/* * parser.c : an XML 1.0 parser, namespaces and validity support are mostly * implemented on top of the SAX interfaces * * References: * The XML specification: * http://www.w3.org/TR/REC-xml * Original 1.0 version: * http://www.w3.org/TR/1998/REC-xml-19980210 * XML second edition working draft * http://www.w3.org/TR/2000/WD-xml-2e-20000814 * * Okay this is a big file, the parser core is around 7000 lines, then it * is followed by the progressive parser top routines, then the various * high level APIs to call the parser and a few miscellaneous functions. * A number of helper functions and deprecated ones have been moved to * parserInternals.c to reduce this file size. * As much as possible the functions are associated with their relative * production in the XML specification. A few productions defining the * different ranges of character are actually implanted either in * parserInternals.h or parserInternals.c * The DOM tree build is realized from the default SAX callbacks in * the module SAX.c. * The routines doing the validation checks are in valid.c and called either * from the SAX callbacks or as standalone functions using a preparsed * document. * * See Copyright for the status of this software. * * daniel@veillard.com */ /* To avoid EBCDIC trouble when parsing on zOS */ #if defined(__MVS__) #pragma convert("ISO8859-1") #endif #define IN_LIBXML #include "libxml.h" #if defined(_WIN32) #define XML_DIR_SEP '\\' #else #define XML_DIR_SEP '/' #endif #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #ifdef LIBXML_CATALOG_ENABLED #include #endif #ifdef LIBXML_SCHEMAS_ENABLED #include #include #endif #if defined(LIBXML_XPATH_ENABLED) || defined(LIBXML_SCHEMAS_ENABLED) #include #endif #include "private/buf.h" #include "private/dict.h" #include "private/enc.h" #include "private/entities.h" #include "private/error.h" #include "private/globals.h" #include "private/html.h" #include "private/io.h" #include "private/memory.h" #include "private/parser.h" #include "private/threads.h" #include "private/xpath.h" struct _xmlStartTag { const xmlChar *prefix; const xmlChar *URI; int line; int nsNr; }; static xmlParserCtxtPtr xmlCreateEntityParserCtxtInternal(xmlSAXHandlerPtr sax, void *userData, const xmlChar *URL, const xmlChar *ID, const xmlChar *base, xmlParserCtxtPtr pctx); static int xmlParseElementStart(xmlParserCtxtPtr ctxt); static void xmlParseElementEnd(xmlParserCtxtPtr ctxt); /************************************************************************ * * * Arbitrary limits set in the parser. See XML_PARSE_HUGE * * * ************************************************************************/ #define XML_PARSER_BIG_ENTITY 1000 #define XML_PARSER_LOT_ENTITY 5000 /* * Constants for protection against abusive entity expansion * ("billion laughs"). */ /* * XML_PARSER_NON_LINEAR is roughly the maximum allowed amplification factor * of serialized output after entity expansion. */ #define XML_PARSER_NON_LINEAR 5 /* * A certain amount is always allowed. */ #define XML_PARSER_ALLOWED_EXPANSION 1000000 /* * Fixed cost for each entity reference. This crudely models processing time * as well to protect, for example, against exponential expansion of empty * or very short entities. */ #define XML_ENT_FIXED_COST 20 /** * xmlParserMaxDepth: * * arbitrary depth limit for the XML documents that we allow to * process. This is not a limitation of the parser but a safety * boundary feature. It can be disabled with the XML_PARSE_HUGE * parser option. */ unsigned int xmlParserMaxDepth = 256; #define SAX2 1 #define XML_PARSER_BIG_BUFFER_SIZE 300 #define XML_PARSER_BUFFER_SIZE 100 #define SAX_COMPAT_MODE BAD_CAST "SAX compatibility mode document" /** * XML_PARSER_CHUNK_SIZE * * When calling GROW that's the minimal amount of data * the parser expected to have received. It is not a hard * limit but an optimization when reading strings like Names * It is not strictly needed as long as inputs available characters * are followed by 0, which should be provided by the I/O level */ #define XML_PARSER_CHUNK_SIZE 100 /* * List of XML prefixed PI allowed by W3C specs */ static const char* const xmlW3CPIs[] = { "xml-stylesheet", "xml-model", NULL }; /* DEPR void xmlParserHandleReference(xmlParserCtxtPtr ctxt); */ static xmlEntityPtr xmlParseStringPEReference(xmlParserCtxtPtr ctxt, const xmlChar **str); static xmlParserErrors xmlParseExternalEntityPrivate(xmlDocPtr doc, xmlParserCtxtPtr oldctxt, xmlSAXHandlerPtr sax, void *user_data, int depth, const xmlChar *URL, const xmlChar *ID, xmlNodePtr *list); static int xmlCtxtUseOptionsInternal(xmlParserCtxtPtr ctxt, int options, const char *encoding); #ifdef LIBXML_LEGACY_ENABLED static void xmlAddEntityReference(xmlEntityPtr ent, xmlNodePtr firstNode, xmlNodePtr lastNode); #endif /* LIBXML_LEGACY_ENABLED */ static xmlParserErrors xmlParseBalancedChunkMemoryInternal(xmlParserCtxtPtr oldctxt, const xmlChar *string, void *user_data, xmlNodePtr *lst); static int xmlLoadEntityContent(xmlParserCtxtPtr ctxt, xmlEntityPtr entity); /************************************************************************ * * * Some factorized error routines * * * ************************************************************************/ /** * xmlErrAttributeDup: * @ctxt: an XML parser context * @prefix: the attribute prefix * @localname: the attribute localname * * Handle a redefinition of attribute error */ static void xmlErrAttributeDup(xmlParserCtxtPtr ctxt, const xmlChar * prefix, const xmlChar * localname) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = XML_ERR_ATTRIBUTE_REDEFINED; if (prefix == NULL) __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, XML_ERR_ATTRIBUTE_REDEFINED, XML_ERR_FATAL, NULL, 0, (const char *) localname, NULL, NULL, 0, 0, "Attribute %s redefined\n", localname); else __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, XML_ERR_ATTRIBUTE_REDEFINED, XML_ERR_FATAL, NULL, 0, (const char *) prefix, (const char *) localname, NULL, 0, 0, "Attribute %s:%s redefined\n", prefix, localname); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlFatalErrMsg: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void LIBXML_ATTR_FORMAT(3,0) xmlFatalErrMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, NULL, NULL, NULL, 0, 0, "%s", msg); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlWarningMsg: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @str1: extra data * @str2: extra data * * Handle a warning. */ static void LIBXML_ATTR_FORMAT(3,0) xmlWarningMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar *str1, const xmlChar *str2) { xmlStructuredErrorFunc schannel = NULL; if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if ((ctxt != NULL) && (ctxt->sax != NULL) && (ctxt->sax->initialized == XML_SAX2_MAGIC)) schannel = ctxt->sax->serror; if (ctxt != NULL) { __xmlRaiseError(schannel, (ctxt->sax) ? ctxt->sax->warning : NULL, ctxt->userData, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_WARNING, NULL, 0, (const char *) str1, (const char *) str2, NULL, 0, 0, msg, (const char *) str1, (const char *) str2); } else { __xmlRaiseError(schannel, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_WARNING, NULL, 0, (const char *) str1, (const char *) str2, NULL, 0, 0, msg, (const char *) str1, (const char *) str2); } } /** * xmlValidityError: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @str1: extra data * * Handle a validity error. */ static void LIBXML_ATTR_FORMAT(3,0) xmlValidityError(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar *str1, const xmlChar *str2) { xmlStructuredErrorFunc schannel = NULL; if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) { ctxt->errNo = error; if ((ctxt->sax != NULL) && (ctxt->sax->initialized == XML_SAX2_MAGIC)) schannel = ctxt->sax->serror; } if (ctxt != NULL) { __xmlRaiseError(schannel, ctxt->vctxt.error, ctxt->vctxt.userData, ctxt, NULL, XML_FROM_DTD, error, XML_ERR_ERROR, NULL, 0, (const char *) str1, (const char *) str2, NULL, 0, 0, msg, (const char *) str1, (const char *) str2); ctxt->valid = 0; } else { __xmlRaiseError(schannel, NULL, NULL, ctxt, NULL, XML_FROM_DTD, error, XML_ERR_ERROR, NULL, 0, (const char *) str1, (const char *) str2, NULL, 0, 0, msg, (const char *) str1, (const char *) str2); } } /** * xmlFatalErrMsgInt: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @val: an integer value * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void LIBXML_ATTR_FORMAT(3,0) xmlFatalErrMsgInt(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, int val) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, NULL, NULL, NULL, val, 0, msg, val); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlFatalErrMsgStrIntStr: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @str1: an string info * @val: an integer value * @str2: an string info * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void LIBXML_ATTR_FORMAT(3,0) xmlFatalErrMsgStrIntStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar *str1, int val, const xmlChar *str2) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, (const char *) str1, (const char *) str2, NULL, val, 0, msg, str1, val, str2); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlFatalErrMsgStr: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @val: a string value * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void LIBXML_ATTR_FORMAT(3,0) xmlFatalErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar * val) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, (const char *) val, NULL, NULL, 0, 0, msg, val); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlErrMsgStr: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @val: a string value * * Handle a non fatal parser error */ static void LIBXML_ATTR_FORMAT(3,0) xmlErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar * val) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_ERROR, NULL, 0, (const char *) val, NULL, NULL, 0, 0, msg, val); } /** * xmlNsErr: * @ctxt: an XML parser context * @error: the error number * @msg: the message * @info1: extra information string * @info2: extra information string * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void LIBXML_ATTR_FORMAT(3,0) xmlNsErr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar * info1, const xmlChar * info2, const xmlChar * info3) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_NAMESPACE, error, XML_ERR_ERROR, NULL, 0, (const char *) info1, (const char *) info2, (const char *) info3, 0, 0, msg, info1, info2, info3); if (ctxt != NULL) ctxt->nsWellFormed = 0; } /** * xmlNsWarn * @ctxt: an XML parser context * @error: the error number * @msg: the message * @info1: extra information string * @info2: extra information string * * Handle a namespace warning error */ static void LIBXML_ATTR_FORMAT(3,0) xmlNsWarn(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar * info1, const xmlChar * info2, const xmlChar * info3) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_NAMESPACE, error, XML_ERR_WARNING, NULL, 0, (const char *) info1, (const char *) info2, (const char *) info3, 0, 0, msg, info1, info2, info3); } static void xmlSaturatedAdd(unsigned long *dst, unsigned long val) { if (val > ULONG_MAX - *dst) *dst = ULONG_MAX; else *dst += val; } static void xmlSaturatedAddSizeT(unsigned long *dst, unsigned long val) { if (val > ULONG_MAX - *dst) *dst = ULONG_MAX; else *dst += val; } /** * xmlParserEntityCheck: * @ctxt: parser context * @extra: sum of unexpanded entity sizes * * Check for non-linear entity expansion behaviour. * * In some cases like xmlStringDecodeEntities, this function is called * for each, possibly nested entity and its unexpanded content length. * * In other cases like xmlParseReference, it's only called for each * top-level entity with its unexpanded content length plus the sum of * the unexpanded content lengths (plus fixed cost) of all nested * entities. * * Summing the unexpanded lengths also adds the length of the reference. * This is by design. Taking the length of the entity name into account * discourages attacks that try to waste CPU time with abusively long * entity names. See test/recurse/lol6.xml for example. Each call also * adds some fixed cost XML_ENT_FIXED_COST to discourage attacks with * short entities. * * Returns 1 on error, 0 on success. */ static int xmlParserEntityCheck(xmlParserCtxtPtr ctxt, unsigned long extra) { unsigned long consumed; xmlParserInputPtr input = ctxt->input; xmlEntityPtr entity = input->entity; /* * Compute total consumed bytes so far, including input streams of * external entities. */ consumed = input->parentConsumed; if ((entity == NULL) || ((entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) && ((entity->flags & XML_ENT_PARSED) == 0))) { xmlSaturatedAdd(&consumed, input->consumed); xmlSaturatedAddSizeT(&consumed, input->cur - input->base); } xmlSaturatedAdd(&consumed, ctxt->sizeentities); /* * Add extra cost and some fixed cost. */ xmlSaturatedAdd(&ctxt->sizeentcopy, extra); xmlSaturatedAdd(&ctxt->sizeentcopy, XML_ENT_FIXED_COST); /* * It's important to always use saturation arithmetic when tracking * entity sizes to make the size checks reliable. If "sizeentcopy" * overflows, we have to abort. */ if ((ctxt->sizeentcopy > XML_PARSER_ALLOWED_EXPANSION) && ((ctxt->sizeentcopy >= ULONG_MAX) || (ctxt->sizeentcopy / XML_PARSER_NON_LINEAR > consumed))) { xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_LOOP, "Maximum entity amplification factor exceeded"); xmlHaltParser(ctxt); return(1); } return(0); } /************************************************************************ * * * Library wide options * * * ************************************************************************/ /** * xmlHasFeature: * @feature: the feature to be examined * * Examines if the library has been compiled with a given feature. * * Returns a non-zero value if the feature exist, otherwise zero. * Returns zero (0) if the feature does not exist or an unknown * unknown feature is requested, non-zero otherwise. */ int xmlHasFeature(xmlFeature feature) { switch (feature) { case XML_WITH_THREAD: #ifdef LIBXML_THREAD_ENABLED return(1); #else return(0); #endif case XML_WITH_TREE: #ifdef LIBXML_TREE_ENABLED return(1); #else return(0); #endif case XML_WITH_OUTPUT: #ifdef LIBXML_OUTPUT_ENABLED return(1); #else return(0); #endif case XML_WITH_PUSH: #ifdef LIBXML_PUSH_ENABLED return(1); #else return(0); #endif case XML_WITH_READER: #ifdef LIBXML_READER_ENABLED return(1); #else return(0); #endif case XML_WITH_PATTERN: #ifdef LIBXML_PATTERN_ENABLED return(1); #else return(0); #endif case XML_WITH_WRITER: #ifdef LIBXML_WRITER_ENABLED return(1); #else return(0); #endif case XML_WITH_SAX1: #ifdef LIBXML_SAX1_ENABLED return(1); #else return(0); #endif case XML_WITH_FTP: #ifdef LIBXML_FTP_ENABLED return(1); #else return(0); #endif case XML_WITH_HTTP: #ifdef LIBXML_HTTP_ENABLED return(1); #else return(0); #endif case XML_WITH_VALID: #ifdef LIBXML_VALID_ENABLED return(1); #else return(0); #endif case XML_WITH_HTML: #ifdef LIBXML_HTML_ENABLED return(1); #else return(0); #endif case XML_WITH_LEGACY: #ifdef LIBXML_LEGACY_ENABLED return(1); #else return(0); #endif case XML_WITH_C14N: #ifdef LIBXML_C14N_ENABLED return(1); #else return(0); #endif case XML_WITH_CATALOG: #ifdef LIBXML_CATALOG_ENABLED return(1); #else return(0); #endif case XML_WITH_XPATH: #ifdef LIBXML_XPATH_ENABLED return(1); #else return(0); #endif case XML_WITH_XPTR: #ifdef LIBXML_XPTR_ENABLED return(1); #else return(0); #endif case XML_WITH_XINCLUDE: #ifdef LIBXML_XINCLUDE_ENABLED return(1); #else return(0); #endif case XML_WITH_ICONV: #ifdef LIBXML_ICONV_ENABLED return(1); #else return(0); #endif case XML_WITH_ISO8859X: #ifdef LIBXML_ISO8859X_ENABLED return(1); #else return(0); #endif case XML_WITH_UNICODE: #ifdef LIBXML_UNICODE_ENABLED return(1); #else return(0); #endif case XML_WITH_REGEXP: #ifdef LIBXML_REGEXP_ENABLED return(1); #else return(0); #endif case XML_WITH_AUTOMATA: #ifdef LIBXML_AUTOMATA_ENABLED return(1); #else return(0); #endif case XML_WITH_EXPR: #ifdef LIBXML_EXPR_ENABLED return(1); #else return(0); #endif case XML_WITH_SCHEMAS: #ifdef LIBXML_SCHEMAS_ENABLED return(1); #else return(0); #endif case XML_WITH_SCHEMATRON: #ifdef LIBXML_SCHEMATRON_ENABLED return(1); #else return(0); #endif case XML_WITH_MODULES: #ifdef LIBXML_MODULES_ENABLED return(1); #else return(0); #endif case XML_WITH_DEBUG: #ifdef LIBXML_DEBUG_ENABLED return(1); #else return(0); #endif case XML_WITH_DEBUG_MEM: #ifdef DEBUG_MEMORY_LOCATION return(1); #else return(0); #endif case XML_WITH_DEBUG_RUN: return(0); case XML_WITH_ZLIB: #ifdef LIBXML_ZLIB_ENABLED return(1); #else return(0); #endif case XML_WITH_LZMA: #ifdef LIBXML_LZMA_ENABLED return(1); #else return(0); #endif case XML_WITH_ICU: #ifdef LIBXML_ICU_ENABLED return(1); #else return(0); #endif default: break; } return(0); } /************************************************************************ * * * SAX2 defaulted attributes handling * * * ************************************************************************/ /** * xmlDetectSAX2: * @ctxt: an XML parser context * * Do the SAX2 detection and specific initialization */ static void xmlDetectSAX2(xmlParserCtxtPtr ctxt) { xmlSAXHandlerPtr sax; /* Avoid unused variable warning if features are disabled. */ (void) sax; if (ctxt == NULL) return; sax = ctxt->sax; #ifdef LIBXML_SAX1_ENABLED if ((sax) && (sax->initialized == XML_SAX2_MAGIC)) ctxt->sax2 = 1; #else ctxt->sax2 = 1; #endif /* LIBXML_SAX1_ENABLED */ ctxt->str_xml = xmlDictLookup(ctxt->dict, BAD_CAST "xml", 3); ctxt->str_xmlns = xmlDictLookup(ctxt->dict, BAD_CAST "xmlns", 5); ctxt->str_xml_ns = xmlDictLookup(ctxt->dict, XML_XML_NAMESPACE, 36); if ((ctxt->str_xml==NULL) || (ctxt->str_xmlns==NULL) || (ctxt->str_xml_ns == NULL)) { xmlErrMemory(ctxt, NULL); } } typedef struct _xmlDefAttrs xmlDefAttrs; typedef xmlDefAttrs *xmlDefAttrsPtr; struct _xmlDefAttrs { int nbAttrs; /* number of defaulted attributes on that element */ int maxAttrs; /* the size of the array */ #if __STDC_VERSION__ >= 199901L /* Using a C99 flexible array member avoids UBSan errors. */ const xmlChar *values[]; /* array of localname/prefix/values/external */ #else const xmlChar *values[5]; #endif }; /** * xmlAttrNormalizeSpace: * @src: the source string * @dst: the target string * * Normalize the space in non CDATA attribute values: * If the attribute type is not CDATA, then the XML processor MUST further * process the normalized attribute value by discarding any leading and * trailing space (#x20) characters, and by replacing sequences of space * (#x20) characters by a single space (#x20) character. * Note that the size of dst need to be at least src, and if one doesn't need * to preserve dst (and it doesn't come from a dictionary or read-only) then * passing src as dst is just fine. * * Returns a pointer to the normalized value (dst) or NULL if no conversion * is needed. */ static xmlChar * xmlAttrNormalizeSpace(const xmlChar *src, xmlChar *dst) { if ((src == NULL) || (dst == NULL)) return(NULL); while (*src == 0x20) src++; while (*src != 0) { if (*src == 0x20) { while (*src == 0x20) src++; if (*src != 0) *dst++ = 0x20; } else { *dst++ = *src++; } } *dst = 0; if (dst == src) return(NULL); return(dst); } /** * xmlAttrNormalizeSpace2: * @src: the source string * * Normalize the space in non CDATA attribute values, a slightly more complex * front end to avoid allocation problems when running on attribute values * coming from the input. * * Returns a pointer to the normalized value (dst) or NULL if no conversion * is needed. */ static const xmlChar * xmlAttrNormalizeSpace2(xmlParserCtxtPtr ctxt, xmlChar *src, int *len) { int i; int remove_head = 0; int need_realloc = 0; const xmlChar *cur; if ((ctxt == NULL) || (src == NULL) || (len == NULL)) return(NULL); i = *len; if (i <= 0) return(NULL); cur = src; while (*cur == 0x20) { cur++; remove_head++; } while (*cur != 0) { if (*cur == 0x20) { cur++; if ((*cur == 0x20) || (*cur == 0)) { need_realloc = 1; break; } } else cur++; } if (need_realloc) { xmlChar *ret; ret = xmlStrndup(src + remove_head, i - remove_head + 1); if (ret == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } xmlAttrNormalizeSpace(ret, ret); *len = strlen((const char *)ret); return(ret); } else if (remove_head) { *len -= remove_head; memmove(src, src + remove_head, 1 + *len); return(src); } return(NULL); } /** * xmlAddDefAttrs: * @ctxt: an XML parser context * @fullname: the element fullname * @fullattr: the attribute fullname * @value: the attribute value * * Add a defaulted attribute for an element */ static void xmlAddDefAttrs(xmlParserCtxtPtr ctxt, const xmlChar *fullname, const xmlChar *fullattr, const xmlChar *value) { xmlDefAttrsPtr defaults; int len; const xmlChar *name; const xmlChar *prefix; /* * Allows to detect attribute redefinitions */ if (ctxt->attsSpecial != NULL) { if (xmlHashLookup2(ctxt->attsSpecial, fullname, fullattr) != NULL) return; } if (ctxt->attsDefault == NULL) { ctxt->attsDefault = xmlHashCreateDict(10, ctxt->dict); if (ctxt->attsDefault == NULL) goto mem_error; } /* * split the element name into prefix:localname , the string found * are within the DTD and then not associated to namespace names. */ name = xmlSplitQName3(fullname, &len); if (name == NULL) { name = xmlDictLookup(ctxt->dict, fullname, -1); prefix = NULL; } else { name = xmlDictLookup(ctxt->dict, name, -1); prefix = xmlDictLookup(ctxt->dict, fullname, len); } /* * make sure there is some storage */ defaults = xmlHashLookup2(ctxt->attsDefault, name, prefix); if (defaults == NULL) { defaults = (xmlDefAttrsPtr) xmlMalloc(sizeof(xmlDefAttrs) + (4 * 5) * sizeof(const xmlChar *)); if (defaults == NULL) goto mem_error; defaults->nbAttrs = 0; defaults->maxAttrs = 4; if (xmlHashUpdateEntry2(ctxt->attsDefault, name, prefix, defaults, NULL) < 0) { xmlFree(defaults); goto mem_error; } } else if (defaults->nbAttrs >= defaults->maxAttrs) { xmlDefAttrsPtr temp; temp = (xmlDefAttrsPtr) xmlRealloc(defaults, sizeof(xmlDefAttrs) + (2 * defaults->maxAttrs * 5) * sizeof(const xmlChar *)); if (temp == NULL) goto mem_error; defaults = temp; defaults->maxAttrs *= 2; if (xmlHashUpdateEntry2(ctxt->attsDefault, name, prefix, defaults, NULL) < 0) { xmlFree(defaults); goto mem_error; } } /* * Split the element name into prefix:localname , the string found * are within the DTD and hen not associated to namespace names. */ name = xmlSplitQName3(fullattr, &len); if (name == NULL) { name = xmlDictLookup(ctxt->dict, fullattr, -1); prefix = NULL; } else { name = xmlDictLookup(ctxt->dict, name, -1); prefix = xmlDictLookup(ctxt->dict, fullattr, len); } defaults->values[5 * defaults->nbAttrs] = name; defaults->values[5 * defaults->nbAttrs + 1] = prefix; /* intern the string and precompute the end */ len = xmlStrlen(value); value = xmlDictLookup(ctxt->dict, value, len); if (value == NULL) goto mem_error; defaults->values[5 * defaults->nbAttrs + 2] = value; defaults->values[5 * defaults->nbAttrs + 3] = value + len; if (ctxt->external) defaults->values[5 * defaults->nbAttrs + 4] = BAD_CAST "external"; else defaults->values[5 * defaults->nbAttrs + 4] = NULL; defaults->nbAttrs++; return; mem_error: xmlErrMemory(ctxt, NULL); return; } /** * xmlAddSpecialAttr: * @ctxt: an XML parser context * @fullname: the element fullname * @fullattr: the attribute fullname * @type: the attribute type * * Register this attribute type */ static void xmlAddSpecialAttr(xmlParserCtxtPtr ctxt, const xmlChar *fullname, const xmlChar *fullattr, int type) { if (ctxt->attsSpecial == NULL) { ctxt->attsSpecial = xmlHashCreateDict(10, ctxt->dict); if (ctxt->attsSpecial == NULL) goto mem_error; } if (xmlHashLookup2(ctxt->attsSpecial, fullname, fullattr) != NULL) return; xmlHashAddEntry2(ctxt->attsSpecial, fullname, fullattr, (void *) (ptrdiff_t) type); return; mem_error: xmlErrMemory(ctxt, NULL); return; } /** * xmlCleanSpecialAttrCallback: * * Removes CDATA attributes from the special attribute table */ static void xmlCleanSpecialAttrCallback(void *payload, void *data, const xmlChar *fullname, const xmlChar *fullattr, const xmlChar *unused ATTRIBUTE_UNUSED) { xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) data; if (((ptrdiff_t) payload) == XML_ATTRIBUTE_CDATA) { xmlHashRemoveEntry2(ctxt->attsSpecial, fullname, fullattr, NULL); } } /** * xmlCleanSpecialAttr: * @ctxt: an XML parser context * * Trim the list of attributes defined to remove all those of type * CDATA as they are not special. This call should be done when finishing * to parse the DTD and before starting to parse the document root. */ static void xmlCleanSpecialAttr(xmlParserCtxtPtr ctxt) { if (ctxt->attsSpecial == NULL) return; xmlHashScanFull(ctxt->attsSpecial, xmlCleanSpecialAttrCallback, ctxt); if (xmlHashSize(ctxt->attsSpecial) == 0) { xmlHashFree(ctxt->attsSpecial, NULL); ctxt->attsSpecial = NULL; } return; } /** * xmlCheckLanguageID: * @lang: pointer to the string value * * DEPRECATED: Internal function, do not use. * * Checks that the value conforms to the LanguageID production: * * NOTE: this is somewhat deprecated, those productions were removed from * the XML Second edition. * * [33] LanguageID ::= Langcode ('-' Subcode)* * [34] Langcode ::= ISO639Code | IanaCode | UserCode * [35] ISO639Code ::= ([a-z] | [A-Z]) ([a-z] | [A-Z]) * [36] IanaCode ::= ('i' | 'I') '-' ([a-z] | [A-Z])+ * [37] UserCode ::= ('x' | 'X') '-' ([a-z] | [A-Z])+ * [38] Subcode ::= ([a-z] | [A-Z])+ * * The current REC reference the successors of RFC 1766, currently 5646 * * http://www.rfc-editor.org/rfc/rfc5646.txt * langtag = language * ["-" script] * ["-" region] * *("-" variant) * *("-" extension) * ["-" privateuse] * language = 2*3ALPHA ; shortest ISO 639 code * ["-" extlang] ; sometimes followed by * ; extended language subtags * / 4ALPHA ; or reserved for future use * / 5*8ALPHA ; or registered language subtag * * extlang = 3ALPHA ; selected ISO 639 codes * *2("-" 3ALPHA) ; permanently reserved * * script = 4ALPHA ; ISO 15924 code * * region = 2ALPHA ; ISO 3166-1 code * / 3DIGIT ; UN M.49 code * * variant = 5*8alphanum ; registered variants * / (DIGIT 3alphanum) * * extension = singleton 1*("-" (2*8alphanum)) * * ; Single alphanumerics * ; "x" reserved for private use * singleton = DIGIT ; 0 - 9 * / %x41-57 ; A - W * / %x59-5A ; Y - Z * / %x61-77 ; a - w * / %x79-7A ; y - z * * it sounds right to still allow Irregular i-xxx IANA and user codes too * The parser below doesn't try to cope with extension or privateuse * that could be added but that's not interoperable anyway * * Returns 1 if correct 0 otherwise **/ int xmlCheckLanguageID(const xmlChar * lang) { const xmlChar *cur = lang, *nxt; if (cur == NULL) return (0); if (((cur[0] == 'i') && (cur[1] == '-')) || ((cur[0] == 'I') && (cur[1] == '-')) || ((cur[0] == 'x') && (cur[1] == '-')) || ((cur[0] == 'X') && (cur[1] == '-'))) { /* * Still allow IANA code and user code which were coming * from the previous version of the XML-1.0 specification * it's deprecated but we should not fail */ cur += 2; while (((cur[0] >= 'A') && (cur[0] <= 'Z')) || ((cur[0] >= 'a') && (cur[0] <= 'z'))) cur++; return(cur[0] == 0); } nxt = cur; while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) nxt++; if (nxt - cur >= 4) { /* * Reserved */ if ((nxt - cur > 8) || (nxt[0] != 0)) return(0); return(1); } if (nxt - cur < 2) return(0); /* we got an ISO 639 code */ if (nxt[0] == 0) return(1); if (nxt[0] != '-') return(0); nxt++; cur = nxt; /* now we can have extlang or script or region or variant */ if ((nxt[0] >= '0') && (nxt[0] <= '9')) goto region_m49; while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) nxt++; if (nxt - cur == 4) goto script; if (nxt - cur == 2) goto region; if ((nxt - cur >= 5) && (nxt - cur <= 8)) goto variant; if (nxt - cur != 3) return(0); /* we parsed an extlang */ if (nxt[0] == 0) return(1); if (nxt[0] != '-') return(0); nxt++; cur = nxt; /* now we can have script or region or variant */ if ((nxt[0] >= '0') && (nxt[0] <= '9')) goto region_m49; while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) nxt++; if (nxt - cur == 2) goto region; if ((nxt - cur >= 5) && (nxt - cur <= 8)) goto variant; if (nxt - cur != 4) return(0); /* we parsed a script */ script: if (nxt[0] == 0) return(1); if (nxt[0] != '-') return(0); nxt++; cur = nxt; /* now we can have region or variant */ if ((nxt[0] >= '0') && (nxt[0] <= '9')) goto region_m49; while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) nxt++; if ((nxt - cur >= 5) && (nxt - cur <= 8)) goto variant; if (nxt - cur != 2) return(0); /* we parsed a region */ region: if (nxt[0] == 0) return(1); if (nxt[0] != '-') return(0); nxt++; cur = nxt; /* now we can just have a variant */ while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) nxt++; if ((nxt - cur < 5) || (nxt - cur > 8)) return(0); /* we parsed a variant */ variant: if (nxt[0] == 0) return(1); if (nxt[0] != '-') return(0); /* extensions and private use subtags not checked */ return (1); region_m49: if (((nxt[1] >= '0') && (nxt[1] <= '9')) && ((nxt[2] >= '0') && (nxt[2] <= '9'))) { nxt += 3; goto region; } return(0); } /************************************************************************ * * * Parser stacks related functions and macros * * * ************************************************************************/ static xmlEntityPtr xmlParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar ** str); #ifdef SAX2 /** * nsPush: * @ctxt: an XML parser context * @prefix: the namespace prefix or NULL * @URL: the namespace name * * Pushes a new parser namespace on top of the ns stack * * Returns -1 in case of error, -2 if the namespace should be discarded * and the index in the stack otherwise. */ static int nsPush(xmlParserCtxtPtr ctxt, const xmlChar *prefix, const xmlChar *URL) { if (ctxt->options & XML_PARSE_NSCLEAN) { int i; for (i = ctxt->nsNr - 2;i >= 0;i -= 2) { if (ctxt->nsTab[i] == prefix) { /* in scope */ if (ctxt->nsTab[i + 1] == URL) return(-2); /* out of scope keep it */ break; } } } if ((ctxt->nsMax == 0) || (ctxt->nsTab == NULL)) { ctxt->nsMax = 10; ctxt->nsNr = 0; ctxt->nsTab = (const xmlChar **) xmlMalloc(ctxt->nsMax * sizeof(xmlChar *)); if (ctxt->nsTab == NULL) { xmlErrMemory(ctxt, NULL); ctxt->nsMax = 0; return (-1); } } else if (ctxt->nsNr >= ctxt->nsMax) { const xmlChar ** tmp; ctxt->nsMax *= 2; tmp = (const xmlChar **) xmlRealloc((char *) ctxt->nsTab, ctxt->nsMax * sizeof(ctxt->nsTab[0])); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); ctxt->nsMax /= 2; return (-1); } ctxt->nsTab = tmp; } ctxt->nsTab[ctxt->nsNr++] = prefix; ctxt->nsTab[ctxt->nsNr++] = URL; return (ctxt->nsNr); } /** * nsPop: * @ctxt: an XML parser context * @nr: the number to pop * * Pops the top @nr parser prefix/namespace from the ns stack * * Returns the number of namespaces removed */ static int nsPop(xmlParserCtxtPtr ctxt, int nr) { int i; if (ctxt->nsTab == NULL) return(0); if (ctxt->nsNr < nr) { xmlGenericError(xmlGenericErrorContext, "Pbm popping %d NS\n", nr); nr = ctxt->nsNr; } if (ctxt->nsNr <= 0) return (0); for (i = 0;i < nr;i++) { ctxt->nsNr--; ctxt->nsTab[ctxt->nsNr] = NULL; } return(nr); } #endif static int xmlCtxtGrowAttrs(xmlParserCtxtPtr ctxt, int nr) { const xmlChar **atts; int *attallocs; int maxatts; if (nr + 5 > ctxt->maxatts) { maxatts = ctxt->maxatts == 0 ? 