_NAME] = "invalid name"; + w->etext[GENX_ALLOC_FAILED] = "memory allocation failed"; + w->etext[GENX_BAD_NAMESPACE_NAME] = "invalid namespace name"; + w->etext[GENX_INTERNAL_ERROR] = "internal error"; + w->etext[GENX_DUPLICATE_PREFIX] = "duplicate prefix"; + w->etext[GENX_SEQUENCE_ERROR] = "call out of sequence"; + w->etext[GENX_NO_START_TAG] = "no start tag for end element call"; + w->etext[GENX_IO_ERROR] = "io error"; + w->etext[GENX_MISSING_VALUE] = "missing attribute value"; + w->etext[GENX_MALFORMED_COMMENT] = "malformed comment body"; + w->etext[GENX_MALFORMED_PI] = "?> in PI"; + w->etext[GENX_XML_PI_TARGET] = "target of PI matches [xX][mM][lL]"; + w->etext[GENX_DUPLICATE_ATTRIBUTE] = "duplicate attribute"; + w->etext[GENX_ATTRIBUTE_IN_DEFAULT_NAMESPACE] = + "attribute is default namespace"; + w->etext[GENX_DUPLICATE_NAMESPACE] = + "namespace declared twice with different prefixes"; + w->etext[GENX_BAD_DEFAULT_DECLARATION] = + "default namespace declared on an element which is not in a namespace"; + + /* the xml: namespace is pre-wired */ + xml = genxDeclareNamespace(w, (utf8) "http://www.w3.org/XML/1998/namespace", + (utf8) "xml", &w->status); + if (xml == NULL) + return NULL; + xml->declCount = 1; + xml->declaration = xml->defaultDecl; + + w->ppIndent = 0; /* Pretty-printing is disabled by default. */ + w->canonical = False; /* No canonicalization by default. */ + + w->firstAttribute = NULL; + w->lastAttribute = NULL; + return w; +} + +genxStatus genxReset (genxWriter w) +{ + int i; + + /* Clean up the stack. */ + w->stack.count = 0; + + /* Reset namespace declaration counts. The first entry is the pre-wired + xml namespace. */ + ((genxNamespace) w->namespaces.pointers[0])->declCount = 1; + + for (i = 1; i < w->namespaces.count; i++) + { + ((genxNamespace) w->namespaces.pointers[i])->declCount = 0; + ((genxNamespace) w->namespaces.pointers[i])->baroque = False; + } + + /* Clear provided attributes. */ + for (i = 0; i < w->attributes.count; i++) + ((genxAttribute) w->attributes.pointers[i])->provided = False; + + /* Clear attribute list. */ + if (!w->canonical) + { + while (w->firstAttribute != NULL) + { + genxAttribute t = w->firstAttribute->next; + w->firstAttribute->next = NULL; + w->firstAttribute = t; + } + + w->lastAttribute = NULL; + } + + w->status = GENX_SUCCESS; + w->sequence = SEQUENCE_NO_DOC; + + return w->status; +} + + +/* + * get/set userData + */ +void genxSetUserData(genxWriter w, void * userData) +{ + w->userData = userData; +} +void * genxGetUserData(genxWriter w) +{ + return w->userData; +} + +/* + * get/set pretty-printing + */ +genxStatus genxSetPrettyPrint(genxWriter w, int ind) +{ + if (w->sequence == SEQUENCE_NO_DOC) + w->ppIndent = ind; + else + w->status = GENX_SEQUENCE_ERROR; + + return w->status; +} + +int genxGetPrettyPrint(genxWriter w) +{ + return w->ppIndent; +} + +/* + * get/set canonicalization. + */ +genxStatus genxSetCanonical(genxWriter w, int flag) +{ + if (w->sequence == SEQUENCE_NO_DOC) + w->canonical = flag; + else + w->status = GENX_SEQUENCE_ERROR; + + return w->status; +} + +int genxGetCanonical(genxWriter w) +{ + return w->canonical; +} + +/* + * get/set allocator + */ +void genxSetAlloc(genxWriter w, genxAlloc alloc) +{ + w->alloc = alloc; +} + +void genxSetDealloc(genxWriter w, genxDealloc dealloc) +{ + w->dealloc = dealloc; +} + +genxAlloc genxGetAlloc(genxWriter w) +{ + return w->alloc; +} + +genxDealloc genxGetDealloc(genxWriter w) +{ + return w->dealloc; +} + +/* + * Clean up + */ +void genxDispose(genxWriter w) +{ + int i; + genxNamespace * nn = (genxNamespace *) w->namespaces.pointers; + genxElement * ee = (genxElement *) w->elements.pointers; + genxAttribute * aa = (genxAttribute *) w->attributes.pointers; + utf8 * pp = (utf8 *) w->prefixes.pointers; + + for (i = 0; i < w->namespaces.count; i++) + { + deallocate(w, nn[i]->name); + deallocate(w, nn[i]); + } + + for (i = 0; i < w->elements.count; i++) + { + deallocate(w, ee[i]->type); + deallocate(w, ee[i]); + } + + for (i = 0; i < w->attributes.count; i++) + { + deallocate(w, aa[i]->name); + deallocate(w, aa[i]->value.buf); + deallocate(w, aa[i]); + } + + for(i = 0; i < w->prefixes.count; i++) + deallocate(w, pp[i]); + + deallocate(w, w->namespaces.pointers); + deallocate(w, w->elements.pointers); + deallocate(w, w->attributes.pointers); + deallocate(w, w->prefixes.pointers); + deallocate(w, w->stack.pointers); + + deallocate(w, w->arec.value.buf); + + deallocate(w, w->empty); + + /* how Oscar dealt with Igli */ + deallocate(w, w); +} + +/******************************* + * External utility routines + */ + +/* + * scan a buffer and report problems with UTF-8 encoding or non-XML characters + */ +genxStatus genxCheckText(genxWriter w, constUtf8 s) +{ + while (*s) + { + int c = genxNextUnicodeChar(&s); + if (c == -1) + return GENX_BAD_UTF8; + + if (!isXMLChar(w, c)) + return GENX_NON_XML_CHARACTER; + } + return GENX_SUCCESS; +} + +/* + * Purify some text + */ +int genxScrubText(genxWriter w, constUtf8 in, utf8 out) +{ + int problems = 0; + constUtf8 last = in; + + while (*in) + { + int c = genxNextUnicodeChar(&in); + if (c == -1) + { + problems++; + last = in; + continue; + } + + if (!isXMLChar(w, c)) + { + problems++; + last = in; + continue; + } + + while (last < in) + *out++ = *last++; + } + *out = 0; + return problems; +} + +/* + * check one character + */ +int genxCharClass(genxWriter w, int c) +{ + int ret = 0; + + if (isXMLChar(w, c)) + ret |= GENX_XML_CHAR; + if (isNameChar(w, c)) + ret |= GENX_NAMECHAR; + if (isLetter(w, c)) + ret |= GENX_LETTER; + return ret; +} + +static genxStatus checkNCName(genxWriter w, constUtf8 name) +{ + int c; + + if (name == NULL || *name == 0) + return GENX_BAD_NAME; + + c = genxNextUnicodeChar(&name); + if (!isLetter(w, c) && c != ':' && c != '_') + return GENX_BAD_NAME; + + while (*name) + { + c = genxNextUnicodeChar(&name); + if (c == -1) + return GENX_BAD_UTF8; + if (!isNameChar(w, c)) + return GENX_BAD_NAME; + } + return GENX_SUCCESS; +} + +char * genxGetErrorMessage(genxWriter w, genxStatus status) +{ + return w->etext[status]; +} +char * genxLastErrorMessage(genxWriter w) +{ + return w->etext[w->status]; +} + +/******************************* + * Declarations: namespace/element/attribute + */ + +/* + * DeclareNamespace - by far the most complex routine in Genx + */ +genxNamespace genxDeclareNamespace(genxWriter w, constUtf8 uri, + constUtf8 defaultPref, + genxStatus * statusP) +{ + genxNamespace ns; + genxAttribute defaultDecl; + unsigned char newPrefix[100]; + + if (uri == NULL || uri[0] == 0) + { + w->status = GENX_BAD_NAMESPACE_NAME; + goto busted; + } + + if ((w->status = genxCheckText(w, uri)) != GENX_SUCCESS) + goto busted; + + /* if a prefix is provided, it has to be an NCname */ + if (defaultPref != NULL && defaultPref[0] != 0 && + (w->status = checkNCName(w, defaultPref)) != GENX_SUCCESS) + goto busted; + + /* previously declared? */ + if ((ns = findNamespace(&w->namespaces, uri))) + { + /* just a lookup, really */ + if ((defaultPref == NULL) || + (defaultPref[0] == 0 && ns->defaultDecl == w->xmlnsEquals) || + (strcmp((const char *) ns->defaultDecl->name + STRLEN_XMLNS_COLON, + (const char *) defaultPref) == 0)) + { + w->status = *statusP = GENX_SUCCESS; + return ns; + } + } + + /* wasn't already declared */ + else + { + + /* make a default prefix if none provided */ + if (defaultPref == NULL) + { + sprintf((char *) newPrefix, "g%d", w->nextPrefix++); + defaultPref = newPrefix; + } + + ns = (genxNamespace) allocate(w, sizeof(struct genxNamespace_rec)); + if (ns == NULL) + { + w->status = GENX_ALLOC_FAILED; + goto busted; + } + ns->writer = w; + ns->baroque = False; + + if ((ns->name = copy(w, uri)) == NULL) + { + w->status = GENX_ALLOC_FAILED; + goto busted; + } + + if ((w->status = listAppend(&w->namespaces, ns)) != GENX_SUCCESS) + goto busted; + ns->defaultDecl = ns->declaration = NULL; + ns->declCount = 0; + } + + if (defaultPref[0] == 0) + { + if (w->defaultNsDeclared) + { + w->status = GENX_DUPLICATE_PREFIX; + goto busted; + } + defaultDecl = w->xmlnsEquals; + w->defaultNsDeclared = True; + } + else + { + /* this catches dupes too */ + if ((defaultPref = storePrefix(w, defaultPref, False)) == NULL) + goto busted; + + defaultDecl = declareAttribute(w, NULL, defaultPref, ns->name, statusP); + if (defaultDecl == NULL || *statusP != GENX_SUCCESS) + { + w->status = *statusP; + return NULL; + } + } + + if (ns->defaultDecl != NULL && defaultDecl != ns->defaultDecl) + ns->baroque = True; + ns->defaultDecl = defaultDecl; + + *statusP = GENX_SUCCESS; + return ns; + +busted: + *statusP = w->status; + return NULL; +} + +/* + * get namespace prefix + */ +utf8 genxGetNamespacePrefix(genxNamespace ns) +{ + if (ns->declaration == NULL) + return NULL; + + if (ns->declaration == ns->writer->xmlnsEquals) + return ns->writer->empty; + + return ns->declaration->name + STRLEN_XMLNS_COLON; +} + +/* + * DeclareElement - see genx.h for details + */ +genxElement genxDeclareElement(genxWriter w, + genxNamespace ns, constUtf8 type, + genxStatus * statusP) +{ + genxElement old; + genxElement el; + + if ((w->status = checkNCName(w, type)) != GENX_SUCCESS) + { + *statusP = w->status; + return NULL; + } + + /* already declared? */ + old = findElement(&w->elements, (ns == NULL) ? NULL : ns->name, type); + if (old) + return old; + + if ((el = (genxElement) allocate(w, sizeof(struct genxElement_rec))) == NULL) + { + w->status = *statusP = GENX_ALLOC_FAILED; + return NULL; + } + + el->writer = w; + el->ns = ns; + if ((el->type = copy(w, type)) == NULL) + { + w->status = *statusP = GENX_ALLOC_FAILED; + return NULL; + } + + if ((w->status = listAppend(&w->elements, el)) != GENX_SUCCESS) + { + *statusP = w->status; + return NULL; + } + + *statusP = GENX_SUCCESS; + return el; +} + +/* + * C14n ordering for attributes: + * - first, namespace declarations by the prefix being declared + * - second, unprefixed attributes by attr name + * - third, prefixed attrs by ns uri then local part + */ +static int orderAttributes(genxAttribute a1, genxAttribute a2) +{ + if (a1->atype == a2->atype) + { + if (a1->atype == ATTR_PREFIXED && a1->ns != a2->ns) + return strcmp((const char *) a1->ns->name, (const char *) a2->ns->name); + else + return strcmp((const char *) a1->name, (const char *) a2->name); + } + + else if (a1->atype == ATTR_NSDECL) + return -1; + + else if (a1->atype == ATTR_NAKED) + { + if (a2->atype == ATTR_NSDECL) + return 1; + else + return -1; + } + + else + return 1; +} + +/* + * internal declare-attribute. This one allows colonized values for + * names, so that you can declare xmlns:-type attributes + */ +static genxAttribute declareAttribute(genxWriter w, genxNamespace ns, + constUtf8 name, constUtf8 valuestr, + genxStatus * statusP) +{ + int high, low; + genxAttribute * aa = (genxAttribute *) w->attributes.pointers; + genxAttribute a; + + w->arec.ns = ns; + w->arec.name = (utf8) name; + + if (ns) + w->arec.atype = ATTR_PREFIXED; + else if (strncmp((const char *) name, "xmlns", STRLEN_XMLNS_COLON - 1) == 0) + w->arec.atype = ATTR_NSDECL; + else + w->arec.atype = ATTR_NAKED; + + if (ns && (ns->defaultDecl == w->xmlnsEquals)) + { + w->status = GENX_ATTRIBUTE_IN_DEFAULT_NAMESPACE; + goto busted; + } + + /* attribute list has to be kept sorted per c14n rules */ + high = w->attributes.count; low = -1; + while (high - low > 1) + { + int probe = (high + low) / 2; + if (orderAttributes(&w->arec, aa[probe]) < 0) + high = probe; + else + low = probe; + } + + /* if it was already there */ + if (low != -1 && orderAttributes(&w->arec, aa[low]) == 0) + return aa[low]; + + /* not there, build it */ + a = (genxAttribute) allocate(w, sizeof(struct genxAttribute_rec)); + if (a == NULL) + { + w->status = GENX_ALLOC_FAILED; + goto busted; + } + + a->writer = w; + a->ns = ns; + a->provided = False; + a->atype = w->arec.atype; + a->next = NULL; + + if ((a->name = copy(w, name)) == NULL) + { + w->status = GENX_ALLOC_FAILED; + goto busted; + } + + if ((w->status = initCollector(w, &a->value)) != GENX_SUCCESS) + goto busted; + + if (valuestr) + if ((w->status = collectString(w, &a->value, valuestr)) != GENX_SUCCESS) + goto busted; + + w->status = listInsert(&w->attributes, a, high); + if (w->status != GENX_SUCCESS) + goto busted; + + *statusP = GENX_SUCCESS; + return a; + +busted: + *statusP = w->status; + return NULL; +} + +/* + * genxDeclareAttribute - see genx.h for details + */ +genxAttribute genxDeclareAttribute(genxWriter w, + genxNamespace ns, constUtf8 name, + genxStatus * statusP) +{ + if ((w->status = checkNCName(w, name)) != GENX_SUCCESS) + { + *statusP = w->status; + return NULL; + } + + return declareAttribute(w, ns, name, NULL, statusP); +} + +/******************************* + * I/O + */ +static genxStatus sendx(genxWriter w, constUtf8 s) +{ + if (w->sender) + return (*w->sender->send)(w->userData, s); + else + return GENX_IO_ERROR; +} + +static genxStatus sendxBounded(genxWriter w, constUtf8 start, constUtf8 end) +{ + if (w->sender) + return (*w->sender->sendBounded)(w->userData, start, end); + else + return GENX_IO_ERROR; +} + +#define SendCheck(w,s) if ((w->status=sendx(w,(constUtf8)s))!