diff options
author | Boris Kolpackov <boris@codesynthesis.com> | 2020-06-25 14:21:51 +0200 |
---|---|---|
committer | Boris Kolpackov <boris@codesynthesis.com> | 2020-06-25 14:21:51 +0200 |
commit | 8557179f8cd9afa9f01079aec954e4ee5958f32c (patch) | |
tree | 25e0915768d7c8f2652e4900100c418748b43769 | |
parent | 8242a224d43203999fceccc27f48989db2276221 (diff) |
Fix security issues in bundled expat
Specifically, the following patches from the Debian 2.1.0-6+deb8u6 package
were applied:
CVE-2018-20843.patch
CVE-2015-1283.patch
CVE-2015-1283-refix.patch
CVE-2016-0718-v2-2-1.patch
cve-2012-6702-plus-cve-2016-5300-v1.patch
CVE-2017-9233.patch
CVE-2016-9063.patch
CVE-2019-15903.patch
Additionally, hashing code has been changed to use size_t instead of
unsigned long (which is 32-bit on 64-bit Windows).
-rw-r--r-- | libstudxml/details/expat/config.h | 1 | ||||
-rw-r--r-- | libstudxml/details/expat/xmlparse.c | 185 | ||||
-rw-r--r-- | libstudxml/details/expat/xmltok.c | 117 | ||||
-rw-r--r-- | libstudxml/details/expat/xmltok.h | 10 | ||||
-rw-r--r-- | libstudxml/details/expat/xmltok_impl.c | 62 |
5 files changed, 260 insertions, 115 deletions
diff --git a/libstudxml/details/expat/config.h b/libstudxml/details/expat/config.h index 99fa673..eb32265 100644 --- a/libstudxml/details/expat/config.h +++ b/libstudxml/details/expat/config.h @@ -24,6 +24,7 @@ #undef WIN32_LEAN_AND_MEAN #define HAVE_MEMMOVE 1 +#define COMPILED_FROM_DSP 1 #else /* POSIX diff --git a/libstudxml/details/expat/xmlparse.c b/libstudxml/details/expat/xmlparse.c index 5bc6373..46871c1 100644 --- a/libstudxml/details/expat/xmlparse.c +++ b/libstudxml/details/expat/xmlparse.c @@ -6,12 +6,19 @@ #include <string.h> /* memset(), memcpy() */ #include <assert.h> #include <limits.h> /* UINT_MAX */ -#include <time.h> /* time() */ #define XML_BUILDING_EXPAT 1 #include <libstudxml/details/expat/config.h> +#ifdef COMPILED_FROM_DSP +#define getpid GetCurrentProcessId +#else +#include <sys/time.h> /* gettimeofday() */ +#include <sys/types.h> /* getpid() */ +#include <unistd.h> /* getpid() */ +#endif + #include <libstudxml/details/expat/ascii.h> #include <libstudxml/details/expat/expat.h> @@ -236,7 +243,7 @@ typedef struct { typedef struct { unsigned long version; - unsigned long hash; + size_t hash; const XML_Char *uriName; } NS_ATT; @@ -321,7 +328,7 @@ 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); + XML_Bool haveMore, XML_Bool allowClosingDoctype); static enum XML_Error processInternalEntity(XML_Parser parser, ENTITY *entity, XML_Bool betweenDecl); @@ -422,7 +429,7 @@ 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 size_t generate_hash_secret_salt(XML_Parser parser); static XML_Bool startParsing(XML_Parser parser); static XML_Parser @@ -546,7 +553,7 @@ struct XML_ParserStruct { XML_Bool m_useForeignDTD; enum XML_ParamEntityParsing m_paramEntityParsing; #endif - unsigned long m_hash_secret_salt; + size_t m_hash_secret_salt; }; #define MALLOC(s) (parser->m_mem.malloc_fcn((s))) @@ -680,12 +687,39 @@ static const XML_Char implicitContext[] = { ASCII_s, ASCII_p, ASCII_a, ASCII_c, ASCII_e, '\0' }; -static unsigned long -generate_hash_secret_salt(void) +static size_t +gather_time_entropy(void) { - unsigned int seed = time(NULL) % UINT_MAX; - srand(seed); - return rand(); +#ifdef COMPILED_FROM_DSP + FILETIME ft; + GetSystemTimeAsFileTime(&ft); /* never fails */ + return ft.dwHighDateTime ^ ft.dwLowDateTime; +#else + struct timeval tv; + int gettimeofday_res; + + gettimeofday_res = gettimeofday(&tv, NULL); + assert (gettimeofday_res == 0); + + /* Microseconds time is <20 bits entropy */ + return tv.tv_usec; +#endif +} + +static size_t +generate_hash_secret_salt(XML_Parser parser) +{ + /* Process ID is 0 bits entropy if attacker has local access + * XML_Parser address is few bits of entropy if attacker has local access */ + const size_t entropy = + gather_time_entropy() ^ getpid() ^ (size_t)parser; + + /* Factors are 2^31-1 and 2^61-1 (Mersenne primes M31 