55 : (nr + 5) * 2; atts = (const xmlChar **) xmlMalloc( maxatts * sizeof(const xmlChar *)); if (atts == NULL) goto mem_error; attallocs = (int *) xmlRealloc((void *) ctxt->attallocs, (maxatts / 5) * sizeof(int)); if (attallocs == NULL) { xmlFree(atts); goto mem_error; } if (ctxt->maxatts > 0) memcpy(atts, ctxt->atts, ctxt->maxatts * sizeof(const xmlChar *)); xmlFree(ctxt->atts); ctxt->atts = atts; ctxt->attallocs = attallocs; ctxt->maxatts = maxatts; } return(ctxt->maxatts); mem_error: xmlErrMemory(ctxt, NULL); return(-1); } /** * inputPush: * @ctxt: an XML parser context * @value: the parser input * * Pushes a new parser input on top of the input stack * * Returns -1 in case of error, the index in the stack otherwise */ int inputPush(xmlParserCtxtPtr ctxt, xmlParserInputPtr value) { if ((ctxt == NULL) || (value == NULL)) return(-1); if (ctxt->inputNr >= ctxt->inputMax) { size_t newSize = ctxt->inputMax * 2; xmlParserInputPtr *tmp; tmp = (xmlParserInputPtr *) xmlRealloc(ctxt->inputTab, newSize * sizeof(*tmp)); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); return (-1); } ctxt->inputTab = tmp; ctxt->inputMax = newSize; } ctxt->inputTab[ctxt->inputNr] = value; ctxt->input = value; return (ctxt->inputNr++); } /** * inputPop: * @ctxt: an XML parser context * * Pops the top parser input from the input stack * * Returns the input just removed */ xmlParserInputPtr inputPop(xmlParserCtxtPtr ctxt) { xmlParserInputPtr ret; if (ctxt == NULL) return(NULL); if (ctxt->inputNr <= 0) return (NULL); ctxt->inputNr--; if (ctxt->inputNr > 0) ctxt->input = ctxt->inputTab[ctxt->inputNr - 1]; else ctxt->input = NULL; ret = ctxt->inputTab[ctxt->inputNr]; ctxt->inputTab[ctxt->inputNr] = NULL; return (ret); } /** * nodePush: * @ctxt: an XML parser context * @value: the element node * * DEPRECATED: Internal function, do not use. * * Pushes a new element node on top of the node stack * * Returns -1 in case of error, the index in the stack otherwise */ int nodePush(xmlParserCtxtPtr ctxt, xmlNodePtr value) { if (ctxt == NULL) return(0); if (ctxt->nodeNr >= ctxt->nodeMax) { xmlNodePtr *tmp; tmp = (xmlNodePtr *) xmlRealloc(ctxt->nodeTab, ctxt->nodeMax * 2 * sizeof(ctxt->nodeTab[0])); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); return (-1); } ctxt->nodeTab = tmp; ctxt->nodeMax *= 2; } if ((((unsigned int) ctxt->nodeNr) > xmlParserMaxDepth) && ((ctxt->options & XML_PARSE_HUGE) == 0)) { xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR, "Excessive depth in document: %d use XML_PARSE_HUGE option\n", xmlParserMaxDepth); xmlHaltParser(ctxt); return(-1); } ctxt->nodeTab[ctxt->nodeNr] = value; ctxt->node = value; return (ctxt->nodeNr++); } /** * nodePop: * @ctxt: an XML parser context * * DEPRECATED: Internal function, do not use. * * Pops the top element node from the node stack * * Returns the node just removed */ xmlNodePtr nodePop(xmlParserCtxtPtr ctxt) { xmlNodePtr ret; if (ctxt == NULL) return(NULL); if (ctxt->nodeNr <= 0) return (NULL); ctxt->nodeNr--; if (ctxt->nodeNr > 0) ctxt->node = ctxt->nodeTab[ctxt->nodeNr - 1]; else ctxt->node = NULL; ret = ctxt->nodeTab[ctxt->nodeNr]; ctxt->nodeTab[ctxt->nodeNr] = NULL; return (ret); } /** * nameNsPush: * @ctxt: an XML parser context * @value: the element name * @prefix: the element prefix * @URI: the element namespace name * @line: the current line number for error messages * @nsNr: the number of namespaces pushed on the namespace table * * Pushes a new element name/prefix/URL on top of the name stack * * Returns -1 in case of error, the index in the stack otherwise */ static int nameNsPush(xmlParserCtxtPtr ctxt, const xmlChar * value, const xmlChar *prefix, const xmlChar *URI, int line, int nsNr) { xmlStartTag *tag; if (ctxt->nameNr >= ctxt->nameMax) { const xmlChar * *tmp; xmlStartTag *tmp2; ctxt->nameMax *= 2; tmp = (const xmlChar * *) xmlRealloc((xmlChar * *)ctxt->nameTab, ctxt->nameMax * sizeof(ctxt->nameTab[0])); if (tmp == NULL) { ctxt->nameMax /= 2; goto mem_error; } ctxt->nameTab = tmp; tmp2 = (xmlStartTag *) xmlRealloc((void * *)ctxt->pushTab, ctxt->nameMax * sizeof(ctxt->pushTab[0])); if (tmp2 == NULL) { ctxt->nameMax /= 2; goto mem_error; } ctxt->pushTab = tmp2; } else if (ctxt->pushTab == NULL) { ctxt->pushTab = (xmlStartTag *) xmlMalloc(ctxt->nameMax * sizeof(ctxt->pushTab[0])); if (ctxt->pushTab == NULL) goto mem_error; } ctxt->nameTab[ctxt->nameNr] = value; ctxt->name = value; tag = &ctxt->pushTab[ctxt->nameNr]; tag->prefix = prefix; tag->URI = URI; tag->line = line; tag->nsNr = nsNr; return (ctxt->nameNr++); mem_error: xmlErrMemory(ctxt, NULL); return (-1); } #ifdef LIBXML_PUSH_ENABLED /** * nameNsPop: * @ctxt: an XML parser context * * Pops the top element/prefix/URI name from the name stack * * Returns the name just removed */ static const xmlChar * nameNsPop(xmlParserCtxtPtr ctxt) { const xmlChar *ret; if (ctxt->nameNr <= 0) return (NULL); ctxt->nameNr--; if (ctxt->nameNr > 0) ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; else ctxt->name = NULL; ret = ctxt->nameTab[ctxt->nameNr]; ctxt->nameTab[ctxt->nameNr] = NULL; return (ret); } #endif /* LIBXML_PUSH_ENABLED */ /** * namePush: * @ctxt: an XML parser context * @value: the element name * * DEPRECATED: Internal function, do not use. * * Pushes a new element name on top of the name stack * * Returns -1 in case of error, the index in the stack otherwise */ int namePush(xmlParserCtxtPtr ctxt, const xmlChar * value) { if (ctxt == NULL) return (-1); if (ctxt->nameNr >= ctxt->nameMax) { const xmlChar * *tmp; tmp = (const xmlChar * *) xmlRealloc((xmlChar * *)ctxt->nameTab, ctxt->nameMax * 2 * sizeof(ctxt->nameTab[0])); if (tmp == NULL) { goto mem_error; } ctxt->nameTab = tmp; ctxt->nameMax *= 2; } ctxt->nameTab[ctxt->nameNr] = value; ctxt->name = value; return (ctxt->nameNr++); mem_error: xmlErrMemory(ctxt, NULL); return (-1); } /** * namePop: * @ctxt: an XML parser context * * DEPRECATED: Internal function, do not use. * * Pops the top element name from the name stack * * Returns the name just removed */ const xmlChar * namePop(xmlParserCtxtPtr ctxt) { const xmlChar *ret; if ((ctxt == NULL) || (ctxt->nameNr <= 0)) return (NULL); ctxt->nameNr--; if (ctxt->nameNr > 0) ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; else ctxt->name = NULL; ret = ctxt->nameTab[ctxt->nameNr]; ctxt->nameTab[ctxt->nameNr] = NULL; return (ret); } static int spacePush(xmlParserCtxtPtr ctxt, int val) { if (ctxt->spaceNr >= ctxt->spaceMax) { int *tmp; ctxt->spaceMax *= 2; tmp = (int *) xmlRealloc(ctxt->spaceTab, ctxt->spaceMax * sizeof(ctxt->spaceTab[0])); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); ctxt->spaceMax /=2; return(-1); } ctxt->spaceTab = tmp; } ctxt->spaceTab[ctxt->spaceNr] = val; ctxt->space = &ctxt->spaceTab[ctxt->spaceNr]; return(ctxt->spaceNr++); } static int spacePop(xmlParserCtxtPtr ctxt) { int ret; if (ctxt->spaceNr <= 0) return(0); ctxt->spaceNr--; if (ctxt->spaceNr > 0) ctxt->space = &ctxt->spaceTab[ctxt->spaceNr - 1]; else ctxt->space = &ctxt->spaceTab[0]; ret = ctxt->spaceTab[ctxt->spaceNr]; ctxt->spaceTab[ctxt->spaceNr] = -1; return(ret); } /* * Macros for accessing the content. Those should be used only by the parser, * and not exported. * * Dirty macros, i.e. one often need to make assumption on the context to * use them * * CUR_PTR return the current pointer to the xmlChar to be parsed. * To be used with extreme caution since operations consuming * characters may move the input buffer to a different location ! * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled * This should be used internally by the parser * only to compare to ASCII values otherwise it would break when * running with UTF-8 encoding. * RAW same as CUR but in the input buffer, bypass any token * extraction that may have been done * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only * to compare on ASCII based substring. * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined * strings without newlines within the parser. * NEXT1(l) Skip 1 xmlChar, and must also be used only to skip 1 non-newline ASCII * defined char within the parser. * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding * * NEXT Skip to the next character, this does the proper decoding * in UTF-8 mode. It also pop-up unfinished entities on the fly. * NEXTL(l) Skip the current unicode character of l xmlChars long. * CUR_CHAR(l) returns the current unicode character (int), set l * to the number of xmlChars used for the encoding [0-5]. * CUR_SCHAR same but operate on a string instead of the context * COPY_BUF copy the current unicode char to the target buffer, increment * the index * GROW, SHRINK handling of input buffers */ #define RAW (*ctxt->input->cur) #define CUR (*ctxt->input->cur) #define NXT(val) ctxt->input->cur[(val)] #define CUR_PTR ctxt->input->cur #define BASE_PTR ctxt->input->base #define CMP4( s, c1, c2, c3, c4 ) \ ( ((unsigned char *) s)[ 0 ] == c1 && ((unsigned char *) s)[ 1 ] == c2 && \ ((unsigned char *) s)[ 2 ] == c3 && ((unsigned char *) s)[ 3 ] == c4 ) #define CMP5( s, c1, c2, c3, c4, c5 ) \ ( CMP4( s, c1, c2, c3, c4 ) && ((unsigned char *) s)[ 4 ] == c5 ) #define CMP6( s, c1, c2, c3, c4, c5, c6 ) \ ( CMP5( s, c1, c2, c3, c4, c5 ) && ((unsigned char *) s)[ 5 ] == c6 ) #define CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) \ ( CMP6( s, c1, c2, c3, c4, c5, c6 ) && ((unsigned char *) s)[ 6 ] == c7 ) #define CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) \ ( CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) && ((unsigned char *) s)[ 7 ] == c8 ) #define CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) \ ( CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) && \ ((unsigned char *) s)[ 8 ] == c9 ) #define CMP10( s, c1, c2, c3, c4, c5, c6, c7, c8, c9, c10 ) \ ( CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) && \ ((unsigned char *) s)[ 9 ] == c10 ) #define SKIP(val) do { \ ctxt->input->cur += (val),ctxt->input->col+=(val); \ if (*ctxt->input->cur == 0) \ xmlParserGrow(ctxt); \ } while (0) #define SKIPL(val) do { \ int skipl; \ for(skipl=0; skiplinput->cur) == '\n') { \ ctxt->input->line++; ctxt->input->col = 1; \ } else ctxt->input->col++; \ ctxt->input->cur++; \ } \ if (*ctxt->input->cur == 0) \ xmlParserGrow(ctxt); \ } while (0) #define SHRINK if ((ctxt->progressive == 0) && \ (ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \ (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \ xmlParserShrink(ctxt); #define GROW if ((ctxt->progressive == 0) && \ (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \ xmlParserGrow(ctxt); #define SKIP_BLANKS xmlSkipBlankChars(ctxt) #define NEXT xmlNextChar(ctxt) #define NEXT1 { \ ctxt->input->col++; \ ctxt->input->cur++; \ if (*ctxt->input->cur == 0) \ xmlParserGrow(ctxt); \ } #define NEXTL(l) do { \ if (*(ctxt->input->cur) == '\n') { \ ctxt->input->line++; ctxt->input->col = 1; \ } else ctxt->input->col++; \ ctxt->input->cur += l; \ } while (0) #define CUR_CHAR(l) xmlCurrentChar(ctxt, &l) #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l) #define COPY_BUF(l,b,i,v) \ if (l == 1) b[i++] = v; \ else i += xmlCopyCharMultiByte(&b[i],v) /** * xmlSkipBlankChars: * @ctxt: the XML parser context * * DEPRECATED: Internal function, do not use. * * skip all blanks character found at that point in the input streams. * It pops up finished entities in the process if allowable at that point. * * Returns the number of space chars skipped */ int xmlSkipBlankChars(xmlParserCtxtPtr ctxt) { int res = 0; /* * It's Okay to use CUR/NEXT here since all the blanks are on * the ASCII range. */ if (((ctxt->inputNr == 1) && (ctxt->instate != XML_PARSER_DTD)) || (ctxt->instate == XML_PARSER_START)) { const xmlChar *cur; /* * if we are in the document content, go really fast */ cur = ctxt->input->cur; while (IS_BLANK_CH(*cur)) { if (*cur == '\n') { ctxt->input->line++; ctxt->input->col = 1; } else { ctxt->input->col++; } cur++; if (res < INT_MAX) res++; if (*cur == 0) { ctxt->input->cur = cur; xmlParserGrow(ctxt); cur = ctxt->input->cur; } } ctxt->input->cur = cur; } else { int expandPE = ((ctxt->external != 0) || (ctxt->inputNr != 1)); while (ctxt->instate != XML_PARSER_EOF) { if (IS_BLANK_CH(CUR)) { /* CHECKED tstblanks.xml */ NEXT; } else if (CUR == '%') { /* * Need to handle support of entities branching here */ if ((expandPE == 0) || (IS_BLANK_CH(NXT(1))) || (NXT(1) == 0)) break; xmlParsePEReference(ctxt); } else if (CUR == 0) { unsigned long consumed; xmlEntityPtr ent; if (ctxt->inputNr <= 1) break; consumed = ctxt->input->consumed; xmlSaturatedAddSizeT(&consumed, ctxt->input->cur - ctxt->input->base); /* * Add to sizeentities when parsing an external entity * for the first time. */ ent = ctxt->input->entity; if ((ent->etype == XML_EXTERNAL_PARAMETER_ENTITY) && ((ent->flags & XML_ENT_PARSED) == 0)) { ent->flags |= XML_ENT_PARSED; xmlSaturatedAdd(&ctxt->sizeentities, consumed); } xmlParserEntityCheck(ctxt, consumed); xmlPopInput(ctxt); } else { break; } /* * Also increase the counter when entering or exiting a PERef. * The spec says: "When a parameter-entity reference is recognized * in the DTD and included, its replacement text MUST be enlarged * by the attachment of one leading and one following space (#x20) * character." */ if (res < INT_MAX) res++; } } return(res); } /************************************************************************ * * * Commodity functions to handle entities * * * ************************************************************************/ /** * xmlPopInput: * @ctxt: an XML parser context * * xmlPopInput: the current input pointed by ctxt->input came to an end * pop it and return the next char. * * Returns the current xmlChar in the parser context */ xmlChar xmlPopInput(xmlParserCtxtPtr ctxt) { xmlParserInputPtr input; if ((ctxt == NULL) || (ctxt->inputNr <= 1)) return(0); if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "Popping input %d\n", ctxt->inputNr); if ((ctxt->inputNr > 1) && (ctxt->inSubset == 0) && (ctxt->instate != XML_PARSER_EOF)) xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "Unfinished entity outside the DTD"); input = inputPop(ctxt); if (input->entity != NULL) input->entity->flags &= ~XML_ENT_EXPANDING; xmlFreeInputStream(input); if (*ctxt->input->cur == 0) xmlParserGrow(ctxt); return(CUR); } /** * xmlPushInput: * @ctxt: an XML parser context * @input: an XML parser input fragment (entity, XML fragment ...). * * xmlPushInput: switch to a new input stream which is stacked on top * of the previous one(s). * Returns -1 in case of error or the index in the input stack */ int xmlPushInput(xmlParserCtxtPtr ctxt, xmlParserInputPtr input) { int ret; if (input == NULL) return(-1); if (xmlParserDebugEntities) { if ((ctxt->input != NULL) && (ctxt->input->filename)) xmlGenericError(xmlGenericErrorContext, "%s(%d): ", ctxt->input->filename, ctxt->input->line); xmlGenericError(xmlGenericErrorContext, "Pushing input %d : %.30s\n", ctxt->inputNr+1, input->cur); } if (((ctxt->inputNr > 40) && ((ctxt->options & XML_PARSE_HUGE) == 0)) || (ctxt->inputNr > 100)) { xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); while (ctxt->inputNr > 1) xmlFreeInputStream(inputPop(ctxt)); return(-1); } ret = inputPush(ctxt, input); if (ctxt->instate == XML_PARSER_EOF) return(-1); GROW; return(ret); } /** * xmlParseCharRef: * @ctxt: an XML parser context * * DEPRECATED: Internal function, don't use. * * Parse a numeric character reference. Always consumes '&'. * * [66] CharRef ::= '&#' [0-9]+ ';' | * '&#x' [0-9a-fA-F]+ ';' * * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. * * Returns the value parsed (as an int), 0 in case of error */ int xmlParseCharRef(xmlParserCtxtPtr ctxt) { int val = 0; int count = 0; /* * Using RAW/CUR/NEXT is okay since we are working on ASCII range here */ if ((RAW == '&') && (NXT(1) == '#') && (NXT(2) == 'x')) { SKIP(3); GROW; while (RAW != ';') { /* loop blocked by count */ if (count++ > 20) { count = 0; GROW; if (ctxt->instate == XML_PARSER_EOF) return(0); } if ((RAW >= '0') && (RAW <= '9')) val = val * 16 + (CUR - '0'); else if ((RAW >= 'a') && (RAW <= 'f') && (count < 20)) val = val * 16 + (CUR - 'a') + 10; else if ((RAW >= 'A') && (RAW <= 'F') && (count < 20)) val = val * 16 + (CUR - 'A') + 10; else { xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); val = 0; break; } if (val > 0x110000) val = 0x110000; NEXT; count++; } if (RAW == ';') { /* on purpose to avoid reentrancy problems with NEXT and SKIP */ ctxt->input->col++; ctxt->input->cur++; } } else if ((RAW == '&') && (NXT(1) == '#')) { SKIP(2); GROW; while (RAW != ';') { /* loop blocked by count */ if (count++ > 20) { count = 0; GROW; if (ctxt->instate == XML_PARSER_EOF) return(0); } if ((RAW >= '0') && (RAW <= '9')) val = val * 10 + (CUR - '0'); else { xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); val = 0; break; } if (val > 0x110000) val = 0x110000; NEXT; count++; } if (RAW == ';') { /* on purpose to avoid reentrancy problems with NEXT and SKIP */ ctxt->input->col++; ctxt->input->cur++; } } else { if (RAW == '&') SKIP(1); xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); } /* * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. */ if (val >= 0x110000) { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseCharRef: character reference out of bounds\n", val); } else if (IS_CHAR(val)) { return(val); } else { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseCharRef: invalid xmlChar value %d\n", val); } return(0); } /** * xmlParseStringCharRef: * @ctxt: an XML parser context * @str: a pointer to an index in the string * * parse Reference declarations, variant parsing from a string rather * than an an input flow. * * [66] CharRef ::= '&#' [0-9]+ ';' | * '&#x' [0-9a-fA-F]+ ';' * * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. * * Returns the value parsed (as an int), 0 in case of error, str will be * updated to the current value of the index */ static int xmlParseStringCharRef(xmlParserCtxtPtr ctxt, const xmlChar **str) { const xmlChar *ptr; xmlChar cur; int val = 0; if ((str == NULL) || (*str == NULL)) return(0); ptr = *str; cur = *ptr; if ((cur == '&') && (ptr[1] == '#') && (ptr[2] == 'x')) { ptr += 3; cur = *ptr; while (cur != ';') { /* Non input consuming loop */ if ((cur >= '0') && (cur <= '9')) val = val * 16 + (cur - '0'); else if ((cur >= 'a') && (cur <= 'f')) val = val * 16 + (cur - 'a') + 10; else if ((cur >= 'A') && (cur <= 'F')) val = val * 16 + (cur - 'A') + 10; else { xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); val = 0; break; } if (val > 0x110000) val = 0x110000; ptr++; cur = *ptr; } if (cur == ';') ptr++; } else if ((cur == '&') && (ptr[1] == '#')){ ptr += 2; cur = *ptr; while (cur != ';') { /* Non input consuming loops */ if ((cur >= '0') && (cur <= '9')) val = val * 10 + (cur - '0'); else { xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); val = 0; break; } if (val > 0x110000) val = 0x110000; ptr++; cur = *ptr; } if (cur == ';') ptr++; } else { xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); return(0); } *str = ptr; /* * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. */ if (val >= 0x110000) { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseStringCharRef: character reference out of bounds\n", val); } else if (IS_CHAR(val)) { return(val); } else { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseStringCharRef: invalid xmlChar value %d\n", val); } return(0); } /** * xmlParserHandlePEReference: * @ctxt: the parser context * * DEPRECATED: Internal function, do not use. * * [69] PEReference ::= '%' Name ';' * * [ WFC: No Recursion ] * A parsed entity must not contain a recursive * reference to itself, either directly or indirectly. * * [ WFC: Entity Declared ] * In a document without any DTD, a document with only an internal DTD * subset which contains no parameter entity references, or a document * with "standalone='yes'", ... ... The declaration of a parameter * entity must precede any reference to it... * * [ VC: Entity Declared ] * In a document with an external subset or external parameter entities * with "standalone='no'", ... ... The declaration of a parameter entity * must precede any reference to it... * * [ WFC: In DTD ] * Parameter-entity references may only appear in the DTD. * NOTE: misleading but this is handled. * * A PEReference may have been detected in the current input stream * the handling is done accordingly to * http://www.w3.org/TR/REC-xml#entproc * i.e. * - Included in literal in entity values * - Included as Parameter Entity reference within DTDs */ void xmlParserHandlePEReference(xmlParserCtxtPtr ctxt) { switch(ctxt->instate) { case XML_PARSER_CDATA_SECTION: return; case XML_PARSER_COMMENT: return; case XML_PARSER_START_TAG: return; case XML_PARSER_END_TAG: return; case XML_PARSER_EOF: xmlFatalErr(ctxt, XML_ERR_PEREF_AT_EOF, NULL); return; case