From c562469eb387946a31e6d80fd169ad36201fa49a Mon Sep 17 00:00:00 2001 From: Guenter Obiltschnig Date: Wed, 21 Jun 2017 15:15:55 +0200 Subject: [PATCH 1/3] upgrade bundled expat to 2.2.1; apply POCO-specific changes --- XML/include/Poco/XML/expat.h | 22 +- XML/include/Poco/XML/expat_external.h | 5 +- XML/src/siphash.h | 344 ++++++++++++++ XML/src/xmlparse.cpp | 631 ++++++++++++++++++++------ XML/src/xmlrole.c | 6 - XML/src/xmltok.c | 30 +- 6 files changed, 874 insertions(+), 164 deletions(-) create mode 100644 XML/src/siphash.h diff --git a/XML/include/Poco/XML/expat.h b/XML/include/Poco/XML/expat.h index 086e24b39..28b0f954d 100644 --- a/XML/include/Poco/XML/expat.h +++ b/XML/include/Poco/XML/expat.h @@ -95,7 +95,9 @@ enum XML_Error { /* Added in 2.0. */ XML_ERROR_RESERVED_PREFIX_XML, XML_ERROR_RESERVED_PREFIX_XMLNS, - XML_ERROR_RESERVED_NAMESPACE_URI + XML_ERROR_RESERVED_NAMESPACE_URI, + /* Added in 2.2.1. */ + XML_ERROR_INVALID_ARGUMENT }; enum XML_Content_Type { @@ -706,6 +708,7 @@ XML_UseParserAsHandlerArg(XML_Parser parser); 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. + Note: If parser == NULL, returns XML_ERROR_INVALID_ARGUMENT. */ XMLPARSEAPI(enum XML_Error) XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD); @@ -729,15 +732,16 @@ XML_GetBase(XML_Parser parser); 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. + XML_StartElementHandler. Returns -1 if parser == NULL. */ 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. + XML_StartElementHandler, or -1 if there is no ID attribute or + parser == NULL. 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); @@ -901,6 +905,7 @@ enum XML_ParamEntityParsing { 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. + Note: If parser == NULL, the function will do nothing and return 0. */ XMLPARSEAPI(int) XML_SetParamEntityParsing(XML_Parser parser, @@ -910,6 +915,7 @@ XML_SetParamEntityParsing(XML_Parser parser, 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. + Note: If parser == NULL, the function will do nothing and return 0. */ XMLPARSEAPI(int) XML_SetHashSalt(XML_Parser parser, @@ -936,6 +942,10 @@ XML_GetErrorCode(XML_Parser parser); 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. + + Note: XML_GetCurrentLineNumber and XML_GetCurrentColumnNumber + return 0 to indicate an error. + Note: XML_GetCurrentByteIndex returns -1 to indicate an error. */ XMLPARSEAPI(XML_Size) XML_GetCurrentLineNumber(XML_Parser parser); XMLPARSEAPI(XML_Size) XML_GetCurrentColumnNumber(XML_Parser parser); @@ -1039,7 +1049,7 @@ XML_GetFeatureList(void); */ #define XML_MAJOR_VERSION 2 #define XML_MINOR_VERSION 2 -#define XML_MICRO_VERSION 0 +#define XML_MICRO_VERSION 1 #ifdef __cplusplus } diff --git a/XML/include/Poco/XML/expat_external.h b/XML/include/Poco/XML/expat_external.h index aa08a2f84..892eb4bbe 100644 --- a/XML/include/Poco/XML/expat_external.h +++ b/XML/include/Poco/XML/expat_external.h @@ -93,7 +93,10 @@ extern "C" { #endif #ifdef XML_UNICODE_WCHAR_T -#define XML_UNICODE +# define XML_UNICODE +# if defined(__SIZEOF_WCHAR_T__) && (__SIZEOF_WCHAR_T__ != 2) +# error "sizeof(wchar_t) != 2; Need -fshort-wchar for both Expat and libc" +# endif #endif #ifdef XML_UNICODE /* Information is UTF-16 encoded. */ diff --git a/XML/src/siphash.h b/XML/src/siphash.h new file mode 100644 index 000000000..23b56d2ae --- /dev/null +++ b/XML/src/siphash.h @@ -0,0 +1,344 @@ +/* ========================================================================== + * siphash.h - SipHash-2-4 in a single header file + * -------------------------------------------------------------------------- + * Derived by William Ahern from the reference implementation[1] published[2] + * by Jean-Philippe Aumasson and Daniel J. Berstein. Licensed in kind. + * by Jean-Philippe Aumasson and Daniel J. Berstein. + * Minimal changes by Sebastian Pipping on top, details below. + * Licensed under the CC0 Public Domain Dedication license. + * + * 1. https://www.131002.net/siphash/siphash24.c + * 2. https://www.131002.net/siphash/ + * -------------------------------------------------------------------------- + * HISTORY: + * + * 2017-06-10 (Sebastian Pipping) + * - Clarify license note in the header + * - Address C89 issues: + * - Stop using inline keyword (and let compiler decide) + * - Turn integer suffix ULL to UL + * - Replace _Bool by int + * - Turn macro siphash24 into a function + * - Address invalid conversion (void pointer) by explicit cast + * - Always expose sip24_valid (for self-tests) + * + * 2012-11-04 - Born. (William Ahern) + * -------------------------------------------------------------------------- + * USAGE: + * + * SipHash-2-4 takes as input two 64-bit words as the key, some number of + * message bytes, and outputs a 64-bit word as the message digest. This + * implementation employs two data structures: a struct sipkey for + * representing the key, and a struct siphash for representing the hash + * state. + * + * For converting a 16-byte unsigned char array to a key, use either the + * macro sip_keyof or the routine sip_tokey. The former instantiates a + * compound literal key, while the latter requires a key object as a + * parameter. + * + * unsigned char secret[16]; + * arc4random_buf(secret, sizeof secret); + * struct sipkey *key = sip_keyof(secret); + * + * For hashing a message, use either the convenience macro siphash24 or the + * routines sip24_init, sip24_update, and sip24_final. + * + * struct siphash state; + * void *msg; + * size_t len; + * uint64_t hash; + * + * sip24_init(&state, key); + * sip24_update(&state, msg, len); + * hash = sip24_final(&state); + * + * or + * + * hash = siphash24(msg, len, key); + * + * To convert the 64-bit hash value to a canonical 8-byte little-endian + * binary representation, use either the macro sip_binof or the routine + * sip_tobin. The former instantiates and returns a compound literal array, + * while the latter requires an array object as a parameter. + * -------------------------------------------------------------------------- + * NOTES: + * + * o Neither sip_keyof, sip_binof, nor siphash24 will work with compilers + * lacking compound literal support. Instead, you must use the