and M61) */ + if (sizeof(size_t) == 4) { + return entropy * (size_t)2147483647; + } else { + return entropy * (size_t)2305843009213693951; + } } static XML_Bool /* only valid for root parser */ @@ -693,7 +727,7 @@ 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(); + hash_secret_salt = generate_hash_secret_salt(parser); if (ns) { /* implicit context only set for root parser, since child parsers (i.e. external entity parsers) will inherit it @@ -1018,7 +1052,7 @@ XML_ExternalEntityParserCreate(XML_Parser oldParser, 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; + size_t oldhash_secret_salt = hash_secret_salt; #ifdef XML_DTD if (!context) @@ -1477,7 +1511,7 @@ XML_SetParamEntityParsing(XML_Parser parser, int XMLCALL XML_SetHashSalt(XML_Parser parser, - unsigned long hash_salt) + unsigned long hash_salt) /* should be size_t */ { /* block after XML_Parse()/XML_ParseBuffer() has been called */ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) @@ -1573,11 +1607,14 @@ XML_Parse(XML_Parser parser, const char *s, int len, int isFinal) 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)); + /* avoid _signed_ integer overflow */ + char *temp = NULL; + const int bytesToAllocate = (int)((unsigned)len * 2U); + if (bytesToAllocate > 0) { + temp = (buffer == NULL + ? (char *)MALLOC(bytesToAllocate) + : (char *)REALLOC(buffer, bytesToAllocate)); + } if (temp == NULL) { errorCode = XML_ERROR_NO_MEMORY; eventPtr = eventEndPtr = NULL; @@ -1585,7 +1622,7 @@ XML_Parse(XML_Parser parser, const char *s, int len, int isFinal) return XML_STATUS_ERROR; } buffer = temp; - bufferLim = buffer + len * 2; + bufferLim = buffer + bytesToAllocate; } memcpy(buffer, end, nLeftOver); } @@ -1668,6 +1705,12 @@ XML_ParseBuffer(XML_Parser parser, int len, int isFinal) void * XMLCALL XML_GetBuffer(XML_Parser parser, int len) { +/* BEGIN MOZILLA CHANGE (sanity check len) */ + if (len < 0) { + errorCode = XML_ERROR_NO_MEMORY; + return NULL; + } +/* END MOZILLA CHANGE */ switch (ps_parsing) { case XML_SUSPENDED: errorCode = XML_ERROR_SUSPENDED; @@ -1679,8 +1722,14 @@ XML_GetBuffer(XML_Parser parser, int len) } if (len > bufferLim - bufferEnd) { - /* FIXME avoid integer overflow */ - int neededSize = len + (int)(bufferEnd - bufferPtr); + /* Do not invoke signed arithmetic overflow: */ + int neededSize = (int) ((unsigned)len + (unsigned)(bufferEnd - bufferPtr)); +/* BEGIN MOZILLA CHANGE (sanity check neededSize) */ + if (neededSize < 0) { + errorCode = XML_ERROR_NO_MEMORY; + return NULL; + } +/* END MOZILLA CHANGE */ #ifdef XML_CONTEXT_BYTES int keep = (int)(bufferPtr - buffer); @@ -1708,8 +1757,17 @@ XML_GetBuffer(XML_Parser parser, int len) if (bufferSize == 0) bufferSize = INIT_BUFFER_SIZE; do { - bufferSize *= 2; - } while (bufferSize < neededSize); + /* Do not invoke signed arithmetic overflow: */ + bufferSize = (int) (2U * (unsigned) bufferSize); +/* BEGIN MOZILLA CHANGE (prevent infinite loop on overflow) */ + } while (bufferSize < neededSize && bufferSize > 0); +/* END MOZILLA CHANGE */ +/* BEGIN MOZILLA CHANGE (sanity check bufferSize) */ + if (bufferSize <= 0) { + errorCode = XML_ERROR_NO_MEMORY; + return NULL; + } +/* END MOZILLA CHANGE */ newBuf = (char *)MALLOC(bufferSize); if (newBuf == 0) { errorCode = XML_ERROR_NO_MEMORY; @@ -2405,11 +2463,11 @@ doContent(XML_Parser parser, for (;;) { int bufSize; int convLen; - XmlConvert(enc, + const enum XML_Convert_Result convert_res = XmlConvert(enc, &fromPtr, rawNameEnd, (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1); convLen = (int)(toPtr - (XML_Char *)tag->buf); - if (fromPtr == rawNameEnd) { + if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) { tag->name.strLen = convLen; break; } @@ -2630,11 +2688,11 @@ doContent(XML_Parser parser, if (MUST_CONVERT(enc, s)) { for (;;) { ICHAR *dataPtr = (ICHAR *)dataBuf; - XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); + const enum XML_Convert_Result convert_res = XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); *eventEndPP = s; charDataHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf)); - if (s == next) + if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) break; *eventPP = s; } @@ -2867,7 +2925,7 @@ storeAtts(XML_Parser parser, const ENCODING *enc, and clear flags that say whether attributes were specified */ i = 0; if (nPrefixes) { - int j; /* hash table index */ + size_t 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) */ @@ -2898,7 +2956,7 @@ storeAtts(XML_Parser parser, const ENCODING *enc, if (s[-1] == 2) { /* prefixed */ ATTRIBUTE_ID *id; const BINDING *b; - unsigned long uriHash = hash_secret_salt; + size_t uriHash = hash_secret_salt; ((XML_Char *)s)[-1] = 0; /* clear flag */ id = (ATTRIBUTE_ID *)lookup(parser, &dtd->attributeIds, s, 0); b = id->prefix->binding; @@ -2906,7 +2964,7 @@ storeAtts(XML_Parser parser, const ENCODING *enc, return XML_ERROR_UNBOUND_PREFIX; /* as we expand the name we also calculate its hash value */ - for (j = 0; j < b->uriLen; j++) { + for (j = 0; j < (size_t)b->uriLen; j++) { const XML_Char c = b->uri[j]; if (!poolAppendChar(&tempPool, c)) return XML_ERROR_NO_MEMORY; @@ -2925,7 +2983,7 @@ storeAtts(XML_Parser parser, const ENCODING *enc, Derived from code in lookup(parser, HASH_TABLE *table, ...). */ unsigned char step = 0; - unsigned long mask = nsAttsSize - 1; + size_t mask = nsAttsSize - 1; j = uriHash & mask; /* index into hash table */ while (nsAtts[j].version == version) { /* for speed we compare stored hash values first */ @@ -3238,11 +3296,11 @@ doCdataSection(XML_Parser parser, if (MUST_CONVERT(enc, s)) { for (;;) { ICHAR *dataPtr = (ICHAR *)dataBuf; - XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); + const enum XML_Convert_Result convert_res = XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); *eventEndPP = next; charDataHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf)); - if (s == next) + if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) break; *eventPP = s; } @@ -3634,6 +3692,14 @@ entityValueInitProcessor(XML_Parser parser, *nextPtr = next; return XML_ERROR_NONE; } + /* If we get this token, we have the start of what might be a + normal tag, but not a declaration (i.e. it doesn't begin with + "<!"). In a DTD context, that isn't legal. + */ + else if (tok == XML_TOK_INSTANCE_START) { + *nextPtr = next; + return XML_ERROR_SYNTAX; + } start = next; eventPtr = start; } @@ -3677,7 +3743,7 @@ externalParEntProcessor(XML_Parser parser, processor = prologProcessor; return doProlog(parser, encoding, s, end, tok, next, - nextPtr, (XML_Bool)!ps_finalBuffer); + nextPtr, (XML_Bool)!ps_finalBuffer, XML_TRUE); } static enum XML_Error PTRCALL @@ -3727,7 +3793,7 @@ prologProcessor(XML_Parser parser, const char *next = s; int tok = XmlPrologTok(encoding, s, end, &next); return doProlog(parser, encoding, s, end, tok, next, - nextPtr, (XML_Bool)!ps_finalBuffer); + nextPtr, (XML_Bool)!ps_finalBuffer, XML_TRUE); } static enum XML_Error @@ -3738,7 +3804,8 @@ doProlog(XML_Parser parser, int tok, const char *next, const char **nextPtr, - XML_Bool haveMore) + XML_Bool haveMore, + XML_Bool allowClosingDoctype) { #ifdef XML_DTD static const XML_Char externalSubsetName[] = { ASCII_HASH , '\0' }; @@ -3914,6 +3981,11 @@ doProlog(XML_Parser parser, } break; case XML_ROLE_DOCTYPE_CLOSE: + if (allowClosingDoctype != XML_TRUE) { + /* Must not close doctype from within expanded parameter entities */ + return XML_ERROR_INVALID_TOKEN; + } + if (doctypeName) { startDoctypeDeclHandler(handlerArg, doctypeName, doctypeSysid, doctypePubid, 0); @@ -4815,7 +4887,7 @@ processInternalEntity(XML_Parser parser, ENTITY *entity, if (entity->is_param) { int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next); result = doProlog(parser, internalEncoding, textStart, textEnd, tok, - next, &next, XML_FALSE); + next, &next, XML_FALSE, XML_FALSE); } else #endif /* XML_DTD */ @@ -4860,7 +4932,7 @@ internalEntityProcessor(XML_Parser parser, if (entity->is_param) { int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next); result = doProlog(parser, internalEncoding, textStart, textEnd, tok, - next, &next, XML_FALSE); + next, &next, XML_FALSE, XML_TRUE); } else #endif /* XML_DTD */ @@ -4887,7 +4959,7 @@ internalEntityProcessor(XML_Parser parser, processor = prologProcessor; tok = XmlPrologTok(encoding, s, end, &next); return doProlog(parser, encoding, s, end, tok, next, nextPtr, - (XML_Bool)!ps_finalBuffer); + (XML_Bool)!ps_finalBuffer, XML_TRUE); } else #endif /* XML_DTD */ @@ -5323,6 +5395,7 @@ reportDefault(XML_Parser parser, const ENCODING *enc, const char *s, const char *end) { if (MUST_CONVERT(enc, s)) { + enum XML_Convert_Result convert_res; const char **eventPP; const char **eventEndPP; if (enc == encoding) { @@ -5335,11 +5408,11 @@ reportDefault(XML_Parser parser, const ENCODING *enc, } do { ICHAR *dataPtr = (ICHAR *)dataBuf; - XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd); + convert_res = XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd); *eventEndPP = s; defaultHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf)); *eventPP = s; - } while (s != end); + } while ((convert_res != XML_CONVERT_COMPLETED) && (convert_res != XML_CONVERT_INPUT_INCOMPLETE)); } else defaultHandler(handlerArg, (XML_Char *)s, (int)((XML_Char *)end - (XML_Char *)s)); @@ -5414,6 +5487,7 @@ setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *elementType) else poolDiscard(&dtd->pool); elementType->prefix = prefix; + break; } } @@ -5951,10 +6025,10 @@ keyeq(KEY s1, KEY s2) return XML_FALSE; } -static unsigned long FASTCALL +static size_t FASTCALL hash(XML_Parser parser, KEY s) { - unsigned long h = hash_secret_salt; + size_t h = hash_secret_salt; while (*s) h = CHAR_HASH(h, *s++); return h; @@ -5981,8 +6055,8 @@ lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize) i = hash(parser, name) & ((unsigned long)table->size - 1); } else { - unsigned long h = hash(parser, name); - unsigned long mask = (unsigned long)table->size - 1; + size_t h = hash(parser, name); + size_t mask = table->size - 1; unsigned char step = 0; i = h & mask; while (table->v[i]) { @@ -5999,7 +6073,7 @@ lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize) 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 newMask = newSize - 1; size_t tsize = newSize * sizeof(NAMED *); NAMED **newV = (NAMED **)table->mem->malloc_fcn(tsize); if (!newV) @@ -6007,7 +6081,7 @@ lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize) 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 newHash = hash(parser, table->v[i]->name); size_t j = newHash & newMask; step = 0; while (newV[j]) { @@ -6142,8 +6216,8 @@ poolAppend(STRING_POOL *pool, const ENCODING *enc, if (!pool->ptr && !poolGrow(pool)) return NULL; for (;;) { - XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end); - if (ptr == end) + const enum XML_Convert_Result convert_res = XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end); + if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) break; if (!poolGrow(pool)) return NULL; @@ -6227,8 +6301,13 @@ poolGrow(STRING_POOL *pool) } } if (pool->blocks && pool->start == pool->blocks->s) { - int blockSize = (int)(pool->end - pool->start)*2; - BLOCK *temp = (BLOCK *) + BLOCK *temp; + int blockSize = (int)((unsigned)(pool->end - pool->start)*2U); + + if (blockSize < 0) + return XML_FALSE; + + temp = (BLOCK *) pool->mem->realloc_fcn(pool->blocks, (offsetof(BLOCK, s) + blockSize * sizeof(XML_Char))); @@ -6243,6 +6322,10 @@ poolGrow(STRING_POOL *pool) else { BLOCK *tem; int blockSize = (int)(pool->end - pool->start); + + if (blockSize < 0) + return XML_FALSE; + if (blockSize < INIT_BLOCK_SIZE) blockSize = INIT_BLOCK_SIZE; else diff --git a/libstudxml/details/expat/xmltok.c b/libstudxml/details/expat/xmltok.c index 5a8f36f..a36735c 100644 --- a/libstudxml/details/expat/xmltok.c +++ b/libstudxml/details/expat/xmltok.c @@ -336,11 +336,12 @@ enum { /* UTF8_cvalN is value of masked first byte of