XML_PARSER_PROLOG: case XML_PARSER_START: case XML_PARSER_MISC: xmlFatalErr(ctxt, XML_ERR_PEREF_IN_PROLOG, NULL); return; case XML_PARSER_ENTITY_DECL: case XML_PARSER_CONTENT: case XML_PARSER_ATTRIBUTE_VALUE: case XML_PARSER_PI: case XML_PARSER_SYSTEM_LITERAL: case XML_PARSER_PUBLIC_LITERAL: /* we just ignore it there */ return; case XML_PARSER_EPILOG: xmlFatalErr(ctxt, XML_ERR_PEREF_IN_EPILOG, NULL); return; case XML_PARSER_ENTITY_VALUE: /* * NOTE: in the case of entity values, we don't do the * substitution here since we need the literal * entity value to be able to save the internal * subset of the document. * This will be handled by xmlStringDecodeEntities */ return; case XML_PARSER_DTD: /* * [WFC: Well-Formedness Constraint: PEs in Internal Subset] * In the internal DTD subset, parameter-entity references * can occur only where markup declarations can occur, not * within markup declarations. * In that case this is handled in xmlParseMarkupDecl */ if ((ctxt->external == 0) && (ctxt->inputNr == 1)) return; if (IS_BLANK_CH(NXT(1)) || NXT(1) == 0) return; break; case XML_PARSER_IGNORE: return; } xmlParsePEReference(ctxt); } /* * Macro used to grow the current buffer. * buffer##_size is expected to be a size_t * mem_error: is expected to handle memory allocation failures */ #define growBuffer(buffer, n) { \ xmlChar *tmp; \ size_t new_size = buffer##_size * 2 + n; \ if (new_size < buffer##_size) goto mem_error; \ tmp = (xmlChar *) xmlRealloc(buffer, new_size); \ if (tmp == NULL) goto mem_error; \ buffer = tmp; \ buffer##_size = new_size; \ } /** * xmlStringDecodeEntitiesInt: * @ctxt: the parser context * @str: the input string * @len: the string length * @what: combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF * @end: an end marker xmlChar, 0 if none * @end2: an end marker xmlChar, 0 if none * @end3: an end marker xmlChar, 0 if none * @check: whether to perform entity checks */ static xmlChar * xmlStringDecodeEntitiesInt(xmlParserCtxtPtr ctxt, const xmlChar *str, int len, int what, xmlChar end, xmlChar end2, xmlChar end3, int check) { xmlChar *buffer = NULL; size_t buffer_size = 0; size_t nbchars = 0; xmlChar *current = NULL; xmlChar *rep = NULL; const xmlChar *last; xmlEntityPtr ent; int c,l; if (str == NULL) return(NULL); last = str + len; if (((ctxt->depth > 40) && ((ctxt->options & XML_PARSE_HUGE) == 0)) || (ctxt->depth > 100)) { xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_LOOP, "Maximum entity nesting depth exceeded"); return(NULL); } /* * allocate a translation buffer. */ buffer_size = XML_PARSER_BIG_BUFFER_SIZE; buffer = (xmlChar *) xmlMallocAtomic(buffer_size); if (buffer == NULL) goto mem_error; /* * OK loop until we reach one of the ending char or a size limit. * we are operating on already parsed values. */ if (str < last) c = CUR_SCHAR(str, l); else c = 0; while ((c != 0) && (c != end) && /* non input consuming loop */ (c != end2) && (c != end3) && (ctxt->instate != XML_PARSER_EOF)) { if (c == 0) break; if ((c == '&') && (str[1] == '#')) { int val = xmlParseStringCharRef(ctxt, &str); if (val == 0) goto int_error; COPY_BUF(0,buffer,nbchars,val); if (nbchars + XML_PARSER_BUFFER_SIZE > buffer_size) { growBuffer(buffer, XML_PARSER_BUFFER_SIZE); } } else if ((c == '&') && (what & XML_SUBSTITUTE_REF)) { if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "String decoding Entity Reference: %.30s\n", str); ent = xmlParseStringEntityRef(ctxt, &str); if ((ent != NULL) && (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { if (ent->content != NULL) { COPY_BUF(0,buffer,nbchars,ent->content[0]); if (nbchars + XML_PARSER_BUFFER_SIZE > buffer_size) { growBuffer(buffer, XML_PARSER_BUFFER_SIZE); } } else { xmlFatalErrMsg(ctxt, XML_ERR_INTERNAL_ERROR, "predefined entity has no content\n"); goto int_error; } } else if ((ent != NULL) && (ent->content != NULL)) { if ((check) && (xmlParserEntityCheck(ctxt, ent->length))) goto int_error; if (ent->flags & XML_ENT_EXPANDING) { xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); xmlHaltParser(ctxt); ent->content[0] = 0; goto int_error; } ent->flags |= XML_ENT_EXPANDING; ctxt->depth++; rep = xmlStringDecodeEntitiesInt(ctxt, ent->content, ent->length, what, 0, 0, 0, check); ctxt->depth--; ent->flags &= ~XML_ENT_EXPANDING; if (rep == NULL) { ent->content[0] = 0; goto int_error; } current = rep; while (*current != 0) { /* non input consuming loop */ buffer[nbchars++] = *current++; if (nbchars + XML_PARSER_BUFFER_SIZE > buffer_size) { growBuffer(buffer, XML_PARSER_BUFFER_SIZE); } } xmlFree(rep); rep = NULL; } else if (ent != NULL) { int i = xmlStrlen(ent->name); const xmlChar *cur = ent->name; buffer[nbchars++] = '&'; if (nbchars + i + XML_PARSER_BUFFER_SIZE > buffer_size) { growBuffer(buffer, i + XML_PARSER_BUFFER_SIZE); } for (;i > 0;i--) buffer[nbchars++] = *cur++; buffer[nbchars++] = ';'; } } else if (c == '%' && (what & XML_SUBSTITUTE_PEREF)) { if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "String decoding PE Reference: %.30s\n", str); ent = xmlParseStringPEReference(ctxt, &str); if (ent != NULL) { if (ent->content == NULL) { /* * Note: external parsed entities will not be loaded, * it is not required for a non-validating parser to * complete external PEReferences coming from the * internal subset */ if (((ctxt->options & XML_PARSE_NOENT) != 0) || ((ctxt->options & XML_PARSE_DTDVALID) != 0) || (ctxt->validate != 0)) { xmlLoadEntityContent(ctxt, ent); } else { xmlWarningMsg(ctxt, XML_ERR_ENTITY_PROCESSING, "not validating will not read content for PE entity %s\n", ent->name, NULL); } } if ((check) && (xmlParserEntityCheck(ctxt, ent->length))) goto int_error; if (ent->flags & XML_ENT_EXPANDING) { xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); xmlHaltParser(ctxt); if (ent->content != NULL) ent->content[0] = 0; goto int_error; } ent->flags |= XML_ENT_EXPANDING; ctxt->depth++; rep = xmlStringDecodeEntitiesInt(ctxt, ent->content, ent->length, what, 0, 0, 0, check); ctxt->depth--; ent->flags &= ~XML_ENT_EXPANDING; if (rep == NULL) { if (ent->content != NULL) ent->content[0] = 0; goto int_error; } current = rep; while (*current != 0) { /* non input consuming loop */ buffer[nbchars++] = *current++; if (nbchars + XML_PARSER_BUFFER_SIZE > buffer_size) { growBuffer(buffer, XML_PARSER_BUFFER_SIZE); } } xmlFree(rep); rep = NULL; } } else { COPY_BUF(l,buffer,nbchars,c); str += l; if (nbchars + XML_PARSER_BUFFER_SIZE > buffer_size) { growBuffer(buffer, XML_PARSER_BUFFER_SIZE); } } if (str < last) c = CUR_SCHAR(str, l); else c = 0; } buffer[nbchars] = 0; return(buffer); mem_error: xmlErrMemory(ctxt, NULL); int_error: if (rep != NULL) xmlFree(rep); if (buffer != NULL) xmlFree(buffer); return(NULL); } /** * xmlStringLenDecodeEntities: * @ctxt: the parser context * @str: the input string * @len: the string length * @what: combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF * @end: an end marker xmlChar, 0 if none * @end2: an end marker xmlChar, 0 if none * @end3: an end marker xmlChar, 0 if none * * DEPRECATED: Internal function, don't use. * * Takes a entity string content and process to do the adequate substitutions. * * [67] Reference ::= EntityRef | CharRef * * [69] PEReference ::= '%' Name ';' * * Returns A newly allocated string with the substitution done. The caller * must deallocate it ! */ xmlChar * xmlStringLenDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int len, int what, xmlChar end, xmlChar end2, xmlChar end3) { if ((ctxt == NULL) || (str == NULL) || (len < 0)) return(NULL); return(xmlStringDecodeEntitiesInt(ctxt, str, len, what, end, end2, end3, 0)); } /** * xmlStringDecodeEntities: * @ctxt: the parser context * @str: the input string * @what: combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF * @end: an end marker xmlChar, 0 if none * @end2: an end marker xmlChar, 0 if none * @end3: an end marker xmlChar, 0 if none * * DEPRECATED: Internal function, don't use. * * Takes a entity string content and process to do the adequate substitutions. * * [67] Reference ::= EntityRef | CharRef * * [69] PEReference ::= '%' Name ';' * * Returns A newly allocated string with the substitution done. The caller * must deallocate it ! */ xmlChar * xmlStringDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int what, xmlChar end, xmlChar end2, xmlChar end3) { if ((ctxt == NULL) || (str == NULL)) return(NULL); return(xmlStringDecodeEntitiesInt(ctxt, str, xmlStrlen(str), what, end, end2, end3, 0)); } /************************************************************************ * * * Commodity functions, cleanup needed ? * * * ************************************************************************/ /** * areBlanks: * @ctxt: an XML parser context * @str: a xmlChar * * @len: the size of @str * @blank_chars: we know the chars are blanks * * Is this a sequence of blank chars that one can ignore ? * * Returns 1 if ignorable 0 otherwise. */ static int areBlanks(xmlParserCtxtPtr ctxt, const xmlChar *str, int len, int blank_chars) { int i, ret; xmlNodePtr lastChild; /* * Don't spend time trying to differentiate them, the same callback is * used ! */ if (ctxt->sax->ignorableWhitespace == ctxt->sax->characters) return(0); /* * Check for xml:space value. */ if ((ctxt->space == NULL) || (*(ctxt->space) == 1) || (*(ctxt->space) == -2)) return(0); /* * Check that the string is made of blanks */ if (blank_chars == 0) { for (i = 0;i < len;i++) if (!