=GENX_SUCCESS) return w->status + +/******************************* + * XML writing routines. The semantics of the externally-facing ones are + * written up in genx.h. Commentary here is implementation notes and + * for internal routines. + */ + +genxStatus genxStartDocSender(genxWriter w, genxSender * sender) +{ + if (w->sequence != SEQUENCE_NO_DOC) + return w->status = GENX_SEQUENCE_ERROR; + + w->sequence = SEQUENCE_PRE_DOC; + w->sender = sender; + + if (w->ppIndent) + { + w->ppSimple = True; + w->ppDepth = 0; + } + + return GENX_SUCCESS; +} + +/* + * Output new line and indentation. + */ +static genxStatus writeIndentation(genxWriter w) +{ + int i, n; + SendCheck(w, "\n"); + n = w->ppDepth * w->ppIndent; + + for (i = 0; i < n; i++) + SendCheck(w, " "); + + return w->status; +} + +/* + * Output attribute. + */ +static genxStatus writeAttribute(genxAttribute a) +{ + genxWriter w = a->writer; + + if (a->ns && a->ns->baroque && a->ns->declaration == w->xmlnsEquals) + return w->status = GENX_ATTRIBUTE_IN_DEFAULT_NAMESPACE; + + SendCheck(w, " "); + + if (a->ns) + { + SendCheck(w, a->ns->declaration->name + STRLEN_XMLNS_COLON); + SendCheck(w, ":"); + } + + SendCheck(w, a->name); + SendCheck(w, "=\""); + SendCheck(w, a->value.buf); + SendCheck(w, "\""); + + return w->status; +} + +/* + * Write out the attributes we've been gathering up for an element. We save + * them until we've gathered them all so they can be writen in canonical + * order. + * Also, we end the start-tag. + * The trick here is that we keep the attribute list properly sorted as + * we build it, then as each attribute is added, we fill in its value and + * mark the fact that it's been added, in the "provided" field. + */ +static genxStatus writeStartTag(genxWriter w, Boolean close) +{ + int i; + genxAttribute * aa = (genxAttribute *) w->attributes.pointers; + genxElement e = w->nowStarting; + + /* + * make sure the right namespace decls are in effect; + * if they are these might create an error, so ignore it + */ + if (e->ns) + addNamespace(e->ns, NULL); + else + unsetDefaultNamespace(w); + w->status = GENX_SUCCESS; + + if (w->ppIndent) + { + if (w->ppDepth) + if (writeIndentation (w) != GENX_SUCCESS) + return w->status; + + if (!close) + { + w->ppDepth++; + w->ppSimple = True; + } + } + + SendCheck(w, "<"); + if (e->ns && (e->ns->declaration != w->xmlnsEquals)) + { + SendCheck(w, e->ns->declaration->name + STRLEN_XMLNS_COLON); + SendCheck(w, ":"); + } + SendCheck(w, e->type); + + /* If we are canonicalizing, then write sorted attributes. Otherwise + write them in the order specified. */ + if (w->canonical) + { + for (i = 0; i < w->attributes.count; i++) + { + if (aa[i]->provided) + { + if (writeAttribute (aa[i]) != GENX_SUCCESS) + return w->status; + + aa[i]->provided = False; + } + } + } + else + { + /* Keep the chain consistent even if we bail out mid way because of + an error. This way we will still be able to clear it in reset().*/ + while (w->firstAttribute != NULL) + { + genxAttribute t = w->firstAttribute->next; + + if (writeAttribute (w->firstAttribute) != GENX_SUCCESS) + return w->status; + + w->firstAttribute->provided = False; + w->firstAttribute->next = NULL; + w->firstAttribute = t; + } + + w->lastAttribute = NULL; + } + + if (close) + SendCheck(w, "/"); + SendCheck(w, ">"); + return GENX_SUCCESS; +} + +/* + * internal clear-er; no sequence checking + */ +static genxStatus unsetDefaultNamespace(genxWriter w) +{ + int i; + Boolean found = False; + + /* don't put it in if not needed */ + i = w->stack.count - 1; + while (found == False && i > 0) + { + while (w->stack.pointers[i] != NULL) + { + genxAttribute decl = (genxAttribute) w->stack.pointers[i--]; + genxNamespace ns = (genxNamespace) w->stack.pointers[i--]; + + /* if already unset */ + if (ns == NULL) + return w->status = GENX_SUCCESS; + + /* + * the default namespace was declared. This namespace now + * becomes baroque + */ + if (decl == w->xmlnsEquals) + { + ns->baroque = True; + found = True; + break; + } + } + i -= 2; + } + + if (!found) + return GENX_SUCCESS; + + /* + * push a signal on the stack + */ + if ((w->status = listAppend(&w->stack, NULL)) != GENX_SUCCESS) + return w->status; + w->status = listAppend(&w->stack, w->xmlnsEquals); + if (w->status != GENX_SUCCESS) + return w->status; + + /* add the xmlns= attribute, it must be the first one */ + return addAttribute(w->xmlnsEquals, w->empty); +} + +/* + * clear the default namespace declaration + */ +genxStatus genxUnsetDefaultNamespace(genxWriter w) +{ + + /* can only do this while in start-tag mode */ + if (w->sequence != SEQUENCE_START_TAG) + return w->status = GENX_SEQUENCE_ERROR; + + return unsetDefaultNamespace(w); +} + +genxStatus genxStartElement(genxElement e) +{ + genxWriter w = e->writer; + + switch (w->sequence) + { + case SEQUENCE_NO_DOC: + case SEQUENCE_POST_DOC: + case SEQUENCE_START_ATTR: + return w->status = GENX_SEQUENCE_ERROR; + case SEQUENCE_START_TAG: + case SEQUENCE_ATTRIBUTES: + if ((w->status = writeStartTag(w, False)) != GENX_SUCCESS) + return w->status; + break; + case SEQUENCE_PRE_DOC: + case SEQUENCE_CONTENT: + break; + } + + w->sequence = SEQUENCE_START_TAG; + + /* + * push the stack. We push a NULL after a pointer to this element + * because the stack will also contain pointers to the namespace + * attributes that got declared here, so we can keep track of what's + * in effect. I.e. a single stack entry consists logically of a pointer + * to an element object, a NULL, then zero or more pairs of pointers to + * namespace objects/declarations + */ + if ((w->status = listAppend(&w->stack, e)) != GENX_SUCCESS) + return w->status; + if ((w->status = listAppend(&w->stack, NULL)) != GENX_SUCCESS) + return w->status; + + w->nowStarting = e; + + return GENX_SUCCESS; +} + +/* + * internal namespace adder; no sequence checking + */ +static genxStatus addNamespace(genxNamespace ns, constUtf8 prefix) +{ + genxWriter w = ns->writer; + genxAttribute decl; + int i; + genxElement e; + + /* + * first, we'll find the declaring attribute + */ + if (prefix == NULL) + decl = ns->defaultDecl; + else + { + if (prefix[0] == 0) + decl = w->xmlnsEquals; + else + { + if ((prefix = storePrefix(w, prefix, True)) == NULL) + return w->status; + decl = declareAttribute(w, NULL, prefix, ns->name, &w->status); + if (decl == NULL || w->status != GENX_SUCCESS) + return w->status; + } + } + + if (decl != ns->defaultDecl) + ns->baroque = True; + + /* + * avoid doing anything if this namespace is already declared. If + * they've shown good taste we can do this cheaply + */ + if (!ns->baroque) + { + if (ns->declCount > 0) + return w->status = GENX_SUCCESS; + } + else + { + + /* + * First, we'll run all the way up the stack to see if there is + * another declaration for this namespace/prefix in scope, in which + * case it's a no-op; or, if there's another declaration for this + * prefix on another namespace, in which case we have to over-ride + */ + i = w->stack.count - 1; + while (i > 0) + { + while (w->stack.pointers[i] != NULL) + { + genxAttribute otherDecl = (genxAttribute) w->stack.pointers[i--]; + genxNamespace otherNs = (genxNamespace) w->stack.pointers[i--]; + + if (ns == otherNs) + { + if (decl == otherDecl) + return w->status = GENX_SUCCESS; + else + { + i = 0; + break; + } + } + else + { + /* different namespace, same prefix? */ + if (decl == otherDecl) + { + i = 0; + break; + } + } + } + i -= 2; + } + } + + /* + * If this namespace is already declared on + * this element (with different prefix/decl) which is an error. + */ + i = w->stack.count - 1; + while (w->stack.pointers[i] != NULL) + { + genxNamespace otherNs; + i--; /* don't need declaration */ + otherNs = (genxNamespace) w->stack.pointers[i--]; + + if (ns == otherNs) + return w->status = GENX_DUPLICATE_NAMESPACE; + } + + /* move pointer from NULL to element */ + --i; + + /* + * It's also an error if this is a default-namespace declaration and the + * element is in no namespace. + */ + e = (genxElement) w->stack.pointers[i]; + if (e->ns == NULL && decl == w->xmlnsEquals) + return w->status = GENX_BAD_DEFAULT_DECLARATION; + + if ((w->status = listAppend(&w->stack, ns)) != GENX_SUCCESS) + return w->status; + if ((w->status = listAppend(&w->stack, decl)) != GENX_SUCCESS) + return w->status; + + ns->declaration = decl; + ns->declCount++; + return addAttribute(decl, ns->name); +} + +/* + * Add a namespace declaration + */ +genxStatus genxAddNamespace(genxNamespace ns, constUtf8 prefix) +{ + if (ns->writer->sequence != SEQUENCE_START_TAG) + return ns->writer->status = GENX_SEQUENCE_ERROR; + + return addNamespace(ns, prefix); +} + +/* + * Private attribute-adding code + * most of the work here is normalizing the value, which is the same + * as regular normalization except for " is replaced by """ + */ +static genxStatus collectAttributeValue (genxWriter w, collector* value, + constUtf8 start, constUtf8 end) +{ + /* If end is NULL then the length of the value is unknown and + the value is 0-terminated. */ + + utf8 last = (utf8) start; + + while (end != NULL ? start < end : *start) + { + int c = genxNextUnicodeChar(&start); + + if (c == -1) + return w->status = GENX_BAD_UTF8; + + if (!isXMLChar(w, c)) + return w->status = GENX_NON_XML_CHARACTER; + + switch(c) + { + case 9: + collectPiece(w, value, "	", 5); + break; + case 0xa: + collectPiece(w, value, "
", 5); + break; + case 0xd: + collectPiece(w, value, "
", 5); + break; + case '"': + collectPiece(w, value, """, 6); + break; + case '<': + collectPiece(w, value, "<", 4); + break; + case '&': + collectPiece(w, value, "&", 5); + break; + /* + case '>': + collectPiece(w, value, ">", 4); + break; + */ + default: + collectPiece(w, value, (const char *) last, start - last); + break; + } + last = (utf8) start; + } + + return GENX_SUCCESS; +} + +static genxStatus addAttribute(genxAttribute a, constUtf8 valuestr) +{ + genxWriter w = a->writer; + + /* if valuestr not provided, this is an xmlns with a pre-cooked value */ + if (valuestr) + { + startCollect(&a->value); + + if (collectAttributeValue (w, &a->value, valuestr, NULL) != GENX_SUCCESS) + return w->status; + + endCollect(&a->value); + } + + /* now add the namespace attribute; might fail if it's been hand-declared */ + if (a->ns) + addNamespace(a->ns, NULL); + + if (valuestr && a->provided) + return w->status = GENX_DUPLICATE_ATTRIBUTE; + + a->provided = True; + + /* Add the attribute to the ordered list if not canonical. */ + if (!w->canonical) + { + if (w->lastAttribute != NULL) + w->lastAttribute = w->lastAttribute->next = a; + else + w->lastAttribute = w->firstAttribute = a; + } + + return GENX_SUCCESS; +} + +/* + * public attribute adder. + * The only difference is that it doesn't allow a NULL value + */ +genxStatus genxAddAttribute(genxAttribute a, constUtf8 valuestr) +{ + if (a->writer->sequence != SEQUENCE_START_TAG && + a->writer->sequence != SEQUENCE_ATTRIBUTES) + return a->writer->status = GENX_SEQUENCE_ERROR; + a->writer->sequence = SEQUENCE_ATTRIBUTES; + + if (valuestr == NULL) + return a->writer->status = GENX_MISSING_VALUE; + + return addAttribute(a, valuestr); +} + +genxStatus genxStartAttribute(genxAttribute a) +{ + genxWriter w = a->writer; + + if (w->sequence != SEQUENCE_START_TAG && + w->sequence != SEQUENCE_ATTRIBUTES) + return w->status = GENX_SEQUENCE_ERROR; + + w->sequence = SEQUENCE_START_ATTR; + w->nowStartingAttr = a; + + startCollect(&a->value); + + return GENX_SUCCESS; +} + +genxStatus genxEndAttribute(genxWriter