lower-level + * interfaces which take as parameters the temporary state objects. + * + * o Uppercase macros may evaluate parameters more than once. Lowercase + * macros should not exhibit any such side effects. + * ========================================================================== + */ +#ifndef SIPHASH_H +#define SIPHASH_H + +#include /* size_t */ +#include /* uint64_t uint32_t uint8_t */ + + +#define SIP_ROTL(x, b) (uint64_t)(((x) << (b)) | ( (x) >> (64 - (b)))) + +#define SIP_U32TO8_LE(p, v) \ + (p)[0] = (uint8_t)((v) >> 0); (p)[1] = (uint8_t)((v) >> 8); \ + (p)[2] = (uint8_t)((v) >> 16); (p)[3] = (uint8_t)((v) >> 24); + +#define SIP_U64TO8_LE(p, v) \ + SIP_U32TO8_LE((p) + 0, (uint32_t)((v) >> 0)); \ + SIP_U32TO8_LE((p) + 4, (uint32_t)((v) >> 32)); + +#define SIP_U8TO64_LE(p) \ + (((uint64_t)((p)[0]) << 0) | \ + ((uint64_t)((p)[1]) << 8) | \ + ((uint64_t)((p)[2]) << 16) | \ + ((uint64_t)((p)[3]) << 24) | \ + ((uint64_t)((p)[4]) << 32) | \ + ((uint64_t)((p)[5]) << 40) | \ + ((uint64_t)((p)[6]) << 48) | \ + ((uint64_t)((p)[7]) << 56)) + + +#define SIPHASH_INITIALIZER { 0, 0, 0, 0, { 0 }, 0, 0 } + +struct siphash { + uint64_t v0, v1, v2, v3; + + unsigned char buf[8], *p; + uint64_t c; +}; /* struct siphash */ + + +#define SIP_KEYLEN 16 + +struct sipkey { + uint64_t k[2]; +}; /* struct sipkey */ + +#define sip_keyof(k) sip_tokey(&(struct sipkey){ { 0 } }, (k)) + +static struct sipkey *sip_tokey(struct sipkey *key, const void *src) { + key->k[0] = SIP_U8TO64_LE((const unsigned char *)src); + key->k[1] = SIP_U8TO64_LE((const unsigned char *)src + 8); + return key; +} /* sip_tokey() */ + + +#define sip_binof(v) sip_tobin((unsigned char[8]){ 0 }, (v)) + +static void *sip_tobin(void *dst, uint64_t u64) { + SIP_U64TO8_LE((unsigned char *)dst, u64); + return dst; +} /* sip_tobin() */ + + +static void sip_round(struct siphash *H, const int rounds) { + int i; + + for (i = 0; i < rounds; i++) { + H->v0 += H->v1; + H->v1 = SIP_ROTL(H->v1, 13); + H->v1 ^= H->v0; + H->v0 = SIP_ROTL(H->v0, 32); + + H->v2 += H->v3; + H->v3 = SIP_ROTL(H->v3, 16); + H->v3 ^= H->v2; + + H->v0 += H->v3; + H->v3 = SIP_ROTL(H->v3, 21); + H->v3 ^= H->v0; + + H->v2 += H->v1; + H->v1 = SIP_ROTL(H->v1, 17); + H->v1 ^= H->v2; + H->v2 = SIP_ROTL(H->v2, 32); + } +} /* sip_round() */ + + +static struct siphash *sip24_init(struct siphash *H, const struct sipkey *key) { + H->v0 = 0x736f6d6570736575UL ^ key->k[0]; + H->v1 = 0x646f72616e646f6dUL ^ key->k[1]; + H->v2 = 0x6c7967656e657261UL ^ key->k[0]; + H->v3 = 0x7465646279746573UL ^ key->k[1]; + + H->p = H->buf; + H->c = 0; + + return H; +} /* sip24_init() */ + + +#define sip_endof(a) (&(a)[sizeof (a) / sizeof *(a)]) + +static struct siphash *sip24_update(struct siphash *H, const void *src, size_t len) { + const unsigned char *p = (const unsigned char *)src, *pe = p + len; + uint64_t m; + + do { + while (p < pe && H->p < sip_endof(H->buf)) + *H->p++ = *p++; + + if (H->p < sip_endof(H->buf)) + break; + + m = SIP_U8TO64_LE(H->buf); + H->v3 ^= m; + sip_round(H, 2); + H->v0 ^= m; + + H->p = H->buf; + H->c += 8; + } while (p < pe); + + return H; +} /* sip24_update() */ + + +static uint64_t sip24_final(struct siphash *H) { + char left = H->p - H->buf; + uint64_t b = (H->c + left) << 56; + + switch (left) { + case 7: b |= (uint64_t)H->buf[6] << 48; + case 6: b |= (uint64_t)H->buf[5] << 40; + case 5: b |= (uint64_t)H->buf[4] << 32; + case 4: b |= (uint64_t)H->buf[3] << 24; + case 3: b |= (uint64_t)H->buf[2] << 16; + case 2: b |= (uint64_t)H->buf[1] << 8; + case 1: b |= (uint64_t)H->buf[0] << 0; + case 0: break; + } + + H->v3 ^= b; + sip_round(H, 2); + H->v0 ^= b; + H->v2 ^= 0xff; + sip_round(H, 4); + + return H->v0 ^ H->v1 ^ H->v2 ^ H->v3; +} /* sip24_final() */ + + +static uint64_t siphash24(const void *src, size_t len, const struct sipkey *key) { + struct siphash state = SIPHASH_INITIALIZER; + return sip24_final(sip24_update(sip24_init(&state, key), src, len)); +} /* siphash24() */ + + +/* + * SipHash-2-4 output with + * k = 00 01 02 ... + * and + * in = (empty string) + * in = 00 (1 byte) + * in = 00 01 (2 bytes) + * in = 00 01 02 (3 bytes) + * ... + * in = 00 01 02 ... 3e (63 bytes) + */ +static int sip24_valid(void) { + static const unsigned char vectors[64][8] = { + { 0x31, 0x0e, 0x0e, 0xdd, 0x47, 0xdb, 0x6f, 0x72, }, + { 0xfd, 0x67, 0xdc, 0x93, 0xc5, 0x39, 0xf8, 0x74, }, + { 0x5a, 0x4f, 0xa9, 0xd9, 0x09, 0x80, 0x6c, 0x0d, }, + { 0x2d, 0x7e, 0xfb, 0xd7, 0x96, 0x66, 0x67, 0x85, }, + { 0xb7, 0x87, 0x71, 0x27, 0xe0, 0x94, 0x27, 0xcf, }, + { 0x8d, 0xa6, 0x99, 0xcd, 0x64, 0x55, 0x76, 0x18, }, + { 0xce, 0xe3, 0xfe, 0x58, 0x6e, 0x46, 0xc9, 0xcb, }, + { 0x37, 0xd1, 0x01, 0x8b, 0xf5, 0x00, 0x02, 0xab, }, + { 0x62, 0x24, 0x93, 0x9a, 0x79, 0xf5, 0xf5, 0x93, }, + { 0xb0, 0xe4, 0xa9, 0x0b, 0xdf, 0x82, 0x00, 0x9e, }, + { 0xf3, 0xb9, 0xdd, 0x94, 0xc5, 0xbb, 0x5d, 0x7a, }, + { 0xa7, 0xad, 0x6b, 0x22, 0x46, 0x2f, 0xb3, 0xf4, }, + { 0xfb, 0xe5, 0x0e, 0x86, 0xbc, 0x8f, 0x1e, 0x75, }, + { 0x90, 0x3d, 0x84, 0xc0, 0x27, 0x56, 0xea, 0x14, }, + { 0xee, 0xf2, 0x7a, 0x8e, 0x90, 0xca, 0x23, 0xf7, }, + { 0xe5, 0x45, 0xbe, 0x49, 0x61, 0xca, 0x29, 0xa1, }, + { 0xdb, 0x9b, 0xc2, 0x57, 0x7f, 0xcc, 0x2a, 0x3f, }, + { 0x94, 0x47, 0xbe, 0x2c, 0xf5, 0xe9, 0x9a, 0x69, }, + { 0x9c, 0xd3, 0x8d, 0x96, 0xf0, 0xb3, 0xc1, 0x4b, }, + { 0xbd, 0x61, 0x79, 0xa7, 0x1d, 0xc9, 0x6d, 0xbb, }, + { 0x98, 0xee, 0xa2, 0x1a, 0xf2, 0x5c, 0xd6, 0xbe, }, + { 0xc7, 0x67, 0x3b, 0x2e, 0xb0, 0xcb, 0xf2, 0xd0, }, + { 0x88, 0x3e, 0xa3, 0xe3, 0x95, 0x67, 0x53, 0x93, }, + { 0xc8, 0xce, 0x5c, 0xcd, 0x8c, 0x03, 0x0c, 0xa8, }, + { 0x94, 0xaf, 0x49, 0xf6, 0xc6, 0x50, 0xad, 0xb8, }, + { 0xea, 0xb8, 0x85, 0x8a, 0xde, 0x92, 0xe1, 0xbc, }, + { 0xf3, 0x15, 0xbb, 0x5b, 0xb8, 0x35, 0xd8, 0x17, }, + { 0xad, 0xcf, 0x6b, 0x07, 0x63, 0x61, 0x2e, 0x2f, }, + { 0xa5, 0xc9, 0x1d, 0xa7, 0xac, 0xaa, 0x4d, 0xde, }, + { 0x71, 0x65, 0x95, 0x87, 0x66, 0x50, 0xa2, 0xa6, }, + { 0x28, 0xef, 0x49, 0x5c, 0x53, 0xa3, 0x87, 0xad, }, + { 0x42, 0xc3, 0x41, 0xd8, 0xfa, 0x92, 0xd8, 0x32, }, + { 0xce, 0x7c, 0xf2, 0x72, 0x2f, 0x51, 0x27, 0x71, }, + { 0xe3, 0x78, 0x59, 0xf9, 0x46, 0x23, 0xf3, 0xa7, }, + { 0x38, 0x12, 0x05, 0xbb, 0x1a, 0xb0, 0xe0, 0x12, }, + { 0xae, 