(IS_BLANK_CH(str[i]))) return(0); } /* * Look if the element is mixed content in the DTD if available */ if (ctxt->node == NULL) return(0); if (ctxt->myDoc != NULL) { ret = xmlIsMixedElement(ctxt->myDoc, ctxt->node->name); if (ret == 0) return(1); if (ret == 1) return(0); } /* * Otherwise, heuristic :-\ */ if ((RAW != '<') && (RAW != 0xD)) return(0); if ((ctxt->node->children == NULL) && (RAW == '<') && (NXT(1) == '/')) return(0); lastChild = xmlGetLastChild(ctxt->node); if (lastChild == NULL) { if ((ctxt->node->type != XML_ELEMENT_NODE) && (ctxt->node->content != NULL)) return(0); } else if (xmlNodeIsText(lastChild)) return(0); else if ((ctxt->node->children != NULL) && (xmlNodeIsText(ctxt->node->children))) return(0); return(1); } /************************************************************************ * * * Extra stuff for namespace support * * Relates to http://www.w3.org/TR/WD-xml-names * * * ************************************************************************/ /** * xmlSplitQName: * @ctxt: an XML parser context * @name: an XML parser context * @prefix: a xmlChar ** * * parse an UTF8 encoded XML qualified name string * * [NS 5] QName ::= (Prefix ':')? LocalPart * * [NS 6] Prefix ::= NCName * * [NS 7] LocalPart ::= NCName * * Returns the local part, and prefix is updated * to get the Prefix if any. */ xmlChar * xmlSplitQName(xmlParserCtxtPtr ctxt, const xmlChar *name, xmlChar **prefix) { xmlChar buf[XML_MAX_NAMELEN + 5]; xmlChar *buffer = NULL; int len = 0; int max = XML_MAX_NAMELEN; xmlChar *ret = NULL; const xmlChar *cur = name; int c; if (prefix == NULL) return(NULL); *prefix = NULL; if (cur == NULL) return(NULL); #ifndef XML_XML_NAMESPACE /* xml: prefix is not really a namespace */ if ((cur[0] == 'x') && (cur[1] == 'm') && (cur[2] == 'l') && (cur[3] == ':')) return(xmlStrdup(name)); #endif /* nasty but well=formed */ if (cur[0] == ':') return(xmlStrdup(name)); c = *cur++; while ((c != 0) && (c != ':') && (len < max)) { /* tested bigname.xml */ buf[len++] = c; c = *cur++; } if (len >= max) { /* * Okay someone managed to make a huge name, so he's ready to pay * for the processing speed. */ max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while ((c != 0) && (c != ':')) { /* tested bigname.xml */ if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max); if (tmp == NULL) { xmlFree(buffer); xmlErrMemory(ctxt, NULL); return(NULL); } buffer = tmp; } buffer[len++] = c; c = *cur++; } buffer[len] = 0; } if ((c == ':') && (*cur == 0)) { if (buffer != NULL) xmlFree(buffer); *prefix = NULL; return(xmlStrdup(name)); } if (buffer == NULL) ret = xmlStrndup(buf, len); else { ret = buffer; buffer = NULL; max = XML_MAX_NAMELEN; } if (c == ':') { c = *cur; *prefix = ret; if (c == 0) { return(xmlStrndup(BAD_CAST "", 0)); } len = 0; /* * Check that the first character is proper to start * a new name */ if (!(((c >= 0x61) && (c <= 0x7A)) || ((c >= 0x41) && (c <= 0x5A)) || (c == '_') || (c == ':'))) { int l; int first = CUR_SCHAR(cur, l); if (!IS_LETTER(first) && (first != '_')) { xmlFatalErrMsgStr(ctxt, XML_NS_ERR_QNAME, "Name %s is not XML Namespace compliant\n", name); } } cur++; while ((c != 0) && (len < max)) { /* tested bigname2.xml */ buf[len++] = c; c = *cur++; } if (len >= max) { /* * Okay someone managed to make a huge name, so he's ready to pay * for the processing speed. */ max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while (c != 0) { /* tested bigname2.xml */ if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buffer); return(NULL); } buffer = tmp; } buffer[len++] = c; c = *cur++; } buffer[len] = 0; } if (buffer == NULL) ret = xmlStrndup(buf, len); else { ret = buffer; } } return(ret); } /************************************************************************ * * * The parser itself * * Relates to http://www.w3.org/TR/REC-xml * * * ************************************************************************/ /************************************************************************ * * * Routines to parse Name, NCName and NmToken * * * ************************************************************************/ #ifdef DEBUG static unsigned long nbParseName = 0; static unsigned long nbParseNmToken = 0; static unsigned long nbParseNCName = 0; static unsigned long nbParseNCNameComplex = 0; static unsigned long nbParseNameComplex = 0; static unsigned long nbParseStringName = 0; #endif /* * The two following functions are related to the change of accepted * characters for Name and NmToken in the Revision 5 of XML-1.0 * They correspond to the modified production [4] and the new production [4a] * changes in that revision. Also note that the macros used for the * productions Letter, Digit, CombiningChar and Extender are not needed * anymore. * We still keep compatibility to pre-revision5 parsing semantic if the * new XML_PARSE_OLD10 option is given to the parser. */ static int xmlIsNameStartChar(xmlParserCtxtPtr ctxt, int c) { if ((ctxt->options & XML_PARSE_OLD10) == 0) { /* * Use the new checks of production [4] [4a] amd [5] of the * Update 5 of XML-1.0 */ if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ (((c >= 'a') && (c <= 'z')) || ((c >= 'A') && (c <= 'Z')) || (c == '_') || (c == ':') || ((c >= 0xC0) && (c <= 0xD6)) || ((c >= 0xD8) && (c <= 0xF6)) || ((c >= 0xF8) && (c <= 0x2FF)) || ((c >= 0x370) && (c <= 0x37D)) || ((c >= 0x37F) && (c <= 0x1FFF)) || ((c >= 0x200C) && (c <= 0x200D)) || ((c >= 0x2070) && (c <= 0x218F)) || ((c >= 0x2C00) && (c <= 0x2FEF)) || ((c >= 0x3001) && (c <= 0xD7FF)) || ((c >= 0xF900) && (c <= 0xFDCF)) || ((c >= 0xFDF0) && (c <= 0xFFFD)) || ((c >= 0x10000) && (c <= 0xEFFFF)))) return(1); } else { if (IS_LETTER(c) || (c == '_') || (c == ':')) return(1); } return(0); } static int xmlIsNameChar(xmlParserCtxtPtr ctxt, int c) { if ((ctxt->options & XML_PARSE_OLD10) == 0) { /* * Use the new checks of production [4] [4a] amd [5] of the * Update 5 of XML-1.0 */ if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ (((c >= 'a') && (c <= 'z')) || ((c >= 'A') && (c <= 'Z')) || ((c >= '0') && (c <= '9')) || /* !start */ (c == '_') || (c == ':') || (c == '-') || (c == '.') || (c == 0xB7) || /* !start */ ((c >= 0xC0) && (c <= 0xD6)) || ((c >= 0xD8) && (c <= 0xF6)) || ((c >= 0xF8) && (c <= 0x2FF)) || ((c >= 0x300) && (c <= 0x36F)) || /* !start */ ((c >= 0x370) && (c <= 0x37D)) || ((c >= 0x37F) && (c <= 0x1FFF)) || ((c >= 0x200C) && (c <= 0x200D)) || ((c >= 0x203F) && (c <= 0x2040)) || /* !start */ ((c >= 0x2070) && (c <= 0x218F)) || ((c >= 0x2C00) && (c <= 0x2FEF)) || ((c >= 0x3001) && (c <= 0xD7FF)) || ((c >= 0xF900) && (c <= 0xFDCF)) || ((c >= 0xFDF0) && (c <= 0xFFFD)) || ((c >= 0x10000) && (c <= 0xEFFFF)))) return(1); } else { if ((IS_LETTER(c)) || (IS_DIGIT(c)) || (c == '.') || (c == '-') || (c == '_') || (c == ':') || (IS_COMBINING(c)) || (IS_EXTENDER(c))) return(1); } return(0); } static xmlChar * xmlParseAttValueInternal(xmlParserCtxtPtr ctxt, int *len, int *alloc, int normalize); static const xmlChar * xmlParseNameComplex(xmlParserCtxtPtr ctxt) { int len = 0, l; int c; int maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; #ifdef DEBUG nbParseNameComplex++; #endif /* * Handler for more complex cases */ c = CUR_CHAR(l); if ((ctxt->options & XML_PARSE_OLD10) == 0) { /* * Use the new checks of production [4] [4a] amd [5] of the * Update 5 of XML-1.0 */ if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */ (!(((c >= 'a') && (c <= 'z')) || ((c >= 'A') && (c <= 'Z')) || (c == '_') || (c == ':') || ((c >= 0xC0) && (c <= 0xD6)) || ((c >= 0xD8) && (c <= 0xF6)) || ((c >= 0xF8) && (c <= 0x2FF)) || ((c >= 0x370) && (c <= 0x37D)) || ((c >= 0x37F) && (c <= 0x1FFF)) || ((c >= 0x200C) && (c <= 0x200D)) || ((c >= 0x2070) && (c <= 0x218F)) || ((c >= 0x2C00) && (c <= 0x2FEF)) || ((c >= 0x3001) && (c <= 0xD7FF)) || ((c >= 0xF900) && (c <= 0xFDCF)) || ((c >= 0xFDF0) && (c <= 0xFFFD)) || ((c >= 0x10000) && (c <= 0xEFFFF))))) { return(NULL); } len += l; NEXTL(l); c = CUR_CHAR(l); while ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ (((c >= 'a') && (c <= 'z')) || ((c >= 'A') && (c <= 'Z')) || ((c >= '0') && (c <= '9')) || /* !start */ (c == '_') || (c == ':') || (c == '-') || (c == '.') || (c == 0xB7) || /* !start */ ((c >= 0xC0) && (c <= 0xD6)) || ((c >= 0xD8) && (c <= 0xF6)) || ((c >= 0xF8) && (c <= 0x2FF)) || ((c >= 0x300) && (c <= 0x36F)) || /* !start */ ((c >= 0x370) && (c <= 0x37D)) || ((c >= 0x37F) && (c <= 0x1FFF)) || ((c >= 0x200C) && (c <= 0x200D)) || ((c >= 0x203F) && (c <= 0x2040)) || /* !start */ ((c >= 0x2070) && (c <= 0x218F)) || ((c >= 0x2C00) && (c <= 0x2FEF)) || ((c >= 0x3001) && (c <= 0xD7FF)) || ((c >= 0xF900) && (c <= 0xFDCF)) || ((c >= 0xFDF0) && (c <= 0xFFFD)) || ((c >= 0x10000) && (c <= 0xEFFFF)) )) { if (len <= INT_MAX - l) len += l; NEXTL(l); c = CUR_CHAR(l); } } else { if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */ (!IS_LETTER(c) && (c != '_') && (c != ':'))) { return(NULL); } len += l; NEXTL(l); c = CUR_CHAR(l); while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */ ((IS_LETTER(c)) || (IS_DIGIT(c)) || (c == '.') || (c == '-') || (c == '_') || (c == ':') || (IS_COMBINING(c)) || (IS_EXTENDER(c)))) { if (len <= INT_MAX - l) len += l; NEXTL(l); c = CUR_CHAR(l); } } if (ctxt->instate == XML_PARSER_EOF) return(NULL); if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Name"); return(NULL); } if (ctxt->input->cur - ctxt->input->base < len) { /* * There were a couple of bugs where PERefs lead to to a change * of the buffer. Check the buffer size to avoid passing an invalid * pointer to xmlDictLookup. */ xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "unexpected change of input buffer"); return (NULL); } if ((*ctxt->input->cur == '\n') && (ctxt->input->cur[-1] == '\r')) return(xmlDictLookup(ctxt->dict, ctxt->input->cur - (len + 1), len)); return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len)); } /** * xmlParseName: * @ctxt: an XML parser context * * DEPRECATED: Internal function, don't use. * * parse an XML name. * * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | * CombiningChar | Extender * * [5] Name ::= (Letter | '_' | ':') (NameChar)* * * [6] Names ::= Name (#x20 Name)* * * Returns the Name parsed or NULL */ const xmlChar * xmlParseName(xmlParserCtxtPtr ctxt) { const xmlChar *in; const xmlChar *ret; size_t count = 0; size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; GROW; if (ctxt->instate == XML_PARSER_EOF) return(NULL); #ifdef DEBUG nbParseName++; #endif /* * Accelerator for simple ASCII names */ in = ctxt->input->cur; if (((*in >= 0x61) && (*in <= 0x7A)) || ((*in >= 0x41) && (*in <= 0x5A)) || (*in == '_') || (*in == ':')) { in++; while (((*in >= 0x61) && (*in <= 0x7A)) || ((*in >= 0x41) && (*in <= 0x5A)) || ((*in >= 0x30) && (*in <= 0x39)) || (*in == '_') || (*in == '-') || (*in == ':') || (*in == '.')) in++; if ((*in > 0) && (*in < 0x80)) { count = in - ctxt->input->cur; if (count > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Name"); return(NULL); } ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count); ctxt->input->cur = in; ctxt->input->col += count; if (ret == NULL) xmlErrMemory(ctxt, NULL); return(ret); } } /* accelerator for special cases */ return(xmlParseNameComplex(ctxt)); } static const xmlChar * xmlParseNCNameComplex(xmlParserCtxtPtr ctxt) { int len = 0, l; int c; int maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; size_t startPosition = 0; #ifdef DEBUG nbParseNCNameComplex++; #endif /* * Handler for