w) +{ + genxAttribute a; + + if (w->sequence != SEQUENCE_START_ATTR) + return w->status = GENX_SEQUENCE_ERROR; + + a = w->nowStartingAttr; + w->sequence = SEQUENCE_ATTRIBUTES; + + endCollect(&a->value); + + /* now add the namespace attribute; might fail if it's been hand-declared */ + if (a->ns) + addNamespace(a->ns, NULL); + + if (a->provided) + return w->status = GENX_DUPLICATE_ATTRIBUTE; + + a->provided = True; + + /* Add the attribute to the ordered list if not canonical. */ + if (!w->canonical) + { + if (w->lastAttribute != NULL) + w->lastAttribute = w->lastAttribute->next = a; + else + w->lastAttribute = w->firstAttribute = a; + } + + return GENX_SUCCESS; +} + +genxStatus genxEndElement(genxWriter w) +{ + int i; + Boolean close = True; + + switch (w->sequence) + { + case SEQUENCE_NO_DOC: + case SEQUENCE_PRE_DOC: + case SEQUENCE_POST_DOC: + case SEQUENCE_START_ATTR: + return w->status = GENX_SEQUENCE_ERROR; + case SEQUENCE_START_TAG: + case SEQUENCE_ATTRIBUTES: + if ((w->status = writeStartTag(w, !w->canonical)) != GENX_SUCCESS) + return w->status; + close = w->canonical; + break; + case SEQUENCE_CONTENT: + break; + } + + /* + * Output the closing tag. + */ + if (close) + { + genxElement e; + + /* + * first peek into the stack to find the right namespace declaration + * (if any) so we can properly prefix the end-tag. Have to do this + * before unwinding the stack because that might reset some xmlns + * prefixes to the context in the parent element + */ + for (i = w->stack.count - 1; w->stack.pointers[i] != NULL; i -= 2) + ; + e = (genxElement) w->stack.pointers[--i]; + + if (w->ppIndent) + { + w->ppDepth--; + + if (!w->ppSimple) + if (writeIndentation (w) != GENX_SUCCESS) + return w->status; + } + + SendCheck(w, "</"); + if (e->ns && e->ns->declaration != w->xmlnsEquals) + { + SendCheck(w, e->ns->declaration->name + STRLEN_XMLNS_COLON); + SendCheck(w, ":"); + } + SendCheck(w, e->type); + SendCheck(w, ">"); + } + + if (w->ppIndent) + w->ppSimple = False; + + /* + * pop zero or more namespace declarations, then a null, then the + * start-element declaration off the stack + */ + w->stack.count--; + while (w->stack.pointers[w->stack.count] != NULL) + { + genxNamespace ns = (genxNamespace) w->stack.pointers[--w->stack.count]; + w->stack.count--; /* don't need decl */ + + /* if not a fake unset-default namespace */ + if (ns) + { + /* + * if they've stupidly jammed in their own namespace-prefix + * declarations, we have to go looking to see if there's another + * one in effect + */ + if (ns->baroque) + { + i = w->stack.count; + while (i > 0) + { + while (w->stack.pointers[i] != NULL) + { + genxAttribute otherDecl = (genxAttribute) w->stack.pointers[i--]; + genxNamespace otherNs = (genxNamespace) w->stack.pointers[i--]; + + if (otherNs == ns) + { + ns->declaration = otherDecl; + i = 0; + break; + } + } + + /* skip NULL & element */ + i -= 2; + } + } + ns->declCount--; + if (ns->declCount == 0) + ns->baroque = False; + } + } + + /* pop the NULL */ + --w->stack.count; + if (w->stack.count < 0) + return w->status = GENX_NO_START_TAG; + + if (w->stack.count == 0) + w->sequence = SEQUENCE_POST_DOC; + else + w->sequence = SEQUENCE_CONTENT; + + return GENX_SUCCESS; +} + +/* + * Internal character-adder. It tries to keep the number of sendx() + * calls down by looking at each character but only doing the output + * when it has to escape something; ordinary text gets saved up in + * chunks the start of which is indicated by *breaker. + * c is the character, next points to the UTF8 representing the next + * lastsP indirectly points to the UTF8 representing the + * character, breakerP* indirectly points to the last place genx + * changed the UTF8, e.g. by escaping a '<' + */ +static genxStatus addChar(genxWriter w, int c, constUtf8 next, + constUtf8 * lastsP, constUtf8 * breakerP) +{ + if (c == -1) + return GENX_BAD_UTF8; + + if (!isXMLChar(w, c)) + return GENX_NON_XML_CHARACTER; + + switch(c) + { + case 0xd: + if ((w->status = sendxBounded(w, *breakerP, *lastsP)) != GENX_SUCCESS) + return w->status; + *breakerP = next; + sendx(w, (utf8) "
"); + break; + case '<': + if ((w->status = sendxBounded(w, *breakerP, *lastsP)) != GENX_SUCCESS) + return w->status; + *breakerP = next; + sendx(w, (utf8) "<"); + break; + case '&': + if ((w->status = sendxBounded(w, *breakerP, *lastsP)) != GENX_SUCCESS) + return w->status; + *breakerP = next; + sendx(w, (utf8) "&"); + break; + case '>': + if ((w->status = sendxBounded(w, *breakerP, *lastsP)) != GENX_SUCCESS) + return w->status; + *breakerP = next; + sendx(w, (utf8) ">"); + break; + default: + break; + } + *lastsP = next; + return GENX_SUCCESS; +} + +genxStatus genxAddText(genxWriter w, constUtf8 start) +{ + constUtf8 lasts = start; + constUtf8 breaker = start; + + if (w->sequence == SEQUENCE_START_TAG || + w->sequence == SEQUENCE_ATTRIBUTES) + { + if ((w->status = writeStartTag(w, False)) != GENX_SUCCESS) + return w->status; + w->sequence = SEQUENCE_CONTENT; + } + + if (w->sequence == SEQUENCE_CONTENT) + { + while (*start) + { + int c = genxNextUnicodeChar(&start); + + w->status = addChar(w, c, start, &lasts, &breaker); + if (w->status != GENX_SUCCESS) + return w->status; + } + return sendxBounded(w, breaker, (utf8) start); + } + else if (w->sequence == SEQUENCE_START_ATTR) + { + return collectAttributeValue (w, &w->nowStartingAttr->value, start, NULL); + } + else + return w->status = GENX_SEQUENCE_ERROR; +} + +genxStatus genxAddBoundedText(genxWriter w, constUtf8 start, constUtf8 end) +{ + constUtf8 lasts = start; + constUtf8 breaker = start; + + if (w->sequence == SEQUENCE_START_TAG || + w->sequence == SEQUENCE_ATTRIBUTES) + { + if ((w->status = writeStartTag(w, False)) != GENX_SUCCESS) + return w->status; + w->sequence = SEQUENCE_CONTENT; + } + + if (w->sequence == SEQUENCE_CONTENT) + { + while (start < end) + { + int c = genxNextUnicodeChar(&start); + + w->status = addChar(w, c, (utf8) start, &lasts, &breaker); + if (w->status != GENX_SUCCESS) + return w->status; + } + return sendxBounded(w, breaker, (utf8) start); + } + else if (w->sequence == SEQUENCE_START_ATTR) + { + return collectAttributeValue (w, &w->nowStartingAttr->value, start, end); + } + else + return w->status = GENX_SEQUENCE_ERROR; +} + +genxStatus genxAddCountedText(genxWriter w, constUtf8 start, int byteCount) +{ + utf8 end = (utf8) (start + byteCount); + + return genxAddBoundedText(w, start, end); +} + +genxStatus genxAddCharacter(genxWriter w, int c) +{ + unsigned char cUTF8[10]; + utf8 lasts, breaker, next; + + if (w->sequence == SEQUENCE_START_TAG || + w->sequence == SEQUENCE_ATTRIBUTES) + { + if ((w->status = writeStartTag(w, False)) != GENX_SUCCESS) + return w->status; + w->sequence = SEQUENCE_CONTENT; + } + + if (!isXMLChar(w, c)) + return w->status = GENX_NON_XML_CHARACTER; + + if (w->sequence == SEQUENCE_START_ATTR) + { + int done = 1; + collector* value = &w->nowStartingAttr->value; + + switch(c) + { + case 9: + collectPiece(w, value, "	", 5); + break; + case 0xa: + collectPiece(w, value, "
", 5); + break; + case 0xd: + collectPiece(w, value, "
", 5); + break; + case '"': + collectPiece(w, value, """, 6); + break; + case '<': + collectPiece(w, value, "<", 4); + break; + case '&': + collectPiece(w, value, "&", 5); + break; + /* + case '>': + collectPiece(w, value, ">", 4); + break; + */ + default: + done = 0; + break; + } + + if (done) + return GENX_SUCCESS; + } + + /* make UTF8 representation of character */ + lasts = breaker = next = cUTF8; + + if (c < 0x80) + *next++ = c; + else if (c < 0x800) + { + *next++ = 0xc0 | (c >> 6); + *next++ = 0x80 | (c & 0x3f); + } + else if (c < 0x10000) + { + *next++ = 0xe0 | (c >> 12); + *next++ = 0x80 | ((c & 0xfc0) >> 6); + *next++ = 0x80 | (c & 0x3f); + } + else + { + *next++ = 0xf0 | (c >> 18); + *next++ = 0x80 | ((c & 0x3f000) >> 12); + *next++ = 0x80 | ((c & 0xfc0) >> 6); + *next++ = 0x80 | (c & 0x3f); + } + *next = 0; + + if (w->sequence == SEQUENCE_CONTENT) + { + w->status = + addChar(w, c, next, (constUtf8 *) &lasts, (constUtf8 *) &breaker); + + if (w->status != GENX_SUCCESS) + return w->status; + + return sendxBounded(w, breaker, next); + } + else if (w->sequence == SEQUENCE_START_ATTR) + { + collectPiece(w, &w->nowStartingAttr->value, + (const char *) cUTF8, next - cUTF8); + return GENX_SUCCESS; + } + else + return w->status = GENX_SEQUENCE_ERROR; +} + +genxStatus genxEndDocument(genxWriter w) +{ + if (w->sequence != SEQUENCE_POST_DOC) + return w->status = GENX_SEQUENCE_ERROR; + + /* Write a newline after the closing tag. */ + SendCheck (w, "\n"); + + if ((w->status = (*w->sender->flush)(w->userData)) != GENX_SUCCESS) + return w->status; + + w->sequence = SEQUENCE_NO_DOC; + return GENX_SUCCESS; +} + +genxStatus genxXmlDeclaration(genxWriter w, + constUtf8 ver, + constUtf8 enc, + constUtf8 stl) +{ + if (w->sequence != SEQUENCE_PRE_DOC) + return w->status = GENX_SEQUENCE_ERROR; + + if ((w->status = genxCheckText(w, ver)) != GENX_SUCCESS) + return w->status; + + if (enc != NULL && (w->status = genxCheckText(w, enc)) != GENX_SUCCESS) + return w->status; + + if (stl != NULL && (w->status = genxCheckText(w, stl)) != GENX_SUCCESS) + return w->status; + + SendCheck (w, "<?xml version=\""); + SendCheck (w, ver); + + if (enc != NULL) + { + SendCheck (w, "\" encoding=\""); + SendCheck (w, enc); + } + + if (stl != NULL) + { + SendCheck (w, "\" standalone=\""); + SendCheck (w, stl); + } + + SendCheck (w, "\" ?>\n"); + + return GENX_SUCCESS; +} + +genxStatus genxComment(genxWriter w, constUtf8 text) +{ + int i; + + if (w->sequence == SEQUENCE_NO_DOC || + w->sequence == SEQUENCE_START_ATTR) + return w->status = GENX_SEQUENCE_ERROR; + + if ((w->status = genxCheckText(w, text)) != GENX_SUCCESS) + return w->status; + + /* no leading '-', no trailing '-', no '--' */ + if (text[0] == '-') + return w->status = GENX_MALFORMED_COMMENT; + for (i = 0; text[i]; i++) + if (text[i] == '-' && (text[i + 1] == '-' || text[i + 1] == 0)) + return w->status = GENX_MALFORMED_COMMENT; + + if (w->sequence == SEQUENCE_START_TAG || + w->sequence == SEQUENCE_ATTRIBUTES) + { + if ((w->status = writeStartTag(w, False)) != GENX_SUCCESS) + return w->status; + w->sequence = SEQUENCE_CONTENT; + } + + else if (w->sequence == SEQUENCE_POST_DOC) + if ((w->status = sendx(w, (utf8) "\n")) != GENX_SUCCESS) + return w->status; + + if ((w->status = sendx(w, (utf8) "<!--")) != GENX_SUCCESS) + return w->status; + if ((w->status = sendx(w, (utf8) text)) != GENX_SUCCESS) + return w->status; + if ((w->status = sendx(w, (utf8) "-->")) != GENX_SUCCESS) + return w->status; + + if (w->sequence == SEQUENCE_PRE_DOC) + if ((w->status = sendx(w, (utf8) "\n")) != GENX_SUCCESS) + return w->status; + + return GENX_SUCCESS; +} + +genxStatus genxPI(genxWriter w, constUtf8 target, constUtf8 text) +{ + int i; + + if (w->sequence == SEQUENCE_NO_DOC || + w->sequence == SEQUENCE_START_ATTR) + return w->status = GENX_SEQUENCE_ERROR; + + if ((w->status = genxCheckText(w, target)) != GENX_SUCCESS) + return w->status; + if ((w->status = checkNCName(w, target)) != GENX_SUCCESS) + return w->status; + if ((strlen((const char *) target) >= 3) && + (target[0] == 'x' || target[0] == 'X') && + (target[1] == 'm' || target[1] == 'M') && + (target[2] == 'l' || target[2] == 'L') && + (target[3] == 0)) + return w->status = GENX_XML_PI_TARGET; + + if ((w->status = genxCheckText(w, text)) != GENX_SUCCESS) + return w->status; + + /* no ?> within */ + for (i = 1; text[i]; i++) + if (text[i] == '>' && text[i - 1] == '?') + return w->status = GENX_MALFORMED_PI; + + if (w->sequence == SEQUENCE_START_TAG || + w->sequence == SEQUENCE_ATTRIBUTES) + { + if ((w->status = writeStartTag(w, False)) != GENX_SUCCESS) + return w->status; + w->sequence = SEQUENCE_CONTENT; + } + + else if (w->sequence == SEQUENCE_POST_DOC) + if ((w->status = sendx(w, (utf8) "\n")) != GENX_SUCCESS) + return w->status; + + if ((w->status = sendx(w, (utf8) "<?")) != GENX_SUCCESS) + return w->status; + if ((w->status = sendx(w, target)) != GENX_SUCCESS) + return w->status; + if (text[0]) + { + if ((w->status = sendx(w, (utf8) " ")) != GENX_SUCCESS) + return w->status; + if ((w->status = sendx(w, text)) != GENX_SUCCESS) + return w->status; + } + if ((w->status = sendx(w, (utf8) "?