N byte sequence */ UTF8_cval4 = 0xf0 }; -static void PTRCALL +static enum XML_Convert_Result PTRCALL utf8_toUtf8(const ENCODING *enc, const char **fromP, const char *fromLim, char **toP, const char *toLim) { + enum XML_Convert_Result res = XML_CONVERT_COMPLETED; char *to; const char *from; @@ -348,30 +349,45 @@ utf8_toUtf8(const ENCODING *enc, if (fromLim - *fromP > toLim - *toP) { /* Avoid copying partial characters. */ + res = XML_CONVERT_OUTPUT_EXHAUSTED; 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++) + for (to = *toP, from = *fromP; (from < fromLim) && (to < toLim); from++, to++) *to = *from; *fromP = from; *toP = to; + + if ((to == toLim) && (from < fromLim)) + return XML_CONVERT_OUTPUT_EXHAUSTED; + else + return res; } -static void PTRCALL +static enum XML_Convert_Result PTRCALL utf8_toUtf16(const ENCODING *enc, const char **fromP, const char *fromLim, unsigned short **toP, const unsigned short *toLim) { + enum XML_Convert_Result res = XML_CONVERT_COMPLETED; unsigned short *to = *toP; const char *from = *fromP; - while (from != fromLim && to != toLim) { + while (from < fromLim && to < toLim) { switch (((struct normal_encoding *)enc)->type[(unsigned char)*from]) { case BT_LEAD2: + if (fromLim - from < 2) { + res = XML_CONVERT_INPUT_INCOMPLETE; + break; + } *to++ = (unsigned short)(((from[0] & 0x1f) << 6) | (from[1] & 0x3f)); from += 2; break; case BT_LEAD3: + if (fromLim - from < 3) { + res = XML_CONVERT_INPUT_INCOMPLETE; + break; + } *to++ = (unsigned short)(((from[0] & 0xf) << 12) | ((from[1] & 0x3f) << 6) | (from[2] & 0x3f)); from += 3; @@ -379,8 +395,14 @@ utf8_toUtf16(const ENCODING *enc, case BT_LEAD4: { unsigned long n; - if (to + 1 == toLim) + if (toLim - to < 2) { + res = XML_CONVERT_OUTPUT_EXHAUSTED; goto after; + } + if (fromLim - from < 4) { + res = XML_CONVERT_INPUT_INCOMPLETE; + goto after; + } n = ((from[0] & 0x7) << 18) | ((from[1] & 0x3f) << 12) | ((from[2] & 0x3f) << 6) | (from[3] & 0x3f); n -= 0x10000; @@ -398,6 +420,7 @@ utf8_toUtf16(const ENCODING *enc, after: *fromP = from; *toP = to; + return res; } #ifdef XML_NS @@ -446,7 +469,7 @@ static const struct normal_encoding internal_utf8_encoding = { STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_) }; -static void PTRCALL +static enum XML_Convert_Result PTRCALL latin1_toUtf8(const ENCODING *enc, const char **fromP, const char *fromLim, char **toP, const char *toLim) @@ -456,32 +479,37 @@ latin1_toUtf8(const ENCODING *enc, for (;;) { unsigned char c; if (*fromP == fromLim) - break; + return XML_CONVERT_COMPLETED; c = (unsigned char)**fromP; if (c & 0x80) { if (toLim - *toP < 2) - break; + return XML_CONVERT_OUTPUT_EXHAUSTED; *(*toP)++ = (char)((c >> 6) | UTF8_cval2); *(*toP)++ = (char)((c & 0x3f) | 0x80); (*fromP)++; } else { if (*toP == toLim) - break; + return XML_CONVERT_OUTPUT_EXHAUSTED; *(*toP)++ = *(*fromP)++; } } } -static void PTRCALL +static enum XML_Convert_Result 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) + while (*fromP < fromLim && *toP < toLim) *(*toP)++ = (unsigned char)*(*fromP)++; + + if ((*toP == toLim) && (*fromP < fromLim)) + return XML_CONVERT_OUTPUT_EXHAUSTED; + else + return XML_CONVERT_COMPLETED; } #ifdef XML_NS @@ -508,15 +536,20 @@ static const struct normal_encoding latin1_encoding = { STANDARD_VTABLE(sb_) ZERO_VTABLE }; -static void PTRCALL +static enum XML_Convert_Result PTRCALL ascii_toUtf8(const ENCODING *enc, const char **fromP, const char *fromLim, char **toP, const char *toLim) { UNUSED(enc); - while (*fromP != fromLim && *toP != toLim) + while (*fromP < fromLim && *toP < toLim) *(*toP)++ = *(*fromP)++; + + if ((*toP == toLim) && (*fromP < fromLim)) + return XML_CONVERT_OUTPUT_EXHAUSTED; + else + return XML_CONVERT_COMPLETED; } #ifdef XML_NS @@ -563,14 +596,15 @@ unicode_byte_type(char hi, char lo) } #define DEFINE_UTF16_TO_UTF8(E) \ -static void PTRCALL \ +static enum XML_Convert_Result PTRCALL \ E ## toUtf8(const ENCODING *enc, \ const char **fromP, const char *fromLim, \ char **toP, const char *toLim) \ { \ - const char *from; \ + const char *from = *fromP; \ UNUSED(enc); \ - for (from = *fromP; from != fromLim; from += 2) { \ + fromLim = from + (((fromLim - from) >> 1) << 1); /* shrink to even */ \ + for (; from < fromLim; from += 2) { \ int plane; \ unsigned char lo2; \ unsigned char lo = GET_LO(from); \ @@ -580,7 +614,7 @@ E ## toUtf8(const ENCODING *enc, \ if (lo < 0x80) { \ if (*toP == toLim) { \ *fromP = from; \ - return; \ + return XML_CONVERT_OUTPUT_EXHAUSTED; \ } \ *(*toP)++ = lo; \ break; \ @@ -590,7 +624,7 @@ E ## toUtf8(const ENCODING *enc, \ case 0x4: case 0x5: case 0x6: case 0x7: \ if (toLim - *toP < 2) { \ *fromP = from; \ - return; \ + return XML_CONVERT_OUTPUT_EXHAUSTED; \ } \ *(*toP)++ = ((lo >> 6) | (hi << 2) | UTF8_cval2); \ *(*toP)++ = ((lo & 0x3f) | 0x80); \ @@ -598,7 +632,7 @@ E ## toUtf8(const ENCODING *enc, \ default: \ if (toLim - *toP < 3) { \ *fromP = from; \ - return; \ + return XML_CONVERT_OUTPUT_EXHAUSTED; \ } \ /* 16 bits divided 4, 6, 6 amongst 3 bytes */ \ *(*toP)++ = ((hi >> 4) | UTF8_cval3); \ @@ -608,7 +642,11 @@ E ## toUtf8(const ENCODING *enc, \ case 0xD8: case 0xD9: case 0xDA: case 0xDB: \ if (toLim - *toP < 4) { \ *fromP = from; \ - return; \ + return XML_CONVERT_OUTPUT_EXHAUSTED; \ + } \ + if (fromLim - from < 4) { \ + *fromP = from; \ + return XML_CONVERT_INPUT_INCOMPLETE; \ } \ plane = (((hi & 0x3) << 2) | ((lo >> 6) & 0x3)) + 1; \ *(*toP)++ = ((plane >> 2) | UTF8_cval4); \ @@ -624,21 +662,33 @@ E ## toUtf8(const ENCODING *enc, \ } \ } \ *fromP = from; \ + if (from < fromLim) \ + return XML_CONVERT_INPUT_INCOMPLETE; \ + else \ + return XML_CONVERT_COMPLETED; \ } #define DEFINE_UTF16_TO_UTF16(E) \ -static void PTRCALL \ +static enum XML_Convert_Result PTRCALL \ E ## toUtf16(const ENCODING *enc, \ const char **fromP, const char *fromLim, \ unsigned short **toP, const unsigned short *toLim) \ { \ - UNUSED(enc); \ + enum XML_Convert_Result res = XML_CONVERT_COMPLETED; \ + UNUSED(enc); \ + fromLim = *fromP + (((fromLim - *fromP) >> 1) << 1); /* shrink to even */ \ /* Avoid copying first half only of surrogate */ \ if (fromLim - *fromP > ((toLim - *toP) << 1) \ - && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) \ + && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) { \ fromLim -= 2; \ - for (; *fromP != fromLim && *toP != toLim; *fromP += 2) \ + res = XML_CONVERT_INPUT_INCOMPLETE; \ + } \ + for (; *fromP < fromLim && *toP < toLim; *fromP += 2) \ *(*toP)++ = (GET_HI(*fromP) << 8) | GET_LO(*fromP); \ + if ((*toP == toLim) && (*fromP < fromLim)) \ + return XML_CONVERT_OUTPUT_EXHAUSTED; \ + else \ + return res; \ } #define SET2(ptr, ch) \ @@ -1318,7 +1368,7 @@ unknown_isInvalid(const ENCODING *enc, const char *p) return (c & ~0xFFFF) || checkCharRefNumber(c) < 0; } -static void PTRCALL +static enum XML_Convert_Result PTRCALL unknown_toUtf8(const ENCODING *enc, const char **fromP, const char *fromLim, char **toP, const char *toLim) @@ -1329,21 +1379,21 @@ unknown_toUtf8(const ENCODING *enc, const char *utf8; int n; if (*fromP == fromLim) - break; + return XML_CONVERT_COMPLETED; 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; + return XML_CONVERT_OUTPUT_EXHAUSTED; utf8 = buf; *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP] - (BT_LEAD2 - 2)); } else { if (n > toLim - *toP) - break; + return XML_CONVERT_OUTPUT_EXHAUSTED; (*fromP)++; } do { @@ -1352,13 +1402,13 @@ unknown_toUtf8(const ENCODING *enc, } } -static void PTRCALL +static enum XML_Convert_Result 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) { + while (*fromP < fromLim && *toP < toLim) { unsigned short c = uenc->utf16[(unsigned char)**fromP]; if (c == 0) { c = (unsigned short) @@ -1370,6 +1420,11 @@ unknown_toUtf16(const ENCODING *enc, (*fromP)++; *(*toP)++ = c; } + + if ((*toP == toLim) && (*fromP < fromLim)) + return XML_CONVERT_OUTPUT_EXHAUSTED; + else + return XML_CONVERT_COMPLETED; } ENCODING * @@ -1533,7 +1588,7 @@ initScan(const ENCODING * const *encodingTable, { const ENCODING **encPtr; - if (ptr == end) + if (ptr >= end) return XML_TOK_NONE; encPtr = enc->encPtr; if (ptr + 1 == end) { diff --git a/libstudxml/details/expat/xmltok.h b/libstudxml/details/expat/xmltok.h index ca867aa..752007e 100644 --- a/libstudxml/details/expat/xmltok.h +++ b/libstudxml/details/expat/xmltok.h @@ -130,6 +130,12 @@ typedef int (PTRCALL *SCANNER)(const ENCODING *, const char *, const char **); +enum XML_Convert_Result { + XML_CONVERT_COMPLETED = 0, + XML_CONVERT_INPUT_INCOMPLETE = 1, + XML_CONVERT_OUTPUT_EXHAUSTED = 2 /* and therefore potentially input remaining as well */ +}; + struct encoding { SCANNER scanners[XML_N_STATES]; SCANNER literalScanners[XML_N_LITERAL_TYPES]; @@ -158,12 +164,12 @@ struct encoding { const char *ptr, const char *end, const char **badPtr); - void (PTRCALL *utf8Convert)(const ENCODING *enc, + enum XML_Convert_Result (PTRCALL *utf8Convert)(const ENCODING *enc, const char **fromP, const char *fromLim, char **toP, const char *toLim); - void (PTRCALL *utf16Convert)(const ENCODING *enc, + enum XML_Convert_Result (PTRCALL *utf16Convert)(const ENCODING *enc, const char **fromP, const char *fromLim, unsigned short **toP, diff --git a/libstudxml/details/expat/xmltok_impl.c b/libstudxml/details/expat/xmltok_impl.c index de4d21a..5410278 100644 --- a/libstudxml/details/expat/xmltok_impl.c +++ b/libstudxml/details/expat/xmltok_impl.c @@ -93,13 +93,13 @@ static int PTRCALL PREFIX(scanComment)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { - if (ptr != end) { + if (ptr < end) { if (!CHAR_MATCHES(enc, ptr, ASCII_MINUS)) { *nextTokPtr = ptr; return XML_TOK_INVALID; } ptr += MINBPC(enc); - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { INVALID_CASES(ptr, nextTokPtr) case BT_MINUS: @@ -147,7 +147,7 @@ PREFIX(scanDecl)(const ENCODING *enc, const char *ptr, *nextTokPtr = ptr; return XML_TOK_INVALID; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { case BT_PERCNT: if (ptr + MINBPC(enc) == end) @@ -236,7 +236,7 @@ PREFIX(scanPi)(const ENCODING *enc, const char *ptr, *nextTokPtr = ptr; return XML_TOK_INVALID; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) case BT_S: case BT_CR: case BT_LF: @@ -245,7 +245,7 @@ PREFIX(scanPi)(const ENCODING *enc, const char *ptr, return XML_TOK_INVALID; } ptr += MINBPC(enc); - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { INVALID_CASES(ptr, nextTokPtr) case BT_QUEST: @@ -311,7 +311,7 @@ static int PTRCALL PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { - if (ptr == end) + if (ptr >= end) return XML_TOK_NONE; if (MINBPC(enc) > 1) { size_t n = end - ptr; @@ -354,7 +354,7 @@ PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr, ptr += MINBPC(enc); break; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { #define LEAD_CASE(n) \ case BT_LEAD ## n: \ @@ -397,11 +397,11 @@ PREFIX(scanEndTag)(const ENCODING *enc, const char *ptr, *nextTokPtr = ptr; return XML_TOK_INVALID; } - while (ptr != end) { + 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)) { + for (ptr += MINBPC(enc); ptr < end; ptr += MINBPC(enc)) { switch (BYTE_TYPE(enc, ptr)) { case BT_S: case BT_CR: case BT_LF: break; @@ -438,7 +438,7 @@ static int PTRCALL PREFIX(scanHexCharRef)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { - if (ptr != end) { + if (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { case BT_DIGIT: case BT_HEX: @@ -447,7 +447,7 @@ PREFIX(scanHexCharRef)(const ENCODING *enc, const char *ptr, *nextTokPtr = ptr; return XML_TOK_INVALID; } - for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) { + for (ptr += MINBPC(enc); ptr < end; ptr += MINBPC(enc)) { switch (BYTE_TYPE(enc, ptr)) { case BT_DIGIT: case BT_HEX: @@ -470,7 +470,7 @@ static int PTRCALL PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { - if (ptr != end) { + if (ptr < end) { if (CHAR_MATCHES(enc, ptr, ASCII_x)) return PREFIX(scanHexCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr); switch (BYTE_TYPE(enc, ptr)) { @@ -480,7 +480,7 @@ PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr, *nextTokPtr = ptr; return XML_TOK_INVALID; } - for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) { + for (ptr += MINBPC(enc); ptr < end; ptr += MINBPC(enc)) { switch (BYTE_TYPE(enc, ptr)) { case BT_DIGIT: break; @@ -512,7 +512,7 @@ PREFIX(scanRef)(const ENCODING *enc, const char *ptr, const char *end, *nextTokPtr = ptr; return XML_TOK_INVALID; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) case BT_SEMI: @@ -535,7 +535,7 @@ PREFIX(scanAtts)(const ENCODING *enc, const char *ptr, const char *end, #ifdef XML_NS int hadColon = 0; #endif - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) #ifdef XML_NS @@ -722,7 +722,7 @@ PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end, hadColon = 0; #endif /* we have a start-tag */ - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) #ifdef XML_NS @@ -746,7 +746,7 @@ PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end, case BT_S: case BT_CR: case BT_LF: { ptr += MINBPC(enc); - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) case BT_GT: @@ -791,7 +791,7 @@ static int PTRCALL PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { - if (ptr == end) + if (ptr >= end) return XML_TOK_NONE; if (MINBPC(enc) > 1) { size_t n = end - ptr; @@ -838,7 +838,7 @@ PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end, ptr += MINBPC(enc); break; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { #define LEAD_CASE(n) \ case BT_LEAD ## n: \ @@ -901,7 +901,7 @@ PREFIX(scanPercent)(const ENCODING *enc, const char *ptr, const char *end, *nextTokPtr = ptr; return XML_TOK_INVALID; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) case BT_SEMI: @@ -927,7 +927,7 @@ PREFIX(scanPoundName)(const ENCODING *enc, const char *ptr, const char *end, *nextTokPtr = ptr; return XML_TOK_INVALID; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) case BT_CR: case BT_LF: case BT_S: @@ -947,7 +947,7 @@ PREFIX(scanLit)(int open, const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { - while (ptr != end) { + while (ptr < end) { int t = BYTE_TYPE(enc, ptr); switch (t) { INVALID_CASES(ptr, nextTokPtr) @@ -979,7 +979,7 @@ PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { int tok; - if (ptr == end) + if (ptr >= end) return XML_TOK_NONE; if (MINBPC(enc) > 1) { size_t n = end - ptr; @@ -1147,7 +1147,7 @@ PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end, *nextTokPtr = ptr; return XML_TOK_INVALID; } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) case BT_GT: case BT_RPAR: case BT_COMMA: @@ -1210,10 +1210,10 @@ PREFIX(attributeValueTok)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { const char *start; - if (ptr == end) + if (ptr >= end) return XML_TOK_NONE; start = ptr; - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { #define LEAD_CASE(n) \ case BT_LEAD ## n: ptr += n; break; @@ -1268,10 +1268,10 @@ PREFIX(entityValueTok)(const ENCODING *enc, const char *ptr, const char *end, const char **nextTokPtr) { const char *start; - if (ptr == end) + if (ptr >= end) return XML_TOK_NONE; start = ptr; - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { #define LEAD_CASE(n) \ case BT_LEAD ## n: ptr += n; break; @@ -1332,7 +1332,7 @@ PREFIX(ignoreSectionTok)(const ENCODING *enc, const char *ptr, end = ptr + n; } } - while (ptr != end) { + while (ptr < end) { switch (BYTE_TYPE(enc, ptr)) { INVALID_CASES(ptr, nextTokPtr) case BT_LT: @@ -1379,7 +1379,7 @@ PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end, { ptr += MINBPC(enc); end -= MINBPC(enc); - for (; ptr != end; ptr += MINBPC(enc)) { + for (; ptr < end; ptr += MINBPC(enc)) { switch (BYTE_TYPE(enc, ptr)) { case BT_DIGIT: case BT_HEX: @@ -1773,7 +1773,7 @@ PREFIX(updatePosition)(const ENCODING *enc, case BT_CR: pos->lineNumber++; ptr += MINBPC(enc); - if (ptr != end && BYTE_TYPE(enc, ptr) == BT_LF) + if (ptr < end && BYTE_TYPE(enc, ptr) == BT_LF) ptr += MINBPC(enc); pos->columnNumber = (XML_Size)-1; break; |