0x97, 0xa1, 0x0f, 0xd4, 0x34, 0xe0, 0x15, }, + { 0xb4, 0xa3, 0x15, 0x08, 0xbe, 0xff, 0x4d, 0x31, }, + { 0x81, 0x39, 0x62, 0x29, 0xf0, 0x90, 0x79, 0x02, }, + { 0x4d, 0x0c, 0xf4, 0x9e, 0xe5, 0xd4, 0xdc, 0xca, }, + { 0x5c, 0x73, 0x33, 0x6a, 0x76, 0xd8, 0xbf, 0x9a, }, + { 0xd0, 0xa7, 0x04, 0x53, 0x6b, 0xa9, 0x3e, 0x0e, }, + { 0x92, 0x59, 0x58, 0xfc, 0xd6, 0x42, 0x0c, 0xad, }, + { 0xa9, 0x15, 0xc2, 0x9b, 0xc8, 0x06, 0x73, 0x18, }, + { 0x95, 0x2b, 0x79, 0xf3, 0xbc, 0x0a, 0xa6, 0xd4, }, + { 0xf2, 0x1d, 0xf2, 0xe4, 0x1d, 0x45, 0x35, 0xf9, }, + { 0x87, 0x57, 0x75, 0x19, 0x04, 0x8f, 0x53, 0xa9, }, + { 0x10, 0xa5, 0x6c, 0xf5, 0xdf, 0xcd, 0x9a, 0xdb, }, + { 0xeb, 0x75, 0x09, 0x5c, 0xcd, 0x98, 0x6c, 0xd0, }, + { 0x51, 0xa9, 0xcb, 0x9e, 0xcb, 0xa3, 0x12, 0xe6, }, + { 0x96, 0xaf, 0xad, 0xfc, 0x2c, 0xe6, 0x66, 0xc7, }, + { 0x72, 0xfe, 0x52, 0x97, 0x5a, 0x43, 0x64, 0xee, }, + { 0x5a, 0x16, 0x45, 0xb2, 0x76, 0xd5, 0x92, 0xa1, }, + { 0xb2, 0x74, 0xcb, 0x8e, 0xbf, 0x87, 0x87, 0x0a, }, + { 0x6f, 0x9b, 0xb4, 0x20, 0x3d, 0xe7, 0xb3, 0x81, }, + { 0xea, 0xec, 0xb2, 0xa3, 0x0b, 0x22, 0xa8, 0x7f, }, + { 0x99, 0x24, 0xa4, 0x3c, 0xc1, 0x31, 0x57, 0x24, }, + { 0xbd, 0x83, 0x8d, 0x3a, 0xaf, 0xbf, 0x8d, 0xb7, }, + { 0x0b, 0x1a, 0x2a, 0x32, 0x65, 0xd5, 0x1a, 0xea, }, + { 0x13, 0x50, 0x79, 0xa3, 0x23, 0x1c, 0xe6, 0x60, }, + { 0x93, 0x2b, 0x28, 0x46, 0xe4, 0xd7, 0x06, 0x66, }, + { 0xe1, 0x91, 0x5f, 0x5c, 0xb1, 0xec, 0xa4, 0x6c, }, + { 0xf3, 0x25, 0x96, 0x5c, 0xa1, 0x6d, 0x62, 0x9f, }, + { 0x57, 0x5f, 0xf2, 0x8e, 0x60, 0x38, 0x1b, 0xe5, }, + { 0x72, 0x45, 0x06, 0xeb, 0x4c, 0x32, 0x8a, 0x95, } + }; + unsigned char in[64]; + struct sipkey k; + size_t i; + + sip_tokey(&k, "\000\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017"); + + for (i = 0; i < sizeof in; ++i) { + in[i] = i; + + if (siphash24(in, i, &k) != SIP_U8TO64_LE(vectors[i])) + return 0; + } + + return 1; +} /* sip24_valid() */ + + +#if SIPHASH_MAIN + +#include + +int main(void) { + int ok = sip24_valid(); + + if (ok) + puts("OK"); + else + puts("FAIL"); + + return !ok; +} /* main() */ + +#endif /* SIPHASH_MAIN */ + + +#endif /* SIPHASH_H */ diff --git a/XML/src/xmlparse.cpp b/XML/src/xmlparse.cpp index 47e41b00d..0a01721d5 100644 --- a/XML/src/xmlparse.cpp +++ b/XML/src/xmlparse.cpp @@ -1,48 +1,42 @@ /* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd See the file COPYING for copying permission. + + 77fea421d361dca90041d0040ecf1dca651167fadf2af79e990e35168d70d933 (2.2.1+) */ +#define _GNU_SOURCE /* syscall prototype */ + #include #include /* memset(), memcpy() */ #include #include /* UINT_MAX */ +#include /* fprintf */ +#include /* getenv */ #define EXPAT_POCO 1 #if defined(EXPAT_POCO) -#include "Poco/Process.h" -#include "Poco/Timestamp.h" -#define platform_getpid Poco::Process::id +#include "Poco/RandomStream.h" +#include "Poco/BinaryReader.h" #elif defined(_WIN32) -#define platform_getpid GetCurrentProcessId +#define getpid GetCurrentProcessId #else #include /* gettimeofday() */ #include /* getpid() */ #include /* getpid() */ -#define platform_getpid getpid #endif #define XML_BUILDING_EXPAT 1 #ifdef EXPAT_WIN32 #include "winconfig.h" -#elif defined(MACOS_CLASSIC) -#include "macconfig.h" -#elif defined(__amigaos__) -#include "amigaconfig.h" -#elif defined(__WATCOMC__) -#include "watcomconfig.h" #elif defined(HAVE_EXPAT_CONFIG_H) #include "expat_config.h" #endif /* ndef EXPAT_WIN32 */ #include "ascii.h" #include "Poco/XML/expat.h" -#if defined(_WIN32) || defined(_WIN32_WCE) -#define WIN32_LEAN_AND_MEAN -#include -#undef WIN32_LEAN_AND_MEAN -#endif +#include "siphash.h" #ifdef XML_UNICODE #define XML_ENCODE_MAX XML_UTF16_ENCODE_MAX @@ -121,17 +115,11 @@ typedef struct { 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. +static size_t +keylen(KEY s); -*/ -#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 +static void +copy_salt_to_sipkey(XML_Parser parser, struct sipkey * key); /* 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, @@ -367,6 +355,8 @@ doIgnoreSection(XML_Parser parser, const ENCODING *, const char **startPtr, const char *end, const char **nextPtr, XML_Bool haveMore); #endif /* XML_DTD */ +static void +freeBindings(XML_Parser parser, BINDING *bindings); static enum XML_Error storeAtts(XML_Parser parser, const ENCODING *, const char *s, TAG_NAME *tagNamePtr, BINDING **bindingsPtr); @@ -709,12 +699,85 @@ static const XML_Char implicitContext[] = { ASCII_s, ASCII_p, ASCII_a, ASCII_c, ASCII_e, '\0' }; + +#if defined(HAVE_GETRANDOM) || defined(HAVE_SYSCALL_GETRANDOM) +# include + +# if defined(HAVE_GETRANDOM) +# include /* getrandom */ +# else +# include /* syscall */ +# include /* SYS_getrandom */ +# endif + +/* Obtain entropy on Linux 3.17+ */ +static int +writeRandomBytes_getrandom(void * target, size_t count) { + int success = 0; /* full count bytes written? */ + size_t bytesWrittenTotal = 0; + const unsigned int getrandomFlags = 0; + + do { + void * const currentTarget = (void*)((char*)target + bytesWrittenTotal); + const size_t bytesToWrite = count - bytesWrittenTotal; + + const int bytesWrittenMore = +#if defined(HAVE_GETRANDOM) + getrandom(currentTarget, bytesToWrite, getrandomFlags); +#else + syscall(SYS_getrandom, currentTarget, bytesToWrite, getrandomFlags); +#endif + + if (bytesWrittenMore > 0) { + bytesWrittenTotal += bytesWrittenMore; + if (bytesWrittenTotal >= count) + success = 1; + } + } while (! success && (errno == EINTR || errno == EAGAIN)); + + return success; +} + +#endif /* defined(HAVE_GETRANDOM) || defined(HAVE_SYSCALL_GETRANDOM) */ + + +#ifdef _WIN32 + +typedef BOOLEAN (APIENTRY *RTLGENRANDOM_FUNC)(PVOID, ULONG); + +/* Obtain entropy on Windows XP / Windows Server 2003 and later. + * Hint on RtlGenRandom and the following article from libsodioum. + * + * Michael Howard: Cryptographically Secure Random number on Windows without using CryptoAPI + * https://blogs.msdn.microsoft.com/michael_howard/2005/01/14/cryptographically-secure-random-number-on-windows-without-using-cryptoapi/ + */ +static int +writeRandomBytes_RtlGenRandom(void * target, size_t count) { + int success = 0; /* full count bytes written? */ + const HMODULE advapi32 = LoadLibrary("ADVAPI32.DLL"); + + if (advapi32) { + const