>")) != GENX_SUCCESS) + return w->status; + + if (w->sequence == SEQUENCE_PRE_DOC) + if ((w->status = sendx(w, (utf8) "\n")) != GENX_SUCCESS) + return w->status; + + return GENX_SUCCESS; +} + +/******************************* + * Literal versions of the writing routines + */ +genxStatus genxStartElementLiteral(genxWriter w, + constUtf8 xmlns, constUtf8 type) +{ + genxNamespace ns = NULL; + genxElement e; + + if (xmlns) + { + ns = genxDeclareNamespace(w, xmlns, NULL, &w->status); + if (ns == NULL || w->status != GENX_SUCCESS) + return w->status; + } + e = genxDeclareElement(w, ns, type, &w->status); + if (e == NULL || w->status != GENX_SUCCESS) + return w->status; + + return genxStartElement(e); +} + +genxStatus genxAddAttributeLiteral(genxWriter w, constUtf8 xmlns, + constUtf8 name, constUtf8 value) +{ + genxNamespace ns = NULL; + genxAttribute a; + + if (xmlns) + { + ns = genxDeclareNamespace(w, xmlns, NULL, &w->status); + if (ns == NULL && w->status != GENX_SUCCESS) + return w->status; + } + + a = genxDeclareAttribute(w, ns, name, &w->status); + if (a == NULL || w->status != GENX_SUCCESS) + return w->status; + + return genxAddAttribute(a, value); +} + +genxStatus genxStartAttributeLiteral(genxWriter w, + constUtf8 xmlns, constUtf8 name) +{ + genxNamespace ns = NULL; + genxAttribute a; + + if (xmlns) + { + ns = genxDeclareNamespace(w, xmlns, NULL, &w->status); + if (ns == NULL && w->status != GENX_SUCCESS) + return w->status; + } + + a = genxDeclareAttribute(w, ns, name, &w->status); + if (a == NULL || w->status != GENX_SUCCESS) + return w->status; + + return genxStartAttribute(a); +} + +genxStatus genxAddNamespaceLiteral(genxWriter w, + constUtf8 uri, constUtf8 prefix) +{ + genxNamespace ns = genxDeclareNamespace(w, uri, prefix, &w->status); + if (ns == NULL && w->status != GENX_SUCCESS) + return w->status; + + return genxAddNamespace(ns, NULL); +} + +/* + * return version + */ +char * genxGetVersion() +{ + return GENX_VERSION; +} diff --git a/xml/details/genx/genx.h b/xml/details/genx/genx.h new file mode 100644 index 0000000..d41c4d0 --- /dev/null +++ b/xml/details/genx/genx.h @@ -0,0 +1,341 @@ +/* + * genx - C-callable library for generating XML documents + */ + +/* + * Copyright (c) 2007-2013 Code Synthesis Tools CC. + * Copyright (c) 2004 by Tim Bray and Sun Microsystems. + * + * For copying permission, see the accompanying COPYING file. + */ + +#ifndef GENX_H +#define GENX_H + +#ifdef __cplusplus +extern "C" { +#endif + +/* + * Note on error handling: genx routines mostly return + * GENX_SUCCESS (guaranteed to be zero) in normal circumstances, one of + * these other GENX_ values on a memory allocation or I/O failure or if the + * call would result in non-well-formed output. + * You can associate an error message with one of these codes explicitly + * or with the most recent error using genxGetErrorMessage() and + * genxLastErrorMessage(); see below. + */ +typedef enum +{ + GENX_SUCCESS = 0, + GENX_BAD_UTF8, + GENX_NON_XML_CHARACTER, + GENX_BAD_NAME, + GENX_ALLOC_FAILED, + GENX_BAD_NAMESPACE_NAME, + GENX_INTERNAL_ERROR, + GENX_DUPLICATE_PREFIX, + GENX_SEQUENCE_ERROR, + GENX_NO_START_TAG, + GENX_IO_ERROR, + GENX_MISSING_VALUE, + GENX_MALFORMED_COMMENT, + GENX_XML_PI_TARGET, + GENX_MALFORMED_PI, + GENX_DUPLICATE_ATTRIBUTE, + GENX_ATTRIBUTE_IN_DEFAULT_NAMESPACE, + GENX_DUPLICATE_NAMESPACE, + GENX_BAD_DEFAULT_DECLARATION +} genxStatus; + +/* character types */ +#define GENX_XML_CHAR 1 +#define GENX_LETTER 2 +#define GENX_NAMECHAR 4 + +/* The size of the character table. Valid values are 0x100 (first 255 + chars are checked) and 0x10000 (all chars are checked). */ +#ifndef GENX_CHAR_TABLE_SIZE +# define GENX_CHAR_TABLE_SIZE 0x100 +#endif + +/* a UTF-8 string */ +typedef unsigned char * utf8; +typedef const unsigned char * constUtf8; + +/* + * genx's own types + */ +typedef struct genxWriter_rec * genxWriter; +typedef struct genxNamespace_rec * genxNamespace; +typedef struct genxElement_rec * genxElement; +typedef struct genxAttribute_rec * genxAttribute; + +typedef void * (*genxAlloc) (void * userData, int bytes); +typedef void (*genxDealloc) (void * userData, void* data); + +/* + * Constructors, set/get + */ + +/* + * Create a new writer. For generating multiple XML documents, it's most + * efficient to re-use the same genx object. However, you can only write + * one document at a time with a writer. + * Returns NULL if it fails, which can only be due to an allocation failure. + */ +genxWriter genxNew(genxAlloc alloc, genxDealloc dealloc, void * userData); + +/* + * Reset the writer object back into usable state after an error or + * interruption. + */ +genxStatus genxReset (genxWriter w); + +/* + * Dispose of a writer, freeing all associated memory + */ +void genxDispose(genxWriter w); + +/* + * Set/get + */ + +/* + * The userdata pointer will be passed to memory-allocation + * and I/O callbacks. If not set, genx will pass NULL + */ +void genxSetUserData(genxWriter w, void * userData); +void * genxGetUserData(genxWriter w); + +/* + * Set/get pretty-printing. If indentation is set to 0, then no pretty- + * printing is performed. + */ +genxStatus genxSetPrettyPrint(genxWriter w, int indentation); +int genxGetPrettyPrint(genxWriter w); + +/* + * Set/get canonicalization. If true, then output explicit closing + * tags and sort attributes. Default is false. + */ +genxStatus genxSetCanonical(genxWriter w, int flag); +int genxGetCanonical(genxWriter w); + +/* + * User-provided memory allocator, if desired. For example, if you were + * in an Apache module, you could arrange for genx to use ap_palloc by + * making the pool accessible via the userData call. + * The "dealloc" is to be used to free memory allocated with "alloc". If + * alloc is provided but dealloc is NULL, genx will not attempt to free + * the memory; this would be appropriate in an Apache context. + * If "alloc" is not provided, genx routines use malloc() to allocate memory + */ +void genxSetAlloc(genxWriter w, genxAlloc alloc); +void genxSetDealloc(genxWriter w, genxDealloc dealloc); +genxAlloc genxGetAlloc(genxWriter w); +genxDealloc genxGetDealloc(genxWriter w); + +/* + * Get the prefix associated with a namespace + */ +utf8 genxGetNamespacePrefix(genxNamespace ns); + +/* + * Declaration functions + */ + +/* + * Declare a namespace. The provided prefix is the default but can be + * overridden by genxAddNamespace. If no default prefiix is provided, + * genx will generate one of the form g-%d. + * On error, returns NULL and signals via statusp + */ +genxNamespace genxDeclareNamespace(genxWriter w, + constUtf8 uri, constUtf8 prefix, + genxStatus * statusP); + +/* + * Declare an element + * If something failed, returns NULL and sets the status code via statusP + */ +genxElement genxDeclareElement(genxWriter w, + genxNamespace ns, constUtf8 type, + genxStatus * statusP); + +/* + * Declare an attribute + */ +genxAttribute genxDeclareAttribute(genxWriter w, + genxNamespace ns, + constUtf8 name, genxStatus * statusP); + +/* + * Writing XML + */ + +/* + * Caller-provided I/O package. + * First form is for a null-terminated string. + * for second, if you have s="abcdef" and want to send "abc", you'd call + * sendBounded(userData, s, s + 3) + */ +typedef struct +{ + genxStatus (* send)(void * userData, constUtf8 s); + genxStatus (* sendBounded)(void * userData, constUtf8 start, constUtf8 end); + genxStatus (* flush)(void * userData); +} genxSender; + +genxStatus genxStartDocSender(genxWriter w, genxSender * sender); + +/* + * End a document. Calls "flush" + */ +genxStatus genxEndDocument(genxWriter w); + +/* + * Write XML declaration. If encoding or standalone are NULL, then those + * attributes are omitted. + */ +genxStatus genxXmlDeclaration(genxWriter w, + constUtf8 version, + constUtf8 encoding, + constUtf8 standalone); +/* + * Write a comment + */ +genxStatus genxComment(genxWriter w, constUtf8 text); + +/* + * Write a PI + */ +genxStatus genxPI(genxWriter w, constUtf8 target, constUtf8 text); + +/* + * Start an element + */ +genxStatus genxStartElementLiteral(genxWriter w, + constUtf8 xmlns, constUtf8 type); + +/* + * Start a predeclared element + * - element must have been declared + */ +genxStatus genxStartElement(genxElement e); + +/* + * Write an attribute + */ +genxStatus genxAddAttributeLiteral(genxWriter w, constUtf8 xmlns, + constUtf8 name, constUtf8 value); + +/* + * Start an attribute + */ +genxStatus genxStartAttributeLiteral(genxWriter w, + constUtf8 xmlns, constUtf8 name); + +/* + * Write a predeclared attribute + */ +genxStatus genxAddAttribute(genxAttribute a, constUtf8 value); + +/* + * Start a predeclared attribute + */ +genxStatus genxStartAttribute(genxAttribute a); + +/* + * End an attribute + */ +genxStatus genxEndAttribute(genxWriter w); + +/* + * add a namespace declaration + */ +genxStatus genxAddNamespaceLiteral(genxWriter w, + constUtf8 uri, constUtf8 prefix); + +/* + * add a predefined namespace declaration + */ +genxStatus genxAddNamespace(genxNamespace ns, constUtf8 prefix); + +/* + * Clear default namespace declaration + */ +genxStatus genxUnsetDefaultNamespace(genxWriter w); + +/* + * Write an end tag + */ +genxStatus genxEndElement(genxWriter w); + +/* + * Write some text + * You can't write any text outside the root element, except with + * genxComment and genxPI + */ +genxStatus genxAddText(genxWriter w, constUtf8 start); +genxStatus genxAddCountedText(genxWriter w, constUtf8 start, int byteCount); +genxStatus genxAddBoundedText(genxWriter w, constUtf8 start, constUtf8 end); + +/* + * Write one character. The integer value is the Unicode character + * value, as usually expressed in U+XXXX notation. + */ +genxStatus genxAddCharacter(genxWriter w, int c); + +/* + * Utility routines + */ + +/* + * Return the Unicode character encoded by the UTF-8 pointed-to by the + * argument, and advance the argument past the encoding of the character. + * Returns -1 if the UTF-8 is malformed, in which case advances the + * argument to point at the first byte past the point past the malformed + * ones. + */ +int genxNextUnicodeChar(constUtf8 * sp); + +/* + * Scan a buffer allegedly full of UTF-8 encoded XML characters; return + * one of GENX_SUCCESS, GENX_BAD_UTF8, or GENX_NON_XML_CHARACTER + */ +genxStatus genxCheckText(genxWriter w, constUtf8 s); + +/* + * return character status, the OR of GENX_XML_CHAR, + * GENX_LETTER, and GENX_NAMECHAR + */ +int genxCharClass(genxWriter w, int c); + +/* + * Silently wipe any non-XML characters out of a chunk of text. + * If you call this on a string before you pass it addText or + * addAttribute, you will never get an error from genx unless + * (a) there's a bug in your software, e.g. a malformed element name, or + * (b) there's a memory allocation or I/O error + * The output can never be longer than the input. + * Returns true if any changes were made. + */ +int genxScrubText(genxWriter w, constUtf8 in, utf8 out); + +/* + * return error messages + */ +char * genxGetErrorMessage(genxWriter w, genxStatus status); +char * genxLastErrorMessage(genxWriter w); + +/* + * return version + */ +char * genxGetVersion(); + +#ifdef __cplusplus +} +#endif + +#endif /* GENX_H */ diff --git a/xml/exception.hxx b/xml/exception.hxx new file mode 100644 index 0000000..adb242f --- /dev/null +++ b/xml/exception.hxx @@ -0,0 +1,17 @@ +// file : xml/exception.hxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#ifndef XML_EXCEPTION_HXX +#define XML_EXCEPTION_HXX + +#include <exception> + +#include <xml/details/export.hxx> + +namespace xml +{ + struct LIBSTUDXML_EXPORT exception: std::exception {}; +} + +#endif // XML_EXCEPTION_HXX diff --git a/xml/makefile b/xml/makefile new file mode 100644 index 0000000..1447886 --- /dev/null +++ b/xml/makefile @@ -0,0 +1,129 @@ +# file : xml/makefile +# copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +# license : MIT; see accompanying LICENSE file + +include $(dir $(lastword $(MAKEFILE_LIST)))../build/bootstrap.make + +sources := qname.cxx parser.cxx serializer.cxx value-traits.cxx + +# Expat. +# +expat_sources := xmlparse.c xmlrole.c xmltok.c +expat_headers := expat.h expat_external.h +expat_dist := ascii.h asciitab.h config.h iasciitab.h internal.h \ +latin1tab.h nametab.h utf8tab.h xmlrole.h xmltok.h xmltok_impl.c \ +xmltok_impl.h xmltok_ns.c README LICENSE + +expat_sources := $(addprefix details/expat/,$(expat_sources)) +expat_headers := $(addprefix details/expat/,$(expat_headers)) +expat_dist := $(addprefix details/expat/,$(expat_dist)) + +# Genx. +# +genx_sources := details/genx/genx.c details/genx/char-props.c +genx_headers := details/genx/genx.h +genx_dist := details/genx/README details/genx/LICENSE + +c_tun := $(expat_sources) $(genx_sources) +c_obj := $(addprefix $(out_base)/,$(c_tun:.c=.o)) +c_od := $(c_obj:.o=.o.d) + +cxx_tun := $(sources) +cxx_obj := $(addprefix $(out_base)/,$(cxx_tun:.cxx=.o)) +cxx_od := $(cxx_obj:.o=.o.d) + +studxml.l := $(out_base)/studxml.l +studxml.l.cpp-options := $(out_base)/studxml.l.cpp-options + +default := $(out_base)/ +dist := $(out_base)/.dist +install := $(out_base)/.install +clean := $(out_base)/.clean + +# Build. +# +$(studxml.l): $(c_obj) $(cxx_obj) + +$(c_obj) $(cxx_obj) $(c_od) $(cxx_od): $(studxml.l.cpp-options) \ + $(out_base)/details/config.h +$(studxml.l.cpp-options): value := -I$(out_root) -I$(src_root) + +$(call include,$(bld_root)/ld/configuration-lib.make) # ld_lib_type + +$(out_base)/details/config.h: | $(out_base)/details/. + @echo '/* file : xml/details/config.h' >$@ + @echo ' * note : automatically generated' >>$@ + @echo ' */' >>$@ + @echo '#ifndef XML_DETAILS_CONFIG_H' >>$@ + @echo '#define XML_DETAILS_CONFIG_H' >>$@ + @echo '' >>$@ + @echo '#define LIBSTUDXML_BYTEORDER 1234' >>$@ +ifeq ($(ld_lib_type),archive) + @echo '#define LIBSTUDXML_STATIC_LIB' >>$@ +endif + @echo '' >>$@ + @echo '#endif /* XML_DETAILS_CONFIG_H */' >>$@ + +$(call include-dep,$(c_od) $(cxx_od),$(c_obj) $(cxx_obj),$(out_base)/details/config.h) + +# Convenience alias for default target. +# +$(out_base)/: $(studxml.l) + +# Dist. +# +$(dist): export sources := $(sources) +$(dist): export expat_sources := $(expat_sources) +$(dist): export expat_headers := $(expat_headers) +$(dist): export genx_sources := $(genx_sources) +$(dist): export genx_headers := $(genx_headers) +$(dist): export headers := $(subst $(src_base)/,,$(shell find $(src_base) \ +-name '*.hxx' -o -name '*.ixx' -o -name '*.txx')) +$(dist): data_dist := details/config-vc.h $(expat_dist) $(genx_dist) +$(dist): export extra_dist := $(data_dist) libstudxml-vc9.vcproj \ +libstudxml-vc10.vcxproj libstudxml-vc10.vcxproj.filters +$(dist): export interface_version = $(shell sed -e \ +'s/^\([0-9]*\.[0-9]*\).*/\1/' $(src_root)/version) + +$(dist): + $(call dist-data,$(sources) $(expat_sources) $(genx_sources)) + $(call dist-data,$(headers) $(expat_headers) $(genx_headers)) + $(call dist-data,$(data_dist) details/config.h.in) + $(call meta-vc9proj,libstudxml-vc9.vcproj) + $(call meta-vc10proj,libstudxml-vc10.vcxproj) + $(call meta-automake) + +# Clean. +# +$(clean): $(studxml.l).o.clean \ + $(studxml.l.cpp-options).clean \ + $(addsuffix .cxx.clean,$(c_obj)) \ + $(addsuffix .cxx.clean,$(c_od)) \ + $(addsuffix .cxx.clean,$(cxx_obj)) \ + $(addsuffix .cxx.clean,$(cxx_od)) + $(call message,rm $$1,rm -f $$1,$(out_base)/details/config.h) + +# Generated .gitignore. +# +ifeq ($(out_base),$(src_base)) +$(studxml.l): | $(out_base)/.gitignore + +$(out_base)/.gitignore: files := details/config.h +$(clean): $(out_base)/.gitignore.clean + +$(call include,$(bld_root)/git/gitignore.make) +endif + +# How to. +# +$(call include,$(bld_root)/dist.make) +$(call include,$(bld_root)/install.make) +$(call include,$(bld_root)/meta/vc9proj.make) +$(call include,$(bld_root)/meta/vc10proj.make) +$(call include,$(bld_root)/meta/automake.make) + +$(call include,$(bld_root)/c/c-o.make) +$(call include,$(bld_root)/c/c-d.make) +$(call include,$(bld_root)/cxx/cxx-o.make) +$(call include,$(bld_root)/cxx/cxx-d.make) +$(call include,$(bld_root)/cxx/o-l.make) diff --git a/xml/parser.cxx b/xml/parser.cxx new file mode 100644 index 0000000..d05add2 --- /dev/null +++ b/xml/parser.cxx @@ -0,0 +1,824 @@ +// file : xml/parser.cxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <new> // std::bad_alloc +#include <cassert> +#include <cstring> // std::strchr +#include <istream> +#include <ostream> +#include <sstream> + +#include <xml/parser.hxx> + +using namespace std; + +namespace xml +{ + // parsing + // + parsing:: + ~parsing () throw () {} + + parsing:: + parsing (const string& n, + unsigned long long l, + unsigned long long c, + const string& d) + : name_ (n), line_ (l), column_ (c), description_ (d) + { + init (); + } + + parsing:: + parsing (const parser& p, const std::string& d) + : name_ (p.input_name ()), + line_ (p.line ()), + column_ (p.column ()), + description_ (d) + { + init (); + } + + void parsing:: + init () + { + ostringstream os; + if (!name_.empty ()) + os << name_ << ':'; + os << line_ << ':' << column_ << ": error: " << description_; + what_ = os.str (); + } + + char const* parsing:: + what () const throw () + { + return what_.c_str (); + } + + // parser::event_type + // + static const char* parser_event_str[] = + { + "start element", + "end element", + "start attribute", + "end attribute", + "characters", + "start namespace declaration", + "end namespace declaration", + "end of file" + }; + + ostream& + operator<< (ostream& os, parser::event_type e) + { + return os << parser_event_str[e]; + } + + // parser + // + parser:: + ~parser () + { + if (p_ != 0) + XML_ParserFree (p_); + } + + parser:: + parser (istream& is, const string& iname, feature_type f) + : is_ (is), iname_ (iname), feature_ (f), + depth_ (0), state_ (state_next), event_ (eof), queue_ (eof), + pqname_ (&qname_), pvalue_ (&value_), + attr_i_ (0), start_ns_i_ (0), end_ns_i_ (0) + { + if ((feature_ & receive_attributes_map) != 0 && + (feature_ & receive_attributes_event) != 0) + feature_ &= ~receive_attributes_map; + + // Allocate the parser. Make sure nothing else can throw after + // this call since otherwise we will leak it. + // + p_ = XML_ParserCreateNS (0, XML_Char (' ')); + + if (p_ == 0) + throw bad_alloc (); + + // Get prefixes in addition to namespaces and local names. + // + XML_SetReturnNSTriplet (p_, true); + + // Set handlers. + // + XML_SetUserData(p_, this); + + if ((f & receive_elements) != 0) + { + XML_SetStartElementHandler (p_, &start_element_); + XML_SetEndElementHandler (p_, &end_element_); + } + + if ((f & receive_characters) != 0) + XML_SetCharacterDataHandler (p_, &characters_); + + if ((f & receive_namespace_decls) != 0) + XML_SetNamespaceDeclHandler (p_, + &start_namespace_decl_, + &end_namespace_decl_); + } + + void parser:: + handle_error () + { + XML_Error e (XML_GetErrorCode (p_)); + + if (e == XML_ERROR_ABORTED) + { + // For now we only abort the parser in the characters_() handler. + // + switch (content ()) + { + case empty: + throw parsing (*this, "character in empty content"); + case complex: + throw parsing (*this, "character in complex content"); + default: + assert (false); + } + } + else + throw parsing (iname_, + XML_GetCurrentLineNumber (p_), + XML_GetCurrentColumnNumber (p_), + XML_ErrorString (e)); + } + + struct stream_exception_controller + { + ~stream_exception_controller () + { + istream::iostate s = is_.rdstate (); + s &= ~istream::failbit; + + // If our error state (sans failbit) intersects with the + // exception state then that means we have an active + // exception and changing error/exception state will + // cause another to be thrown. + // + if (!(old_state_ & s)) + { + // Clear failbit if it was caused by eof. + // + if (is_.fail () && is_.eof ()) + is_.clear (s); + + is_.exceptions (old_state_); + } + } + + stream_exception_controller (istream& is) + : is_ (is), old_state_ (is_.exceptions ()) + { + is_.exceptions (old_state_ & ~istream::failbit); + } + + private: + stream_exception_controller (const stream_exception_controller&); + + stream_exception_controller& + operator= (const stream_exception_controller&); + + private: + istream& is_; + istream::iostate old_state_; + }; + + const string& parser:: + attribute (const qname_type& qn) const + { + if (const element_entry* e = get_element ()) + { + attribute_map_type::const_iterator i (e->attr_map_.find (qn)); + + if (i != e->attr_map_.end ()) + { + if (!i->second.handled) + { + i->second.handled = true; + e->attr_unhandled_--; + } + return i->second.value; + } + } + + throw parsing (*this, "attribute '" + qn.string () + "' expected"); + } + + string parser:: + attribute (const qname_type& qn, const string& dv) const + { + if (const element_entry* e = get_element ()) + { + attribute_map_type::const_iterator i (e->attr_map_.find (qn)); + + if (i != e->attr_map_.end ()) + { + if (!i->second.handled) + { + i->second.handled = true; + e->attr_unhandled_--; + } + return i->second.value; + } + } + + return dv; + } + + bool parser:: + attribute_present (const qname_type& qn) const + { + if (const element_entry* e = get_element ()) + { + attribute_map_type::const_iterator i (e->attr_map_.find (qn)); + + if (i != e->attr_map_.end ()) + { + if (!i->second.handled) + { + i->second.handled = true; + e->attr_unhandled_--; + } + return true; + } + } + + return false; + } + + void parser:: + next_expect (event_type e) + { + if (next () != e) + throw parsing (*this, string (parser_event_str[e]) + " expected"); + } + + void parser:: + next_expect (event_type e, const string& ns, const string& n) + { + if (next () != e || namespace_ () != ns || name () != n) + throw parsing (*this, + string (parser_event_str[e]) + " '" + + qname_type (ns, n).string () + "' expected"); + } + + const parser::element_entry* parser:: + get_element () const + { + // The start_element_() Expat handler may have already provisioned + // an entry in the element stack. In this case, we need to get the + // one before it, if any. + // + const element_entry* r (0); + element_state::size_type n (element_state_.size ()); + if (n != 0) + { + n--; + if (element_state_[n].depth == depth_) + r = &element_state_[n]; + else if (n != 0 && element_state_[n].depth > depth_) + { + n--; + if (element_state_[n].depth == depth_) + r = &element_state_[n]; + } + } + return r; + } + + void parser:: + pop_element () + { + // Make sure there are no unhandled attributes left. + // + const element_entry& e (element_state_.back ()); + if (e.attr_unhandled_ != 0) + { + // Find the first unhandled attribute and report it. + // + for (attribute_map_type::const_iterator i (e.attr_map_.begin ()); + i != e.attr_map_.end (); ++i) + { + if (!i->second.handled) + throw parsing ( + *this, "unexpected attribute '" + i->first.string () + "'"); + } + assert (false); + } + + element_state_.pop_back (); + } + + parser::event_type parser:: + next_ (bool peek) + { + event_type e (next_body ()); + + // Content-specific processing. Note that we handle characters in the + // characters_() Expat handler for two reasons. Firstly, it is faster + // to ignore the whitespaces at the source. Secondly, this allows us + // to distinguish between element and attribute characters. We can + // move this processing to the handler because the characters event + // is never queued. + // + switch (e) + { + case end_element: + { + // If this is a peek, then avoid popping the stack just yet. + // This way, the attribute map will still be valid until we + // call next(). + // + if (!peek) + { + if (!element_state_.empty () && + element_state_.back ().depth == depth_) + pop_element (); + + depth_--; + } + break; + } + case start_element: + { + const element_entry* e (get_element ()); + switch (e != 0 ? e->content : mixed) + { + case empty: + throw parsing (*this, "element in empty content"); + case simple: + throw parsing (*this, "element in simple content"); + default: + break; + } + + // If this is a peek, then delay adjusting the depth. + // + if (!peek) + depth_++; + + break; + } + default: + break; + } + + return e; + } + + parser::event_type parser:: + next_body () + { + // See if we have any start namespace declarations we need to return. + // + if (start_ns_i_ < start_ns_.size ()) + { + // Based on the previous event determine what's the next one must be. + // + switch (event_) + { + case start_namespace_decl: + { + if (++start_ns_i_ == start_ns_.size ()) + { + start_ns_i_ = 0; + start_ns_.clear (); + pqname_ = &qname_; + break; // No more declarations. + } + // Fall through. + } + case start_element: + { + event_ = start_namespace_decl; + pqname_ = &start_ns_[start_ns_i_]; + return event_; + } + default: + { + assert (false); + return event_ = eof; + } + } + } + + // See if we have any attributes we need to