RTLGENRANDOM_FUNC RtlGenRandom + = (RTLGENRANDOM_FUNC)GetProcAddress(advapi32, "SystemFunction036"); + if (RtlGenRandom) { + if (RtlGenRandom((PVOID)target, (ULONG)count) == TRUE) { + success = 1; + } + } + FreeLibrary(advapi32); + } + + return success; +} + +#endif /* _WIN32 */ + + +#if !defined(EXPAT_POCO) static unsigned long -platform_gather_time_entropy(void) +gather_time_entropy(void) { -#if defined(EXPAT_POCO) - return static_cast(Poco::Timestamp().utcTime()); -#elif defined(_WIN32) +#ifdef _WIN32 FILETIME ft; GetSystemTimeAsFileTime(&ft); /* never fails */ return ft.dwHighDateTime ^ ft.dwLowDateTime; @@ -729,21 +792,71 @@ platform_gather_time_entropy(void) return tv.tv_usec; #endif } +#endif + +#if defined(HAVE_ARC4RANDOM_BUF) && defined(HAVE_LIBBSD) +# include +#endif + +static unsigned long +ENTROPY_DEBUG(const char * label, unsigned long entropy) { + const char * const EXPAT_ENTROPY_DEBUG = getenv("EXPAT_ENTROPY_DEBUG"); + if (EXPAT_ENTROPY_DEBUG && ! strcmp(EXPAT_ENTROPY_DEBUG, "1")) { + fprintf(stderr, "Entropy: %s --> 0x%0*lx (%lu bytes)\n", + label, + (int)sizeof(entropy) * 2, entropy, + (unsigned long)sizeof(entropy)); + } + return entropy; +} static unsigned long 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 unsigned long entropy = - platform_gather_time_entropy() ^ platform_getpid() ^ (unsigned long)parser; + unsigned long entropy; + (void)parser; +#if defined(EXPAT_POCO) + Poco::RandomInputStream rstr; + Poco::BinaryReader rrdr(rstr); + rrdr >> entropy; + return ENTROPY_DEBUG("RandomInputStream", entropy); +#else +#if defined(HAVE_ARC4RANDOM_BUF) || defined(__CloudABI__) + (void)gather_time_entropy; + arc4random_buf(&entropy, sizeof(entropy)); + return ENTROPY_DEBUG("arc4random_buf", entropy); +#else + /* Try high quality providers first .. */ +#ifdef _WIN32 + if (writeRandomBytes_RtlGenRandom((void *)&entropy, sizeof(entropy))) { + return ENTROPY_DEBUG("RtlGenRandom", entropy); + } +#elif defined(HAVE_GETRANDOM) || defined(HAVE_SYSCALL_GETRANDOM) + if (writeRandomBytes_getrandom((void *)&entropy, sizeof(entropy))) { + return ENTROPY_DEBUG("getrandom", entropy); + } +#endif + /* .. and self-made low quality for backup: */ + + /* Process ID is 0 bits entropy if attacker has local access */ + entropy = gather_time_entropy() ^ getpid(); /* Factors are 2^31-1 and 2^61-1 (Mersenne primes M31 and M61) */ -#ifdef POCO_LONG_IS_64_BIT - return entropy * 2305843009213693951ULL; -#else - return entropy * 2147483647; + if (sizeof(unsigned long) == 4) { + return ENTROPY_DEBUG("fallback(4)", entropy * 2147483647); + } else { + return ENTROPY_DEBUG("fallback(8)", + entropy * (unsigned long)2305843009213693951); + } #endif +#endif +} + +static unsigned long +get_hash_secret_salt(XML_Parser parser) { + if (parser->m_parentParser != NULL) + return get_hash_secret_salt(parser->m_parentParser); + return parser->m_hash_secret_salt; } static XML_Bool /* only valid for root parser */ @@ -974,6 +1087,10 @@ XML_ParserReset(XML_Parser parser, const XML_Char *encodingName) { TAG *tStk; OPEN_INTERNAL_ENTITY *openEntityList; + + if (parser == NULL) + return XML_FALSE; + if (parentParser) return XML_FALSE; /* move tagStack to freeTagList */ @@ -1008,6 +1125,8 @@ XML_ParserReset(XML_Parser parser, const XML_Char *encodingName) enum XML_Status XMLCALL XML_SetEncoding(XML_Parser parser, const XML_Char *encodingName) { + if (parser == NULL) + return XML_STATUS_ERROR; /* 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. @@ -1031,52 +1150,88 @@ XML_ExternalEntityParserCreate(XML_Parser oldParser, { 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; + DTD *oldDtd; + XML_StartElementHandler oldStartElementHandler; + XML_EndElementHandler oldEndElementHandler; + XML_CharacterDataHandler oldCharacterDataHandler; + XML_ProcessingInstructionHandler oldProcessingInstructionHandler; + XML_CommentHandler oldCommentHandler; + XML_StartCdataSectionHandler oldStartCdataSectionHandler; + XML_EndCdataSectionHandler oldEndCdataSectionHandler; + XML_DefaultHandler oldDefaultHandler; + XML_UnparsedEntityDeclHandler oldUnparsedEntityDeclHandler; + XML_NotationDeclHandler oldNotationDeclHandler; + XML_StartNamespaceDeclHandler oldStartNamespaceDeclHandler; + XML_EndNamespaceDeclHandler oldEndNamespaceDeclHandler; + XML_NotStandaloneHandler oldNotStandaloneHandler; + XML_ExternalEntityRefHandler oldExternalEntityRefHandler; + XML_SkippedEntityHandler oldSkippedEntityHandler; + XML_UnknownEncodingHandler oldUnknownEncodingHandler; + XML_ElementDeclHandler oldElementDeclHandler; + XML_AttlistDeclHandler oldAttlistDeclHandler; + XML_EntityDeclHandler oldEntityDeclHandler; + XML_XmlDeclHandler oldXmlDeclHandler; + ELEMENT_TYPE * oldDeclElementType; - void *oldUserData = userData; - void *oldHandlerArg = handlerArg; - XML_Bool oldDefaultExpandInternalEntities = defaultExpandInternalEntities; - XML_Parser oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg; + void *oldUserData; + void *oldHandlerArg; + XML_Bool oldDefaultExpandInternalEntities; + XML_Parser oldExternalEntityRefHandlerArg; #ifdef XML_DTD - enum XML_ParamEntityParsing oldParamEntityParsing = paramEntityParsing; - int oldInEntityValue = prologState.inEntityValue; + enum XML_ParamEntityParsing oldParamEntityParsing; + int oldInEntityValue; #endif - XML_Bool oldns_triplets = ns_triplets; + XML_Bool oldns_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; + unsigned long oldhash_secret_salt; + + /* Validate the oldParser parameter before we pull everything out of it */ + if (oldParser == NULL) + return NULL; + + /* Stash the original parser contents on the stack */ + oldDtd = _dtd; + oldStartElementHandler = startElementHandler; + oldEndElementHandler = endElementHandler; + oldCharacterDataHandler = characterDataHandler; + oldProcessingInstructionHandler = processingInstructionHandler; + oldCommentHandler = commentHandler; + oldStartCdataSectionHandler = startCdataSectionHandler; + oldEndCdataSectionHandler = endCdataSectionHandler; + oldDefaultHandler = defaultHandler; + oldUnparsedEntityDeclHandler = unparsedEntityDeclHandler; + oldNotationDeclHandler = notationDeclHandler; + oldStartNamespaceDeclHandler = startNamespaceDeclHandler; + oldEndNamespaceDeclHandler = endNamespaceDeclHandler; + oldNotStandaloneHandler = notStandaloneHandler; + oldExternalEntityRefHandler = externalEntityRefHandler; + oldSkippedEntityHandler = skippedEntityHandler; + oldUnknownEncodingHandler = unknownEncodingHandler; + oldElementDeclHandler = elementDeclHandler; + oldAttlistDeclHandler = attlistDeclHandler; + oldEntityDeclHandler = entityDeclHandler; + oldXmlDeclHandler = xmlDeclHandler; + oldDeclElementType = declElementType; + + oldUserData = userData; + oldHandlerArg = handlerArg; + oldDefaultExpandInternalEntities = defaultExpandInternalEntities; + oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg; +#ifdef XML_DTD + oldParamEntityParsing = paramEntityParsing; + oldInEntityValue = prologState.inEntityValue; +#endif + 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. + */ + oldhash_secret_salt = hash_secret_salt; #ifdef XML_DTD if (!context) @@ -1242,12 +1397,15 @@ XML_ParserFree(XML_Parser parser) void XMLCALL XML_UseParserAsHandlerArg(XML_Parser parser) { - handlerArg = parser; + if (parser != NULL) + handlerArg = parser; } enum XML_Error XMLCALL XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD) { + if (parser == NULL) + return XML_ERROR_INVALID_ARGUMENT; #ifdef XML_DTD /* block after XML_Parse()/XML_ParseBuffer() has been called */ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) @@ -1262,6 +1420,8 @@ XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD) void XMLCALL XML_SetReturnNSTriplet(XML_Parser parser, int do_nst) { + if (parser == NULL) + return; /* block after XML_Parse()/XML_ParseBuffer() has been called */ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) return; @@ -1271,6 +1431,8 @@ XML_SetReturnNSTriplet(XML_Parser parser, int do_nst) void XMLCALL XML_SetUserData(XML_Parser parser, void *p) { + if (parser == NULL) + return; if (handlerArg == userData) handlerArg = userData = p; else @@ -1280,6 +1442,8 @@ XML_SetUserData(XML_Parser parser, void *p) enum XML_Status XMLCALL XML_SetBase(XML_Parser parser, const XML_Char *p) { + if (parser == NULL) + return XML_STATUS_ERROR; if (p) { p = poolCopyString(&_dtd->pool, p); if (!p) @@ -1294,18 +1458,24 @@ XML_SetBase(XML_Parser parser, const XML_Char *p) const XML_Char * XMLCALL XML_GetBase(XML_Parser parser) { + if (parser == NULL) + return NULL; return curBase; } int XMLCALL XML_GetSpecifiedAttributeCount(XML_Parser parser) { + if (parser == NULL) + return -1; return nSpecifiedAtts; } int XMLCALL XML_GetIdAttributeIndex(XML_Parser parser) { + if (parser == NULL) + return -1; return idAttIndex; } @@ -1313,6 +1483,8 @@ XML_GetIdAttributeIndex(XML_Parser parser) const XML_AttrInfo * XMLCALL XML_GetAttributeInfo(XML_Parser parser) { + if (parser == NULL) + return NULL; return attInfo; } #endif @@ -1322,6 +1494,8 @@ XML_SetElementHandler(XML_Parser parser, XML_StartElementHandler start, XML_EndElementHandler end) { + if (parser == NULL) + return; startElementHandler = start; endElementHandler = end; } @@ -1329,34 +1503,39 @@ XML_SetElementHandler(XML_Parser parser, void XMLCALL XML_SetStartElementHandler(XML_Parser parser, XML_StartElementHandler start) { - startElementHandler = start; + if (parser != NULL) + startElementHandler = start; } void XMLCALL XML_SetEndElementHandler(XML_Parser parser, XML_EndElementHandler end) { - endElementHandler = end; + if (parser != NULL) + endElementHandler = end; } void XMLCALL XML_SetCharacterDataHandler(XML_Parser parser, XML_CharacterDataHandler handler) { - characterDataHandler = handler; + if (parser != NULL) + characterDataHandler = handler; } void XMLCALL XML_SetProcessingInstructionHandler(XML_Parser parser, XML_ProcessingInstructionHandler handler) { - processingInstructionHandler = handler; + if (parser != NULL) + processingInstructionHandler = handler; } void XMLCALL XML_SetCommentHandler(XML_Parser parser, XML_CommentHandler handler) { - commentHandler = handler; + if (parser != NULL) + commentHandler = handler; } void XMLCALL @@ -1364,6 +1543,8 @@ XML_SetCdataSectionHandler(XML_Parser parser, XML_StartCdataSectionHandler start, XML_EndCdataSectionHandler end) { + if (parser == NULL) + return; startCdataSectionHandler = start; endCdataSectionHandler = end; } @@ -1371,19 +1552,23 @@ XML_SetCdataSectionHandler(XML_Parser parser, void XMLCALL XML_SetStartCdataSectionHandler(XML_Parser parser, XML_StartCdataSectionHandler start) { - startCdataSectionHandler = start; + if (parser != NULL) + startCdataSectionHandler = start; } void XMLCALL XML_SetEndCdataSectionHandler(XML_Parser parser, XML_EndCdataSectionHandler end) { - endCdataSectionHandler = end; + if (parser != NULL) + endCdataSectionHandler = end; } void XMLCALL XML_SetDefaultHandler(XML_Parser parser, XML_DefaultHandler handler) { + if (parser == NULL) + return; defaultHandler = handler; defaultExpandInternalEntities = XML_FALSE; } @@ -1392,6 +1577,8 @@ void XMLCALL XML_SetDefaultHandlerExpand(XML_Parser parser, XML_DefaultHandler handler) { + if (parser == NULL) + return; defaultHandler = handler; defaultExpandInternalEntities = XML_TRUE; } @@ -1401,6 +1588,8 @@ XML_SetDoctypeDeclHandler(XML_Parser parser, XML_StartDoctypeDeclHandler start, XML_EndDoctypeDeclHandler end) { + if (parser == NULL) + return; startDoctypeDeclHandler = start; endDoctypeDeclHandler = end; } @@ -1408,27 +1597,31 @@ XML_SetDoctypeDeclHandler(XML_Parser parser, void XMLCALL XML_SetStartDoctypeDeclHandler(XML_Parser parser, XML_StartDoctypeDeclHandler start) { - startDoctypeDeclHandler = start; + if (parser != NULL) + startDoctypeDeclHandler = start; } void XMLCALL XML_SetEndDoctypeDeclHandler(XML_Parser parser, XML_EndDoctypeDeclHandler end) { - endDoctypeDeclHandler = end; + if (parser != NULL) + endDoctypeDeclHandler = end; } void XMLCALL XML_SetUnparsedEntityDeclHandler(XML_Parser parser, XML_UnparsedEntityDeclHandler handler) { - unparsedEntityDeclHandler = handler; + if (parser != NULL) + unparsedEntityDeclHandler = handler; } void XMLCALL XML_SetNotationDeclHandler(XML_Parser parser, XML_NotationDeclHandler handler) { - notationDeclHandler = handler; + if (parser != NULL) + notationDeclHandler = handler; } void XMLCALL @@ -1436,6 +1629,8 @@ XML_SetNamespaceDeclHandler(XML_Parser parser, XML_StartNamespaceDeclHandler start, XML_EndNamespaceDeclHandler end) { + if (parser == NULL) + return; startNamespaceDeclHandler = start; endNamespaceDeclHandler = end; } @@ -1443,32 +1638,38 @@ XML_SetNamespaceDeclHandler(XML_Parser parser, void XMLCALL XML_SetStartNamespaceDeclHandler(XML_Parser parser, XML_StartNamespaceDeclHandler start) { - startNamespaceDeclHandler = start; + if (parser != NULL) + startNamespaceDeclHandler = start; } void XMLCALL XML_SetEndNamespaceDeclHandler(XML_Parser parser, XML_EndNamespaceDeclHandler end) { - endNamespaceDeclHandler = end; + if (parser != NULL) + endNamespaceDeclHandler = end; } void XMLCALL XML_SetNotStandaloneHandler(XML_Parser parser, XML_NotStandaloneHandler handler) { - notStandaloneHandler = handler; + if (parser != NULL) + notStandaloneHandler = handler; } void XMLCALL XML_SetExternalEntityRefHandler(XML_Parser parser, XML_ExternalEntityRefHandler handler) { - externalEntityRefHandler = handler; + if (parser != NULL) + externalEntityRefHandler = handler; } void XMLCALL XML_SetExternalEntityRefHandlerArg(XML_Parser parser, void *arg) { + if (parser == NULL) + return; if (arg) externalEntityRefHandlerArg = (XML_Parser)arg; else @@ -1479,7 +1680,8 @@ void XMLCALL XML_SetSkippedEntityHandler(XML_Parser parser, XML_SkippedEntityHandler handler) { - skippedEntityHandler = handler; + if (parser != NULL) + skippedEntityHandler = handler; } void XMLCALL @@ -1487,6 +1689,8 @@ XML_SetUnknownEncodingHandler(XML_Parser parser, XML_UnknownEncodingHandler handler, void *data) { + if (parser == NULL) + return; unknownEncodingHandler = handler; unknownEncodingHandlerData = data; } @@ -1495,33 +1699,39 @@ void XMLCALL XML_SetElementDeclHandler(XML_Parser parser, XML_ElementDeclHandler eldecl) { - elementDeclHandler = eldecl; + if (parser != NULL) + elementDeclHandler = eldecl; } void XMLCALL XML_SetAttlistDeclHandler(XML_Parser parser, XML_AttlistDeclHandler attdecl) { - attlistDeclHandler = attdecl; + if (parser != NULL) + attlistDeclHandler = attdecl; } void XMLCALL XML_SetEntityDeclHandler(XML_Parser parser, XML_EntityDeclHandler handler) { - entityDeclHandler = handler; + if (parser != NULL) + entityDeclHandler = handler; } void XMLCALL XML_SetXmlDeclHandler(XML_Parser parser, XML_XmlDeclHandler handler) { - xmlDeclHandler = handler; + if (parser != NULL) + xmlDeclHandler = handler; } int XMLCALL XML_SetParamEntityParsing(XML_Parser parser, enum XML_ParamEntityParsing peParsing) { + if (parser == NULL) + return 0; /* block after XML_Parse()/XML_ParseBuffer() has been called */ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) return 0; @@ -1537,6 +1747,10 @@ int XMLCALL XML_SetHashSalt(XML_Parser parser, unsigned long hash_salt) { + if (parser == NULL) + return 0; + if (parser->m_parentParser) + return XML_SetHashSalt(parser->m_parentParser, hash_salt); /* block after XML_Parse()/XML_ParseBuffer() has been called */ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) return 0; @@ -1547,6 +1761,10 @@ XML_SetHashSalt(XML_Parser parser, enum XML_Status XMLCALL XML_Parse(XML_Parser parser, const char *s, int len, int isFinal) { + if ((parser == NULL) || (len < 0) || ((s == NULL) && (len != 0))) { + errorCode = XML_ERROR_INVALID_ARGUMENT; + return XML_STATUS_ERROR; + } switch (ps_parsing) { case XML_SUSPENDED: errorCode = XML_ERROR_SUSPENDED; @@ -1599,6 +1817,13 @@ XML_Parse(XML_Parser parser, const char *s, int len, int isFinal) const char *end; int nLeftOver; enum XML_Status result; + /* Detect overflow (a+b > MAX <==> b > MAX-a) */ + if (len > ((XML_Size)-1) / 2 - parseEndByteIndex) { + errorCode = XML_ERROR_NO_MEMORY; + eventPtr = eventEndPtr = NULL; + processor = errorProcessor; + return XML_STATUS_ERROR; + } parseEndByteIndex += len; positionPtr = s; ps_finalBuffer = (XML_Bool)isFinal; @@ -1631,11 +1856,14 @@ 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; @@ -1643,7 +1871,7 @@ 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); } @@ -1673,6 +1901,8 @@ XML_ParseBuffer(XML_Parser parser, int len, int isFinal) const char *start; enum XML_Status result = XML_STATUS_OK; + if (parser == NULL) + return XML_STATUS_ERROR; switch (ps_parsing) { case XML_SUSPENDED: errorCode = XML_ERROR_SUSPENDED; @@ -1726,6 +1956,8 @@ XML_ParseBuffer(XML_Parser parser, int len, int isFinal) void * XMLCALL XML_GetBuffer(XML_Parser parser, int len) { + if (parser == NULL) + return NULL; if (len < 0) { errorCode = XML_ERROR_NO_MEMORY; return NULL; @@ -1822,6 +2054,8 @@ XML_GetBuffer(XML_Parser parser, int len) enum XML_Status XMLCALL XML_StopParser(XML_Parser parser, XML_Bool resumable) { + if (parser == NULL) + return XML_STATUS_ERROR; switch (ps_parsing) { case XML_SUSPENDED: if (resumable) { @@ -1854,6 +2088,8 @@ XML_ResumeParser(XML_Parser parser) { enum XML_Status result = XML_STATUS_OK; + if (parser == NULL) + return XML_STATUS_ERROR; if (ps_parsing != XML_SUSPENDED) { errorCode = XML_ERROR_NOT_SUSPENDED; return XML_STATUS_ERROR; @@ -1890,6 +2126,8 @@ XML_ResumeParser(XML_Parser parser) void XMLCALL XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status) { + if (parser == NULL) + return; assert(status != NULL); *status = parser->m_parsingStatus; } @@ -1897,12 +2135,16 @@ XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status) enum XML_Error XMLCALL XML_GetErrorCode(XML_Parser parser) { + if (parser == NULL) + return XML_ERROR_INVALID_ARGUMENT; return errorCode; } XML_Index XMLCALL XML_GetCurrentByteIndex(XML_Parser parser) { + if (parser == NULL) + return -1; if (eventPtr) return (XML_Index)(parseEndByteIndex - (parseEndPtr - eventPtr)); return -1; @@ -1911,6 +2153,8 @@ XML_GetCurrentByteIndex(XML_Parser parser) int XMLCALL XML_GetCurrentByteCount(XML_Parser parser) { + if (parser == NULL) + return 0; if (eventEndPtr && eventPtr) return (int)(eventEndPtr - eventPtr); return 0; @@ -1920,11 +2164,19 @@ const char * XMLCALL XML_GetInputContext(XML_Parser parser, int *offset, int *size) { #ifdef XML_CONTEXT_BYTES + if (parser == NULL) + return NULL; if (eventPtr && buffer) { - *offset = (int)(eventPtr - buffer); - *size = (int)(bufferEnd - buffer); + if (offset != NULL) + *offset = (int)(eventPtr - buffer); + if (size != NULL) + *size = (int)(bufferEnd - buffer); return buffer; } +#else + (void)parser; + (void)offset; + (void)size; #endif /* defined XML_CONTEXT_BYTES */ return (char *) 0; } @@ -1932,6 +2184,8 @@ XML_GetInputContext(XML_Parser parser, int *offset, int *size) XML_Size XMLCALL XML_GetCurrentLineNumber(XML_Parser parser) { + if (parser == NULL) + return 0; if (eventPtr && eventPtr >= positionPtr) { XmlUpdatePosition(encoding, positionPtr, eventPtr, &position); positionPtr = eventPtr; @@ -1942,6 +2196,8 @@ XML_GetCurrentLineNumber(XML_Parser parser) XML_Size XMLCALL XML_GetCurrentColumnNumber(XML_Parser parser) { + if (parser == NULL) + return 0; if (eventPtr && eventPtr >= positionPtr) { XmlUpdatePosition(encoding, positionPtr, eventPtr, &position); positionPtr = eventPtr; @@ -1952,30 +2208,38 @@ XML_GetCurrentColumnNumber(XML_Parser parser) void XMLCALL XML_FreeContentModel(XML_Parser parser, XML_Content *model) { - FREE(model); + if (parser != NULL) + FREE(model); } void * XMLCALL XML_MemMalloc(XML_Parser parser, size_t size) { + if (parser == NULL) + return NULL; return MALLOC(size); } void * XMLCALL XML_MemRealloc(XML_Parser parser, void *ptr, size_t size) { + if (parser == NULL) + return NULL; return REALLOC(ptr, size); } void XMLCALL XML_MemFree(XML_Parser parser, void *ptr) { - FREE(ptr); + if (parser != NULL) + FREE(ptr); } void XMLCALL XML_DefaultCurrent(XML_Parser parser) { + if (parser == NULL) + return; if (defaultHandler) { if (openInternalEntities) reportDefault(parser, @@ -2482,7 +2746,7 @@ doContent(XML_Parser parser, &fromPtr, rawNameEnd, (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1); convLen = (int)(toPtr - (XML_Char *)tag->buf); - if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) { + if ((fromPtr >= rawNameEnd) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) { tag->name.strLen = convLen; break; } @@ -2525,8 +2789,10 @@ doContent(XML_Parser parser, return XML_ERROR_NO_MEMORY; poolFinish(&tempPool); result = storeAtts(parser, enc, s, &name, &bindings); - if (result) + if (result != XML_ERROR_NONE) { + freeBindings(parser, bindings); return result; + } poolFinish(&tempPool); if (startElementHandler) { startElementHandler(handlerArg, name.str, (const XML_Char **)atts); @@ -2541,15 +2807,7 @@ doContent(XML_Parser parser, 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; - } + freeBindings(parser, bindings); } if (tagLevel == 0) return epilogProcessor(parser, next, end, nextPtr); @@ -2747,6 +3005,29 @@ doContent(XML_Parser parser, /* not reached */ } +/* This function does not call free() on the allocated memory, merely + * moving it to the parser's freeBindingList where it can be freed or + * reused as appropriate. + */ +static void +freeBindings(XML_Parser parser, BINDING *bindings) +{ + while (bindings) { + BINDING *b = bindings; + + /* startNamespaceDeclHandler will have been called for this + * binding in addBindings(), so call the end handler now. + */ + if (endNamespaceDeclHandler) + endNamespaceDeclHandler(handlerArg, b->prefix->name); + + bindings = bindings->nextTagBinding; + b->nextTagBinding = freeBindingList; + freeBindingList = b; + b->prefix->binding = b->prevPrefixBinding; + } +} + /* Precondition: all arguments must be non-NULL; Purpose: - normalize attributes @@ -2971,7 +3252,13 @@ storeAtts(XML_Parser parser, const ENCODING *enc, if (s[-1] == 2) { /* prefixed */ ATTRIBUTE_ID *id; const BINDING *b; - unsigned long uriHash = hash_secret_salt; + unsigned long uriHash; + struct siphash sip_state; + struct sipkey sip_key; + + copy_salt_to_sipkey(parser, &sip_key); + sip24_init(&sip_state, &sip_key); + ((XML_Char *)s)[-1] = 0; /* clear flag */ id = (ATTRIBUTE_ID *)lookup(parser, &dtd->attributeIds, s, 0); if (!id || !id->prefix) @@ -2980,22 +3267,26 @@ storeAtts(XML_Parser parser, const ENCODING *enc, 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); } + + sip24_update(&sip_state, b->uri, b->uriLen * sizeof(XML_Char)); + while (*s++ != XML_T(ASCII_COLON)) ; + + sip24_update(&sip_state, s, keylen(s) * sizeof(XML_Char)); + 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++); + uriHash = (unsigned long)sip24_final(&sip_state); + { /* Check hash table for duplicate of expanded name (uriName). Derived from code in lookup(parser, HASH_TABLE *table, ...). */ @@ -3709,6 +4000,14 @@ 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 + "internalEventEndPtr = NULL; textStart = (char *)entity->textPtr; textEnd = (char *)(entity->textPtr + entity->textLen); + /* Set a safe default value in case 'next' does not get set */ + next = textStart; #ifdef XML_DTD if (entity->is_param) { @@ -4930,6 +5231,8 @@ internalEntityProcessor(XML_Parser parser, entity = openEntity->entity; textStart = ((char *)entity->textPtr) + entity->processed; textEnd = (char *)(entity->textPtr + entity->textLen); + /* Set a safe default value in case 'next' does not get set */ + next = textStart; #ifdef XML_DTD if (entity->is_param) { @@ -5890,7 +6193,6 @@ dtdCopy(XML_Parser oldParser, DTD *newDtd, const DTD *oldDtd, const XML_Memory_H newE->defaultAtts = (DEFAULT_ATTRIBUTE *) ms->malloc_fcn(oldE->nDefaultAtts * sizeof(DEFAULT_ATTRIBUTE)); if (!newE->defaultAtts) { - ms->free_fcn(newE); return 0; } } @@ -6025,13 +6327,32 @@ keyeq(KEY s1, KEY s2) return XML_FALSE; } +static size_t +keylen(KEY s) +{ + size_t len = 0; + for (; *s; s++, len++); + return len; +} + +static void +copy_salt_to_sipkey(XML_Parser parser, struct sipkey * key) +{ + key->k[0] = 0; + key->k[1] = get_hash_secret_salt(parser); +} + 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; + struct siphash state; + struct sipkey key; + (void)sip_tobin; + (void)sip24_valid; + copy_salt_to_sipkey(parser, &key); + sip24_init(&state, &key); + sip24_update(&state, s, keylen(s) * sizeof(XML_Char)); + return (unsigned long)sip24_final(&state); } static NAMED * @@ -6274,6 +6595,35 @@ poolStoreString(STRING_POOL *pool, const ENCODING *enc, return pool->start; } +static size_t +poolBytesToAllocateFor(int blockSize) +{ + /* Unprotected math would be: + ** return offsetof(BLOCK, s) + blockSize * sizeof(XML_Char); + ** + ** Detect overflow, avoiding _signed_ overflow undefined behavior + ** For a + b * c we check b * c in isolation first, so that addition of a + ** on top has no chance of making us accept a small non-negative number + */ + const size_t stretch = sizeof(XML_Char); /* can be 4 bytes */ + + if (blockSize <= 0) + return 0; + + if (blockSize > (int)(INT_MAX / stretch)) + return 0; + + { + const int stretchedBlockSize = blockSize * (int)stretch; + const int bytesToAllocate = (int)( + offsetof(BLOCK, s) + (unsigned)stretchedBlockSize); + if (bytesToAllocate < 0) + return 0; + + return (size_t)bytesToAllocate; + } +} + static XML_Bool FASTCALL poolGrow(STRING_POOL *pool) { @@ -6303,14 +6653,17 @@ poolGrow(STRING_POOL *pool) if (pool->blocks && pool->start == pool->blocks->s) { BLOCK *temp; int blockSize = (int)((unsigned)(pool->end - pool->start)*2U); + size_t bytesToAllocate; if (blockSize < 0) return XML_FALSE; + bytesToAllocate = poolBytesToAllocateFor(blockSize); + if (bytesToAllocate == 0) + return XML_FALSE; + temp = (BLOCK *) - pool->mem->realloc_fcn(pool->blocks, - (offsetof(BLOCK, s) - + blockSize * sizeof(XML_Char))); + pool->mem->realloc_fcn(pool->blocks, (unsigned)bytesToAllocate); if (temp == NULL) return XML_FALSE; pool->blocks = temp; @@ -6322,16 +6675,26 @@ poolGrow(STRING_POOL *pool) else { BLOCK *tem; int blockSize = (int)(pool->end - pool->start); + size_t bytesToAllocate; if (blockSize < 0) return XML_FALSE; if (blockSize < INIT_BLOCK_SIZE) blockSize = INIT_BLOCK_SIZE; - else + else { + /* Detect overflow, avoiding _signed_ overflow undefined behavior */ + if ((int)((unsigned)blockSize * 2U) < 0) { + return XML_FALSE; + } blockSize *= 2; - tem = (BLOCK *)pool->mem->malloc_fcn(offsetof(BLOCK, s) - + blockSize * sizeof(XML_Char)); + } + + bytesToAllocate = poolBytesToAllocateFor(blockSize); + if (bytesToAllocate == 0) + return XML_FALSE; + + tem = (BLOCK *)pool->mem->malloc_fcn(bytesToAllocate); if (!tem) return XML_FALSE; tem->size = blockSize; diff --git a/XML/src/xmlrole.c b/XML/src/xmlrole.c index 0d8566823..0b15a87ae 100644 --- a/XML/src/xmlrole.c +++ b/XML/src/xmlrole.c @@ -6,12 +6,6 @@ #ifdef EXPAT_WIN32 #include "winconfig.h" -#elif defined(MACOS_CLASSIC) -#include "macconfig.h" -#elif defined(__amigaos__) -#include "amigaconfig.h" -#elif defined(__WATCOMC__) -#include "watcomconfig.h" #else #ifdef HAVE_EXPAT_CONFIG_H #include "expat_config.h" diff --git a/XML/src/xmltok.c b/XML/src/xmltok.c index 88aca3d7e..abfff6983 100644 --- a/XML/src/xmltok.c +++ b/XML/src/xmltok.c @@ -6,12 +6,6 @@ #ifdef EXPAT_WIN32 #include "winconfig.h" -#elif defined(MACOS_CLASSIC) -#include "macconfig.h" -#elif defined(__amigaos__) -#include "amigaconfig.h" -#elif defined(__WATCOMC__) -#include "watcomconfig.h" #else #ifdef HAVE_EXPAT_CONFIG_H #include "expat_config.h" @@ -369,24 +363,24 @@ utf8_toUtf8(const ENCODING *UNUSED_P(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; - if (fromLim - *fromP > toLim - *toP) { - /* Avoid copying partial characters. */ - res = XML_CONVERT_OUTPUT_EXHAUSTED; - fromLim = *fromP + (toLim - *toP); - align_limit_to_full_utf8_characters(*fromP, &fromLim); - } + const char *fromLimInitial = fromLim; + + /* Avoid copying partial characters. */ + align_limit_to_full_utf8_characters(*fromP, &fromLim); + for (to = *toP, from = *fromP; (from < fromLim) && (to < toLim); from++, to++) *to = *from; *fromP = from; *toP = to; - if ((to == toLim) && (from < fromLim)) + if (fromLim < fromLimInitial) + return XML_CONVERT_INPUT_INCOMPLETE; + else if ((to == toLim) && (from < fromLim)) return XML_CONVERT_OUTPUT_EXHAUSTED; else - return res; + return XML_CONVERT_COMPLETED; } static enum XML_Convert_Result PTRCALL @@ -402,7 +396,7 @@ utf8_toUtf16(const ENCODING *enc, case BT_LEAD2: if (fromLim - from < 2) { res = XML_CONVERT_INPUT_INCOMPLETE; - break; + goto after; } *to++ = (unsigned short)(((from[0] & 0x1f) << 6) | (from[1] & 0x3f)); from += 2; @@ -410,7 +404,7 @@ utf8_toUtf16(const ENCODING *enc, case BT_LEAD3: if (fromLim - from < 3) { res = XML_CONVERT_INPUT_INCOMPLETE; - break; + goto after; } *to++ = (unsigned short)(((from[0] & 0xf) << 12) | ((from[1] & 0x3f) << 6) | (from[2] & 0x3f)); @@ -441,6 +435,8 @@ utf8_toUtf16(const ENCODING *enc, break; } } + if (from < fromLim) + res = XML_CONVERT_OUTPUT_EXHAUSTED; after: *fromP = from; *toP = to; From d18e400a88bdc4e5adcb00764c6a00accb78a904 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?G=C3=BCnter=20Obiltschnig?= Date: Wed, 21 Jun 2017 15:42:48 +0200 Subject: [PATCH 2/3] fix for Windows --- XML/src/xmlparse.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/XML/src/xmlparse.cpp b/XML/src/xmlparse.cpp index 0a01721d5..0c7d09f55 100644 --- a/XML/src/xmlparse.cpp +++ b/XML/src/xmlparse.cpp @@ -700,6 +700,7 @@ static const XML_Char implicitContext[] = { }; +#if !defined(EXPAT_POCO) #if defined(HAVE_GETRANDOM) || defined(HAVE_SYSCALL_GETRANDOM) # include @@ -773,7 +774,6 @@ writeRandomBytes_RtlGenRandom(void * target, size_t count) { #endif /* _WIN32 */ -#if !defined(EXPAT_POCO) static unsigned long gather_time_entropy(void) { From 2867d4c34b8edc96442a6e7944300f592f1d0546 Mon Sep 17 00:00:00 2001 From: Guenter Obiltschnig Date: Wed, 21 Jun 2017 15:48:03 +0200 Subject: [PATCH 3/3] bump version number --- DLLVersion.rc | 4 ++-- Foundation/include/Poco/Version.h | 2 +- VERSION | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/DLLVersion.rc b/DLLVersion.rc index 94d10dbe3..52a61638b 100644 --- a/DLLVersion.rc +++ b/DLLVersion.rc @@ -4,8 +4,8 @@ #include "winres.h" -#define POCO_VERSION 1,7,8,2 -#define POCO_VERSION_STR "1.7.8p2" +#define POCO_VERSION 1,7,8,3 +#define POCO_VERSION_STR "1.7.8p3" VS_VERSION_INFO VERSIONINFO FILEVERSION POCO_VERSION diff --git a/Foundation/include/Poco/Version.h b/Foundation/include/Poco/Version.h index ebf75f821..0aaf2720a 100644 --- a/Foundation/include/Poco/Version.h +++ b/Foundation/include/Poco/Version.h @@ -37,7 +37,7 @@ // Ax: alpha releases // Bx: beta releases // -#define POCO_VERSION 0x01070802 +#define POCO_VERSION 0x01070803 #endif // Foundation_Version_INCLUDED diff --git a/VERSION b/VERSION index 0f55946b7..b5cd65cfa 100644 --- a/VERSION +++ b/VERSION @@ -1 +1 @@ -1.7.8p2 +1.7.8p3