return as events. + // + if (attr_i_ < attr_.size ()) + { + // Based on the previous event determine what's the next one must be. + // + switch (event_) + { + case start_attribute: + { + event_ = characters; + pvalue_ = &attr_[attr_i_].value; + return event_; + } + case characters: + { + event_ = end_attribute; // Name is already set. + return event_; + } + case end_attribute: + { + if (++attr_i_ == attr_.size ()) + { + attr_i_ = 0; + attr_.clear (); + pqname_ = &qname_; + pvalue_ = &value_; + break; // No more attributes. + } + // Fall through. + } + case start_element: + case start_namespace_decl: + { + event_ = start_attribute; + pqname_ = &attr_[attr_i_].qname; + return event_; + } + default: + { + assert (false); + return event_ = eof; + } + } + } + + // See if we have any end namespace declarations we need to return. + // + if (end_ns_i_ < end_ns_.size ()) + { + // Based on the previous event determine what's the next one must be. + // + switch (event_) + { + case end_namespace_decl: + { + if (++end_ns_i_ == end_ns_.size ()) + { + end_ns_i_ = 0; + end_ns_.clear (); + pqname_ = &qname_; + break; // No more declarations. + } + // Fall through. + } + // The end namespace declaration comes before the end element + // which means it can follow pretty much any other event. + // + default: + { + event_ = end_namespace_decl; + pqname_ = &end_ns_[end_ns_i_]; + return event_; + } + } + } + + // Check the queue. + // + if (queue_ != eof) + { + event_ = queue_; + queue_ = eof; + return event_; + } + + XML_ParsingStatus ps; + XML_GetParsingStatus (p_, &ps); + + switch (ps.parsing) + { + case XML_INITIALIZED: + { + // As if we finished the previous chunk. + break; + } + case XML_PARSING: + { + assert (false); + return event_ = eof; + } + case XML_FINISHED: + { + return event_ = eof; + } + case XML_SUSPENDED: + { + switch (XML_ResumeParser (p_)) + { + case XML_STATUS_SUSPENDED: + { + // If the parser is again in the suspended state, then + // that means we have the next event. + // + return event_; + } + case XML_STATUS_OK: + { + // Otherwise, we need to get and parse the next chunk of data + // unless this was the last chunk, in which case this is eof. + // + if (ps.finalBuffer) + return event_ = eof; + + break; + } + case XML_STATUS_ERROR: + handle_error (); + } + + break; + } + } + + // Get and parse the next chunk of data until we get the next event + // or reach eof. + // + event_ = eof; + XML_Status s; + do + { + const size_t cap (4096); + + char* b (static_cast<char*> (XML_GetBuffer (p_, cap))); + if (b == 0) + throw bad_alloc (); + + // Temporarily unset the exception failbit. Also clear the fail bit + // when we reset the old state if it was caused by eof. + // + { + stream_exception_controller sec (is_); + is_.read (b, static_cast<streamsize> (cap)); + } + + s = XML_ParseBuffer (p_, static_cast<int> (is_.gcount ()), is_.eof ()); + + if (s == XML_STATUS_ERROR) + handle_error (); + + } while (s != XML_STATUS_SUSPENDED && !is_.eof ()); + + return event_; + } + + static void + split_name (const XML_Char* s, qname& qn) + { + string& ns (qn.namespace_ ()); + string& name (qn.name ()); + string& prefix (qn.prefix ()); + + const char* p (strchr (s, ' ')); + + if (p == 0) + { + ns.clear (); + name = s; + prefix.clear (); + } + else + { + ns.assign (s, 0, p - s); + + s = p + 1; + p = strchr (s, ' '); + + if (p == 0) + { + name = s; + prefix.clear (); + } + else + { + name.assign (s, 0, p - s); + prefix = p + 1; + } + } + } + + void XMLCALL parser:: + start_element_ (void* v, const XML_Char* name, const XML_Char** atts) + { + parser& p (*static_cast<parser*> (v)); + + XML_ParsingStatus ps; + XML_GetParsingStatus (p.p_, &ps); + + // Expat has a (mis)-feature of a possibily calling handlers even + // after the non-resumable XML_StopParser call. + // + if (ps.parsing == XML_FINISHED) + return; + + // Cannot be a followup event. + // + assert (ps.parsing == XML_PARSING); + + p.event_ = start_element; + split_name (name, p.qname_); + + p.line_ = XML_GetCurrentLineNumber (p.p_); + p.column_ = XML_GetCurrentColumnNumber (p.p_); + + // Handle attributes. + // + if (*atts != 0) + { + bool am ((p.feature_ & receive_attributes_map) != 0); + bool ae ((p.feature_ & receive_attributes_event) != 0); + + // Provision an entry for this element. + // + element_entry* pe (0); + if (am) + { + p.element_state_.push_back (element_entry (p.depth_ + 1)); + pe = &p.element_state_.back (); + } + + if (am || ae) + { + for (; *atts != 0; atts += 2) + { + if (am) + { + qname_type qn; + split_name (*atts, qn); + attribute_map_type::value_type v (qn, attribute_value_type ()); + v.second.value = *(atts + 1); + v.second.handled = false; + pe->attr_map_.insert (v); + } + else + { + p.attr_.push_back (attribute_type ()); + split_name (*atts, p.attr_.back ().qname); + p.attr_.back ().value = *(atts + 1); + } + } + + if (am) + pe->attr_unhandled_ = pe->attr_map_.size (); + } + } + + XML_StopParser (p.p_, true); + } + + void XMLCALL parser:: + end_element_ (void* v, const XML_Char* name) + { + parser& p (*static_cast<parser*> (v)); + + XML_ParsingStatus ps; + XML_GetParsingStatus (p.p_, &ps); + + // Expat has a (mis)-feature of a possibily calling handlers even + // after the non-resumable XML_StopParser call. + // + if (ps.parsing == XML_FINISHED) + return; + + // This can be a followup event for empty elements (<foo/>). In this + // case the element name is already set. + // + if (ps.parsing != XML_PARSING) + p.queue_ = end_element; + else + { + // We may also have the end namespace declaration events which + // should come before the end element. If that's the case, then + // queue the end element and return the end namespace as the next + // event. + // + if (p.end_ns_i_ < p.end_ns_.size ()) + { + p.event_ = end_namespace_decl; + p.queue_ = end_element; + } + else + p.event_ = end_element; + + split_name (name, p.qname_); + + p.line_ = XML_GetCurrentLineNumber (p.p_); + p.column_ = XML_GetCurrentColumnNumber (p.p_); + + XML_StopParser (p.p_, true); + } + } + + void XMLCALL parser:: + characters_ (void* v, const XML_Char* s, int n) + { + parser& p (*static_cast<parser*> (v)); + + XML_ParsingStatus ps; + XML_GetParsingStatus (p.p_, &ps); + + // Expat has a (mis)-feature of a possibily calling handlers even + // after the non-resumable XML_StopParser call. + // + if (ps.parsing == XML_FINISHED) + return; + + // If this is empty or complex content, see if these are whitespaces. + // + switch (p.content ()) + { + case empty: + case complex: + { + for (int i (0); i != n; ++i) + { + char c (s[i]); + if (c == 0x20 || c == 0x0A || c == 0x0D || c == 0x09) + continue; + + // It would have been easier to throw the exception directly, + // however, the Expat code is most likely not exception safe. + // + p.line_ = XML_GetCurrentLineNumber (p.p_); + p.column_ = XML_GetCurrentColumnNumber (p.p_); + XML_StopParser (p.p_, false); + break; + } + return; + } + default: + break; + } + + // This can be a followup event for another character event. In + // this case simply append the data. + // + if (ps.parsing != XML_PARSING) + { + assert (p.event_ == characters); + p.value_.append (s, n); + } + else + { + p.event_ = characters; + p.value_.assign (s, n); + + p.line_ = XML_GetCurrentLineNumber (p.p_); + p.column_ = XML_GetCurrentColumnNumber (p.p_); + + XML_StopParser (p.p_, true); + } + } + + void XMLCALL parser:: + start_namespace_decl_ (void* v, const XML_Char* prefix, const XML_Char* ns) + { + parser& p (*static_cast<parser*> (v)); + + XML_ParsingStatus ps; + XML_GetParsingStatus (p.p_, &ps); + + // Expat has a (mis)-feature of a possibily calling handlers even + // after the non-resumable XML_StopParser call. + // + if (ps.parsing == XML_FINISHED) + return; + + p.start_ns_.push_back (qname_type ()); + p.start_ns_.back ().prefix () = (prefix != 0 ? prefix : ""); + p.start_ns_.back ().namespace_ () = (ns != 0 ? ns : ""); + } + + void XMLCALL parser:: + end_namespace_decl_ (void* v, const XML_Char* prefix) + { + parser& p (*static_cast<parser*> (v)); + + XML_ParsingStatus ps; + XML_GetParsingStatus (p.p_, &ps); + + // Expat has a (mis)-feature of a possibily calling handlers even + // after the non-resumable XML_StopParser call. + // + if (ps.parsing == XML_FINISHED) + return; + + p.end_ns_.push_back (qname_type ()); + p.end_ns_.back ().prefix () = (prefix != 0 ? prefix : ""); + } +} diff --git a/xml/parser.hxx b/xml/parser.hxx new file mode 100644 index 0000000..09afc56 --- /dev/null +++ b/xml/parser.hxx @@ -0,0 +1,422 @@ +// file : xml/parser.hxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#ifndef XML_PARSER_HXX +#define XML_PARSER_HXX + +#include <map> +#include <vector> +#include <string> +#include <iosfwd> +#include <cstddef> // std::size_t +#include <cassert> + +#include <xml/details/config.hxx> // LIBSTUDXML_EXTERNAL_EXPAT + +#ifndef LIBSTUDXML_EXTERNAL_EXPAT +# include <xml/details/expat/expat.h> +#else +# include <expat.h> +#endif + +// We only support UTF-8 Expat. +// +#ifdef XML_UNICODE +# error UTF-16 expat (XML_UNICODE defined) is not supported +#endif + +#include <xml/qname.hxx> +#include <xml/exception.hxx> + +#include <xml/details/export.hxx> + +namespace xml +{ + class parser; + + struct LIBSTUDXML_EXPORT parsing: exception + { + virtual + ~parsing () throw (); + + parsing (const std::string& name, + unsigned long long line, + unsigned long long column, + const std::string& description); + + parsing (const parser&, const std::string& description); + + const std::string& + name () const {return name_;} + + unsigned long long + line () const {return line_;} + + unsigned long long + column () const {return column_;} + + const std::string& + description () const {return description_;} + + virtual const char* + what () const throw (); + + private: + void + init (); + + private: + std::string name_; + unsigned long long line_; + unsigned long long column_; + std::string description_; + std::string what_; + }; + + class LIBSTUDXML_EXPORT parser + { + public: + typedef xml::qname qname_type; + typedef unsigned short feature_type; + + // If both receive_attributes_event and receive_attributes_map are + // specified, then receive_attributes_event is assumed. + // + static const feature_type receive_elements = 0x0001; + static const feature_type receive_characters = 0x0002; + static const feature_type receive_attributes_map = 0x0004; + static const feature_type receive_attributes_event = 0x0008; + static const feature_type receive_namespace_decls = 0x0010; + + static const feature_type receive_default = receive_elements | + receive_characters | + receive_attributes_map; + + ~parser (); + + // Parse std::istream. Input name is used in diagnostics to identify + // the document being parsed. std::ios_base::failure exception is + // used to report io errors (badbit and failbit). + // + parser (std::istream&, + const std::string& input_name, + feature_type = receive_default); + + const std::string& + input_name () const {return iname_;} + + // Parsing events. + // + public: + enum event_type + { + // If adding new events, also update the stream insertion operator. + // + start_element, + end_element, + start_attribute, + end_attribute, + characters, + start_namespace_decl, + end_namespace_decl, + eof + }; + + event_type + next () + { + // Move to .ixx. + + if (state_ == state_next) + return next_ (false); + else + { + // If we previously peeked at start/end_element, then adjust + // state accordingly. + // + switch (event_) + { + case end_element: + { + if (!element_state_.empty () && + element_state_.back ().depth == depth_) + pop_element (); + + depth_--; + break; + } + case start_element: + { + depth_++; + break; + } + default: + break; + } + + state_ = state_next; + return event_; + } + } + + // Get the next event and make sure that it's what's expected. If it + // is not, then throw an appropriate parsing exception. + // + void + next_expect (event_type); + + void + next_expect (event_type, const qname_type& qname); + + void + next_expect (event_type, const std::string& name); + + void + next_expect (event_type, const std::string& ns, const std::string& name); + + event_type + peek () + { + //@@ move to .ixx + + if (state_ == state_peek) + return event_; + else + { + event_type e (next_ (true)); + state_ = state_peek; // Set it after the call to next_(). + return e; + } + } + + // Return the even that was last returned by the call to next() or + // peek(). + // + event_type + event () {return event_;} + + // Event data. + // + public: + const qname_type& qname () const {return *pqname_;} + + const std::string& namespace_ () const {return pqname_->namespace_ ();} + const std::string& name () const {return pqname_->name ();} + const std::string& prefix () const {return pqname_->prefix ();} + + const std::string& value () const {return *pvalue_;} + + unsigned long long line () const {return line_;} + unsigned long long column () const {return column_;} + + // Attribute map lookup. If attribute is not found, then the version + // without the default value throws an appropriate parsing exception + // while the version with the default value returns that value. + // + // Note also that there is no attribute(ns,name) version since it + // would conflict with attribute(name,dv) (qualified attributes + // are not very common). + // + // Attribute map is valid throughout at the "element level" until + // end_element and not just during start_element. As a special case, + // the map is still valid after peek() that returned end_element until + // this end_element event is retrieved with next(). + // + const std::string& + attribute (const std::string& name) const; + + template <typename T> + T + attribute (const std::string& name) const; + + std::string + attribute (const std::string& name, const std::string& dv) const; + + template <typename T> + T + attribute (const std::string& name, const T& dv) const; + + const std::string& + attribute (const qname_type& qname) const; + + template <typename T> + T + attribute (const qname_type& qname) const; + + std::string + attribute (const qname_type& qname, const std::string& dv) const; + + template <typename T> + T + attribute (const qname_type& qname, const T& dv) const; + + bool + attribute_present (const std::string& name) const; + + bool + attribute_present (const qname_type& qname) const; + + // Low-level attribute map access. Note that this API assumes + // all attributes are handled. + // + struct attribute_value_type + { + std::string value; + mutable bool handled; + }; + + typedef std::map<qname_type, attribute_value_type> attribute_map_type; + + const attribute_map_type& + attribute_map () const; + + // Optional content processing. + // + public: + enum content_type + { + // element characters whitespaces + empty, // no no ignored + simple, // no yes preserved + complex, // yes no ignored + mixed // yes yes preserved + }; + + // Note that you cannot get/set content while peeking. + // + void + content (content_type c) + { + //@@ move to .ixx + + assert (state_ == state_next); + + if (!element_state_.empty () && element_state_.back ().depth == depth_) + element_state_.back ().content = c; + else + element_state_.push_back (element_entry (depth_, c)); + } + + content_type + content () const + { + assert (state_ == state_next); + + return + !element_state_.empty () && element_state_.back ().depth == depth_ + ? element_state_.back ().content + : mixed; + } + + private: + static void XMLCALL + start_element_ (void*, const XML_Char*, const XML_Char**); + + static void XMLCALL + end_element_ (void*, const XML_Char*); + + static void XMLCALL + characters_ (void*, const XML_Char*, int); + + static void XMLCALL + start_namespace_decl_ (void*, const XML_Char*, const XML_Char*); + + static void XMLCALL + end_namespace_decl_ (void*, const XML_Char*); + + private: + event_type + next_ (bool peek); + + event_type + next_body (); + + void + handle_error (); + + private: + std::istream& is_; + const std::string iname_; + feature_type feature_; + + XML_Parser p_; + std::size_t depth_; + enum {state_next, state_peek} state_; + event_type event_; + event_type queue_; + + qname_type qname_; + std::string value_; + + // These are used to avoid copying when we are handling attributes + // and namespace decls. + // + const qname_type* pqname_; + const std::string* pvalue_; + + unsigned long long line_; + unsigned long long column_; + + // Attributes as events. + // + struct attribute_type + { + qname_type qname; + std::string value; + }; + + typedef std::vector<attribute_type> attributes; + + attributes attr_; + attributes::size_type attr_i_; // Index of the current attribute. + + // Namespace declarations. + // + typedef std::vector<qname_type> namespace_decls; + + namespace_decls start_ns_; + namespace_decls::size_type start_ns_i_; // Index of the current decl. + + namespace_decls end_ns_; + namespace_decls::size_type end_ns_i_; // Index of the current decl. + + // Element state consisting of the content model and attribute map. + // + struct element_entry + { + element_entry (std::size_t d, content_type c = mixed) + : depth (d), content (c), attr_unhandled_ (0) {} + + std::size_t depth; + content_type content; + attribute_map_type attr_map_; + mutable attribute_map_type::size_type attr_unhandled_; + }; + + typedef std::vector<element_entry> element_state; + std::vector<element_entry> element_state_; + + // Empty attribute map to return when an element has no attributes. + // + const attribute_map_type empty_attr_map_; + + // Return the element entry corresponding to the current depth, if + // exists, and NULL otherwise. + // + const element_entry* + get_element () const; + + void + pop_element (); + }; + + LIBSTUDXML_EXPORT + std::ostream& + operator<< (std::ostream&, parser::event_type); +} + +#include <xml/parser.ixx> +#include <xml/parser.txx> + +#endif // XML_PARSER_HXX diff --git a/xml/parser.ixx b/xml/parser.ixx new file mode 100644 index 0000000..18cdf69 --- /dev/null +++ b/xml/parser.ixx @@ -0,0 +1,71 @@ +// file : xml/parser.ixx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <xml/value-traits.hxx> + +namespace xml +{ + inline const std::string& parser:: + attribute (const std::string& n) const + { + return attribute (qname_type (n)); + } + + template <typename T> + inline T parser:: + attribute (const std::string& n) const + { + return attribute<T> (qname_type (n)); + } + + inline std::string parser:: + attribute (const std::string& n, const std::string& dv) const + { + return attribute (qname_type (n), dv); + } + + template <typename T> + inline T parser:: + attribute (const std::string& n, const T& dv) const + { + return attribute<T> (qname_type (n), dv); + } + + template <typename T> + inline T parser:: + attribute (const qname_type& qn) const + { + return value_traits<T>::parse (attribute (qn), *this); + } + + inline bool parser:: + attribute_present (const std::string& n) const + { + return attribute_present (qname_type (n)); + } + + inline const parser::attribute_map_type& parser:: + attribute_map () const + { + if (const element_entry* e = get_element ()) + { + e->attr_unhandled_ = 0; // Assume all handled. + return e->attr_map_; + } + + return empty_attr_map_; + } + + inline void parser:: + next_expect (event_type e, const qname_type& qn) + { + return next_expect (e, qn.namespace_ (), qn.name ()); + } + + inline void parser:: + next_expect (event_type e, const std::string& n) + { + return next_expect (e, std::string (), n); + } +} diff --git a/xml/parser.txx b/xml/parser.txx new file mode 100644 index 0000000..35ed8e5 --- /dev/null +++ b/xml/parser.txx @@ -0,0 +1,30 @@ +// file : xml/parser.txx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <xml/value-traits.hxx> + +namespace xml +{ + template <typename T> + T parser:: + attribute (const qname_type& qn, const T& dv) const + { + if (const element_entry* e = get_element ()) + { + attribute_map_type::const_iterator i (e->attr_map_.find (qn)); + + if (i != e->attr_map_.end ()) + { + if (!i->second.handled) + { + i->second.handled = true; + e->attr_unhandled_--; + } + return value_traits<T>::parse (i->second.value, *this); + } + } + + return dv; + } +} diff --git a/xml/qname.cxx b/xml/qname.cxx new file mode 100644 index 0000000..c900efc --- /dev/null +++ b/xml/qname.cxx @@ -0,0 +1,32 @@ +// file : xml/qname.cxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <ostream> + +#include <xml/qname.hxx> + +using namespace std; + +namespace xml +{ + string qname:: + string () const + { + std::string r; + if (!ns_.empty ()) + { + r += ns_; + r += '#'; + } + + r += name_; + return r; + } + + ostream& + operator<< (ostream& os, const qname& qn) + { + return os << qn.string (); + } +} diff --git a/xml/qname.hxx b/xml/qname.hxx new file mode 100644 index 0000000..be1bf00 --- /dev/null +++ b/xml/qname.hxx @@ -0,0 +1,78 @@ +// file : xml/qname.hxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#ifndef XML_QNAME_HXX +#define XML_QNAME_HXX + +#include <string> +#include <iosfwd> + +#include <xml/details/export.hxx> + +namespace xml +{ + // Note that the optional prefix is just a "syntactic sugar". In + // particular, it is ignored by the comparison operators and the + // std::ostream insertion operator. + // + class LIBSTUDXML_EXPORT qname + { + public: + qname () {} + qname (const std::string& name): name_ (name) {} + qname (const std::string& ns, const std::string& name) + : ns_ (ns), name_ (name) {} + qname (const std::string& ns, + const std::string& name, + const std::string& prefix) + : ns_ (ns), name_ (name), prefix_ (prefix) {} + + const std::string& namespace_ () const {return ns_;} + const std::string& name () const {return name_;} + const std::string& prefix () const {return prefix_;} + + std::string& namespace_ () {return ns_;} + std::string& name () {return name_;} + std::string& prefix () {return prefix_;} + + // String representation in the [<namespace>#]<name> form. + // + std::string + string () const; + + // Note that comparison operators ignore prefixes. + // + public: + friend bool + operator< (const qname& x, const qname& y) + { + return x.ns_ < y.ns_ || (x.ns_ == y.ns_ && x.name_ < y.name_); + } + + friend bool + operator== (const qname& x, const qname& y) + { + return x.ns_ == y.ns_ && x.name_ == y.name_; + } + + friend bool + operator!= (const qname& x, const qname& y) + { + return !(x == y); + } + + private: + std::string ns_; + std::string name_; + std::string prefix_; + }; + + // Print the string representation ([<namespace>#]<name>). + // + LIBSTUDXML_EXPORT + std::ostream& + operator<< (std::ostream&, const qname&); +} + +#endif // XML_QNAME_HXX diff --git a/xml/serializer.cxx b/xml/serializer.cxx new file mode 100644 index 0000000..59b47e5 --- /dev/null +++ b/xml/serializer.cxx @@ -0,0 +1,255 @@ +// file : xml/serializer.cxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <new> // std::bad_alloc +#include <cstring> // std::strlen + +#include <xml/serializer.hxx> + +using namespace std; + +namespace xml +{ + // serialization + // + serialization:: + ~serialization () throw () {} + + serialization:: + serialization (const string& n, const string& d) + : name_ (n), description_ (d) + { + init (); + } + + serialization:: + serialization (const serializer& s, const std::string& d) + : name_ (s.output_name ()), description_ (d) + { + init (); + } + + void serialization:: + init () + { + if (!name_.empty ()) + { + what_ += name_; + what_ += ": "; + } + + what_ += "error: "; + what_ += description_; + } + + char const* serialization:: + what () const throw () + { + return what_.c_str (); + } + + // serializer + // + extern "C" genxStatus + genx_write (void* p, constUtf8 us) + { + // It would have been easier to throw the exception directly, + // however, the Genx code is most likely not exception safe. + // + ostream* os (static_cast<ostream*> (p)); + const char* s (reinterpret_cast<const char*> (us)); + os->write (s, static_cast<streamsize> (strlen (s))); + return os->good () ? GENX_SUCCESS : GENX_IO_ERROR; + } + + extern "C" genxStatus + genx_write_bound (void* p, constUtf8 start, constUtf8 end) + { + ostream* os (static_cast<ostream*> (p)); + const char* s (reinterpret_cast<const char*> (start)); + streamsize n (static_cast<streamsize> (end - start)); + os->write (s, n); + return os->good () ? GENX_SUCCESS : GENX_IO_ERROR; + } + + extern "C" genxStatus + genx_flush (void* p) + { + ostream* os (static_cast<ostream*> (p)); + os->flush (); + return os->good () ? GENX_SUCCESS : GENX_IO_ERROR; + } + + serializer:: + ~serializer () + { + if (s_ != 0) + genxDispose (s_); + } + + serializer:: + serializer (ostream& os, const string& oname, unsigned short ind) + : os_ (os), os_state_ (os.exceptions ()), oname_ (oname), depth_ (0) + { + // Temporarily disable exceptions on the stream. + // + os_.exceptions (ostream::goodbit); + + // Allocate the serializer. Make sure nothing else can throw after + // this call since otherwise we will leak it. + // + s_ = genxNew (0, 0, 0); + + if (s_ == 0) + throw bad_alloc (); + + genxSetUserData (s_, &os_); + + if (ind != 0) + genxSetPrettyPrint (s_, ind); + + sender_.send = &genx_write; + sender_.sendBounded = &genx_write_bound; + sender_.flush = &genx_flush; + + if (genxStatus e = genxStartDocSender (s_, &sender_)) + { + string m (genxGetErrorMessage (s_, e)); + genxDispose (s_); + throw serialization (oname, m); + } + } + + void serializer:: + handle_error (genxStatus e) + { + switch (e) + { + case GENX_ALLOC_FAILED: + throw bad_alloc (); + case GENX_IO_ERROR: + // Restoring the original exception state should trigger the + // exception. If it doesn't (e.g., because the user didn't + // configure the stream to throw), then fall back to the + // serialiation exception. + // + os_.exceptions (os_state_); + // Fall through. + default: + throw serialization (oname_, genxGetErrorMessage (s_, e)); + } + } + + void serializer:: + start_element (const string& ns, const string& name) + { + if (genxStatus e = genxStartElementLiteral ( + s_, + reinterpret_cast<constUtf8> (ns.empty () ? 0 : ns.c_str ()), + reinterpret_cast<constUtf8> (name.c_str ()))) + handle_error (e); + + depth_++; + } + + void serializer:: + end_element () + { + if (genxStatus e = genxEndElement (s_)) + handle_error (e); + + // Call EndDocument() if we are past the root element. + // + if (--depth_ == 0) + { + if (genxStatus e = genxEndDocument (s_)) + handle_error (e); + + // Also restore the original exception state on the stream. + // + os_.exceptions (os_state_); + } + } + + void serializer:: + start_attribute (const string& ns, const string& name) + { + if (genxStatus e = genxStartAttributeLiteral ( + s_, + reinterpret_cast<constUtf8> (ns.empty () ? 0 : ns.c_str ()), + reinterpret_cast<constUtf8> (name.c_str ()))) + handle_error (e); + } + + void serializer:: + end_attribute () + { + if (genxStatus e = genxEndAttribute (s_)) + handle_error (e); + } + + void serializer:: + attribute (const string& ns, + const string& name, + const string& value) + { + if (genxStatus e = genxAddAttributeLiteral ( + s_, + reinterpret_cast<constUtf8> (ns.empty () ? 0 : ns.c_str ()), + reinterpret_cast<constUtf8> (name.c_str ()), + reinterpret_cast<constUtf8> (value.c_str ()))) + handle_error (e); + } + + void serializer:: + characters (const string& value) + { + if (genxStatus e = genxAddCountedText ( + s_, + reinterpret_cast<constUtf8> (value.c_str ()), + static_cast<int> (value.size ()))) + handle_error (e); + } + + void serializer:: + namespace_decl (const string& ns, const string& p) + { + if (genxStatus e = ns.empty () && p.empty () + ? genxUnsetDefaultNamespace (s_) + : genxAddNamespaceLiteral ( + s_, + reinterpret_cast<constUtf8> (ns.c_str ()), + reinterpret_cast<constUtf8> (p.c_str ()))) + handle_error (e); + } + + void serializer:: + xml_decl (const string& ver, const string& enc, const string& stl) + { + if (genxStatus e = genxXmlDeclaration ( + s_, + reinterpret_cast<constUtf8> (ver.c_str ()), + (enc.empty () ? 0 : reinterpret_cast<constUtf8> (enc.c_str ())), + (stl.empty () ? 0 : reinterpret_cast<constUtf8> (stl.c_str ())))) + handle_error (e); + } + + bool serializer:: + lookup_namespace_prefix (const string& ns, string& p) + { + // Currently Genx will create a namespace mapping if one doesn't + // already exist. + // + genxStatus e; + genxNamespace gns ( + genxDeclareNamespace ( + s_, reinterpret_cast<constUtf8> (ns.c_str ()), 0, &e)); + + if (e != GENX_SUCCESS) + handle_error (e); + + p = reinterpret_cast<const char*> (genxGetNamespacePrefix (gns)); + return true; + } +} diff --git a/xml/serializer.hxx b/xml/serializer.hxx new file mode 100644 index 0000000..59baff8 --- /dev/null +++ b/xml/serializer.hxx @@ -0,0 +1,180 @@ +// file : xml/serializer.hxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#ifndef XML_SERIALIZER_HXX +#define XML_SERIALIZER_HXX + +#include <string> +#include <ostream> +#include <cstddef> // std::size_t + +#include <xml/details/genx/genx.h> + +#include <xml/qname.hxx> +#include <xml/exception.hxx> + +#include <xml/details/export.hxx> + +namespace xml +{ + class serializer; + + struct LIBSTUDXML_EXPORT serialization: exception + { + virtual + ~serialization () throw (); + + serialization (const std::string& name, + const std::string& description); + + serialization (const serializer&, const std::string& description); + + const std::string& + name () const {return name_;} + + const std::string& + description () const {return description_;} + + virtual const char* + what () const throw (); + + private: + void + init (); + + private: + std::string name_; + std::string description_; + std::string what_; + }; + + class LIBSTUDXML_EXPORT serializer + { + public: + typedef xml::qname qname_type; + + ~serializer (); + + // Serialize to std::ostream. Output name is used in diagnostics to + // identify the document being serialized. std::ios_base::failure + // exception is used to report io errors (badbit and failbit). The + // indentation argument specifies the number of indentation spaces + // that should be used for pretty-printing. If 0 is passed, no + // pretty-printing is performed. + // + serializer (std::ostream&, + const std::string& output_name, + unsigned short indentation = 2); + + const std::string& + output_name () const {return oname_;} + + // Serialization functions. + // + public: + + // Elements. + // + void + start_element (const qname_type& qname); + + void + start_element (const std::string& name); + + void + start_element (const std::string& ns, const std::string& name); + + void + end_element (); + + // Attributes. + // + void + start_attribute (const qname_type& qname); + + void + start_attribute (const std::string& name); + + void + start_attribute (const std::string& ns, const std::string& name); + + void + end_attribute (); + + void + attribute (const qname_type& qname, const std::string& value); + + template <typename T> + void + attribute (const qname_type& qname, const T& value); + + void + attribute (const std::string& name, const std::string& value); + + template <typename T> + void + attribute (const std::string& name, const T& value); + + void + attribute (const std::string& ns, + const std::string& name, + const std::string& value); + + template <typename T> + void + attribute (const std::string& ns, + const std::string& name, + const T& value); + + // Characters. + // + void + characters (const std::string& value); + + template <typename T> + void + characters (const T& value); + + // Namespaces declaration. If prefix is empty, then the default + // namespace is declared. If both prefix and namespace are empty, + // then the default namespace declaration is cleared (xmlns=""). + // + void + namespace_decl (const std::string& ns, const std::string& prefix); + + // XML Declaration. If encoding or standalone are not specified, + // then these attributes are omitted from the output. + // + void + xml_decl (const std::string& version = "1.0", + const std::string& encoding = "UTF-8", + const std::string& standalone = ""); + + // Utility functions. + // + public: + // Return true if there is a mapping. In this case, prefix contains + // the mapped prefix. + // + bool + lookup_namespace_prefix (const std::string& ns, std::string& prefix); + + private: + void + handle_error (genxStatus); + + private: + std::ostream& os_; + std::ostream::iostate os_state_; // Original exception state. + const std::string oname_; + + genxWriter s_; + genxSender sender_; + std::size_t depth_; + }; +} + +#include <xml/serializer.ixx> + +#endif // XML_SERIALIZER_HXX diff --git a/xml/serializer.ixx b/xml/serializer.ixx new file mode 100644 index 0000000..882cfba --- /dev/null +++ b/xml/serializer.ixx @@ -0,0 +1,72 @@ +// file : xml/serializer.ixx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <xml/value-traits.hxx> + +namespace xml +{ + inline void serializer:: + start_element (const qname_type& qname) + { + start_element (qname.namespace_ (), qname.name ()); + } + + inline void serializer:: + start_element (const std::string& name) + { + start_element (std::string (), name); + } + + inline void serializer:: + start_attribute (const qname_type& qname) + { + start_attribute (qname.namespace_ (), qname.name ()); + } + + inline void serializer:: + start_attribute (const std::string& name) + { + start_attribute (std::string (), name); + } + + inline void serializer:: + attribute (const qname_type& qname, const std::string& value) + { + attribute (qname.namespace_ (), qname.name (), value); + } + + template <typename T> + inline void serializer:: + attribute (const qname_type& qname, const T& value) + { + attribute (qname, value_traits<T>::serialize (value, *this)); + } + + inline void serializer:: + attribute (const std::string& name, const std::string& value) + { + attribute (std::string (), name, value); + } + + template <typename T> + inline void serializer:: + attribute (const std::string& name, const T& value) + { + attribute (name, value_traits<T>::serialize (value, *this)); + } + + template <typename T> + inline void serializer:: + attribute (const std::string& ns, const std::string& name, const T& value) + { + attribute (ns, name, value_traits<T>::serialize (value, *this)); + } + + template <typename T> + inline void serializer:: + characters (const T& value) + { + characters (value_traits<T>::serialize (value, *this)); + } +} diff --git a/xml/value-traits.cxx b/xml/value-traits.cxx new file mode 100644 index 0000000..ea118ae --- /dev/null +++ b/xml/value-traits.cxx @@ -0,0 +1,21 @@ +// file : xml/value-traits.cxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <xml/parser.hxx> + +using namespace std; + +namespace xml +{ + bool default_value_traits<bool>:: + parse (string s, const parser& p) + { + if (s == "true" || s == "1" || s == "True" || s == "TRUE") + return true; + else if (s == "false" || s == "0" || s == "False" || s == "FALSE") + return false; + else + throw parsing (p, "invalid bool value '" + s + "'"); + } +} diff --git a/xml/value-traits.hxx b/xml/value-traits.hxx new file mode 100644 index 0000000..7531f2c --- /dev/null +++ b/xml/value-traits.hxx @@ -0,0 +1,51 @@ +// file : xml/value-traits.hxx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#ifndef XML_VALUE_TRAITS_HXX +#define XML_VALUE_TRAITS_HXX + +#include <string> +#include <cstddef> // std::size_t + +#include <xml/details/export.hxx> + +namespace xml +{ + //@@ forward + class parser; + class serializer; + + template <typename T> + struct default_value_traits + { + static T + parse (std::string, const parser&); + + static std::string + serialize (const T&, const serializer&); + }; + + template <> + struct LIBSTUDXML_EXPORT default_value_traits<bool> + { + static bool + parse (std::string, const parser&); + + static std::string + serialize (bool v, const serializer&) + { + return v ? "true" : "false"; + } + }; + + template <typename T> + struct value_traits: default_value_traits<T> {}; + + template <typename T, std::size_t N> + struct value_traits<T[N]>: default_value_traits<const T*> {}; +} + +#include <xml/value-traits.txx> + +#endif // XML_VALUE_TRAITS_HXX diff --git a/xml/value-traits.txx b/xml/value-traits.txx new file mode 100644 index 0000000..9cc0777 --- /dev/null +++ b/xml/value-traits.txx @@ -0,0 +1,32 @@ +// file : xml/value-traits.txx +// copyright : Copyright (c) 2009-2013 Code Synthesis Tools CC +// license : MIT; see accompanying LICENSE file + +#include <sstream> + +#include <xml/parser.hxx> +#include <xml/serializer.hxx> + +namespace xml +{ + template <typename T> + T default_value_traits<T>:: + parse (std::string s, const parser& p) + { + T r; + std::istringstream is (s); + if (!(is >> r && is.eof ()) ) + throw parsing (p, "invalid value '" + s + "'"); + return r; + } + + template <typename T> + std::string default_value_traits<T>:: + serialize (const T& v, const serializer& s) + { + std::ostringstream os; + if (!(os << v)) + throw serialization (s, "invalid value"); + return os.str (); + } +} |