more complex cases */ startPosition = CUR_PTR - BASE_PTR; c = CUR_CHAR(l); if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */ (!xmlIsNameStartChar(ctxt, c) || (c == ':'))) { return(NULL); } while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */ (xmlIsNameChar(ctxt, c) && (c != ':'))) { if (len <= INT_MAX - l) len += l; NEXTL(l); c = CUR_CHAR(l); } if (ctxt->instate == XML_PARSER_EOF) return(NULL); if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); return(NULL); } return(xmlDictLookup(ctxt->dict, (BASE_PTR + startPosition), len)); } /** * xmlParseNCName: * @ctxt: an XML parser context * @len: length of the string parsed * * parse an XML name. * * [4NS] NCNameChar ::= Letter | Digit | '.' | '-' | '_' | * CombiningChar | Extender * * [5NS] NCName ::= (Letter | '_') (NCNameChar)* * * Returns the Name parsed or NULL */ static const xmlChar * xmlParseNCName(xmlParserCtxtPtr ctxt) { const xmlChar *in, *e; const xmlChar *ret; size_t count = 0; size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; #ifdef DEBUG nbParseNCName++; #endif /* * Accelerator for simple ASCII names */ in = ctxt->input->cur; e = ctxt->input->end; if ((((*in >= 0x61) && (*in <= 0x7A)) || ((*in >= 0x41) && (*in <= 0x5A)) || (*in == '_')) && (in < e)) { in++; while ((((*in >= 0x61) && (*in <= 0x7A)) || ((*in >= 0x41) && (*in <= 0x5A)) || ((*in >= 0x30) && (*in <= 0x39)) || (*in == '_') || (*in == '-') || (*in == '.')) && (in < e)) in++; if (in >= e) goto complex; if ((*in > 0) && (*in < 0x80)) { count = in - ctxt->input->cur; if (count > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); return(NULL); } ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count); ctxt->input->cur = in; ctxt->input->col += count; if (ret == NULL) { xmlErrMemory(ctxt, NULL); } return(ret); } } complex: return(xmlParseNCNameComplex(ctxt)); } /** * xmlParseNameAndCompare: * @ctxt: an XML parser context * * parse an XML name and compares for match * (specialized for endtag parsing) * * Returns NULL for an illegal name, (xmlChar*) 1 for success * and the name for mismatch */ static const xmlChar * xmlParseNameAndCompare(xmlParserCtxtPtr ctxt, xmlChar const *other) { register const xmlChar *cmp = other; register const xmlChar *in; const xmlChar *ret; GROW; if (ctxt->instate == XML_PARSER_EOF) return(NULL); in = ctxt->input->cur; while (*in != 0 && *in == *cmp) { ++in; ++cmp; } if (*cmp == 0 && (*in == '>' || IS_BLANK_CH (*in))) { /* success */ ctxt->input->col += in - ctxt->input->cur; ctxt->input->cur = in; return (const xmlChar*) 1; } /* failure (or end of input buffer), check with full function */ ret = xmlParseName (ctxt); /* strings coming from the dictionary direct compare possible */ if (ret == other) { return (const xmlChar*) 1; } return ret; } /** * xmlParseStringName: * @ctxt: an XML parser context * @str: a pointer to the string pointer (IN/OUT) * * parse an XML name. * * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | * CombiningChar | Extender * * [5] Name ::= (Letter | '_' | ':') (NameChar)* * * [6] Names ::= Name (#x20 Name)* * * Returns the Name parsed or NULL. The @str pointer * is updated to the current location in the string. */ static xmlChar * xmlParseStringName(xmlParserCtxtPtr ctxt, const xmlChar** str) { xmlChar buf[XML_MAX_NAMELEN + 5]; const xmlChar *cur = *str; int len = 0, l; int c; int maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; #ifdef DEBUG nbParseStringName++; #endif c = CUR_SCHAR(cur, l); if (!xmlIsNameStartChar(ctxt, c)) { return(NULL); } COPY_BUF(l,buf,len,c); cur += l; c = CUR_SCHAR(cur, l); while (xmlIsNameChar(ctxt, c)) { COPY_BUF(l,buf,len,c); cur += l; c = CUR_SCHAR(cur, l); if (len >= XML_MAX_NAMELEN) { /* test bigentname.xml */ /* * Okay someone managed to make a huge name, so he's ready to pay * for the processing speed. */ xmlChar *buffer; int max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while (xmlIsNameChar(ctxt, c)) { if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buffer); return(NULL); } buffer = tmp; } COPY_BUF(l,buffer,len,c); cur += l; c = CUR_SCHAR(cur, l); if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); xmlFree(buffer); return(NULL); } } buffer[len] = 0; *str = cur; return(buffer); } } if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); return(NULL); } *str = cur; return(xmlStrndup(buf, len)); } /** * xmlParseNmtoken: * @ctxt: an XML parser context * * DEPRECATED: Internal function, don't use. * * parse an XML Nmtoken. * * [7] Nmtoken ::= (NameChar)+ * * [8] Nmtokens ::= Nmtoken (#x20 Nmtoken)* * * Returns the Nmtoken parsed or NULL */ xmlChar * xmlParseNmtoken(xmlParserCtxtPtr ctxt) { xmlChar buf[XML_MAX_NAMELEN + 5]; int len = 0, l; int c; int maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; #ifdef DEBUG nbParseNmToken++; #endif c = CUR_CHAR(l); while (xmlIsNameChar(ctxt, c)) { COPY_BUF(l,buf,len,c); NEXTL(l); c = CUR_CHAR(l); if (len >= XML_MAX_NAMELEN) { /* * Okay someone managed to make a huge token, so he's ready to pay * for the processing speed. */ xmlChar *buffer; int max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while (xmlIsNameChar(ctxt, c)) { if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buffer); return(NULL); } buffer = tmp; } COPY_BUF(l,buffer,len,c); if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NmToken"); xmlFree(buffer); return(NULL); } NEXTL(l); c = CUR_CHAR(l); } buffer[len] = 0; if (ctxt->instate == XML_PARSER_EOF) { xmlFree(buffer); return(NULL); } return(buffer); } } if (ctxt->instate == XML_PARSER_EOF) return(NULL); if (len == 0) return(NULL); if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NmToken"); return(NULL); } return(xmlStrndup(buf, len)); } /** * xmlParseEntityValue: * @ctxt: an XML parser context * @orig: if non-NULL store a copy of the original entity value * * DEPRECATED: Internal function, don't use. * * parse a value for ENTITY declarations * * [9] EntityValue ::= '"' ([^%&"] | PEReference | Reference)* '"' | * "'" ([^%&'] | PEReference | Reference)* "'" * * Returns the EntityValue parsed with reference substituted or NULL */ xmlChar * xmlParseEntityValue(xmlParserCtxtPtr ctxt, xmlChar **orig) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; int c, l; int maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_HUGE_LENGTH : XML_MAX_TEXT_LENGTH; xmlChar stop; xmlChar *ret = NULL; const xmlChar *cur = NULL; xmlParserInputPtr input; if (RAW == '"') stop = '"'; else if (RAW == '\'') stop = '\''; else { xmlFatalErr(ctxt, XML_ERR_ENTITY_NOT_STARTED, NULL); return(NULL); } buf = (xmlChar *) xmlMallocAtomic(size); if (buf == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } /* * The content of the entity definition is copied in a buffer. */ ctxt->instate = XML_PARSER_ENTITY_VALUE; input = ctxt->input; GROW; if (ctxt->instate == XML_PARSER_EOF) goto error; NEXT; c = CUR_CHAR(l); /* * NOTE: 4.4.5 Included in Literal * When a parameter entity reference appears in a literal entity * value, ... a single or double quote character in the replacement * text is always treated as a normal data character and will not * terminate the literal. * In practice it means we stop the loop only when back at parsing * the initial entity and the quote is found */ while (((IS_CHAR(c)) && ((c != stop) || /* checked */ (ctxt->input != input))) && (ctxt->instate != XML_PARSER_EOF)) { if (len + 5 >= size) { xmlChar *tmp; size *= 2; tmp = (xmlChar *) xmlRealloc(buf, size); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); goto error; } buf = tmp; } COPY_BUF(l,buf,len,c); NEXTL(l); GROW; c = CUR_CHAR(l); if (c == 0) { GROW; c = CUR_CHAR(l); } if (len > maxLength) { xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_NOT_FINISHED, "entity value too long\n"); goto error; } } buf[len] = 0; if (ctxt->instate == XML_PARSER_EOF) goto error; if (c != stop) { xmlFatalErr(ctxt, XML_ERR_ENTITY_NOT_FINISHED, NULL); goto error; } NEXT; /* * Raise problem w.r.t. '&' and '%' being used in non-entities * reference constructs. Note Charref will be handled in * xmlStringDecodeEntities() */ cur = buf; while (*cur != 0) { /* non input consuming */ if ((*cur == '%') || ((*cur == '&') && (cur[1] != '#'))) { xmlChar *name; xmlChar tmp = *cur; int nameOk = 0; cur++; name = xmlParseStringName(ctxt, &cur); if (name != NULL) { nameOk = 1; xmlFree(name); } if ((nameOk == 0) || (*cur != ';')) { xmlFatalErrMsgInt(ctxt, XML_ERR_ENTITY_CHAR_ERROR, "EntityValue: '%c' forbidden except for entities references\n", tmp); goto error; } if ((tmp == '%') && (ctxt->inSubset == 1) && (ctxt->inputNr == 1)) { xmlFatalErr(ctxt, XML_ERR_ENTITY_PE_INTERNAL, NULL); goto error; } if (*cur == 0) break; } cur++; } /* * Then PEReference entities are substituted. * * NOTE: 4.4.7 Bypassed * When a general entity reference appears in the EntityValue in * an entity declaration, it is bypassed and left as is. * so XML_SUBSTITUTE_REF is not set here. */ ++ctxt->depth; ret = xmlStringDecodeEntitiesInt(ctxt, buf, len, XML_SUBSTITUTE_PEREF, 0, 0, 0, /* check */ 1); --ctxt->depth; if (orig != NULL) { *orig = buf; buf = NULL; } error: if (buf != NULL) xmlFree(buf); return(ret); } /** * xmlParseAttValueComplex: * @ctxt: an XML parser context * @len: the resulting attribute len * @normalize: whether to apply the inner normalization * * parse a value for an attribute, this is the fallback function * of xmlParseAttValue() when the attribute parsing requires handling * of non-ASCII characters, or normalization compaction. * * Returns the AttValue parsed or NULL. The value has to be freed by the caller. */ static xmlChar * xmlParseAttValueComplex(xmlParserCtxtPtr ctxt, int *attlen, int normalize) { xmlChar limit = 0; xmlChar *buf = NULL; xmlChar *rep = NULL; size_t len = 0; size_t buf_size = 0; size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_HUGE_LENGTH : XML_MAX_TEXT_LENGTH; int c, l, in_space = 0; xmlChar *current = NULL; xmlEntityPtr ent; if (NXT(0) == '"') { ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; limit = '"'; NEXT; } else if (NXT(0) == '\'') { limit = '\''; ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; NEXT; } else { xmlFatalErr(ctxt, XML_ERR_ATTRIBUTE_NOT_STARTED, NULL); return(NULL); } /* * allocate a translation buffer. */ buf_size = XML_PARSER_BUFFER_SIZE; buf = (xmlChar *) xmlMallocAtomic(buf_size); if (buf == NULL) goto mem_error; /* * OK loop until we reach one of the ending char or a size limit. */ c = CUR_CHAR(l); while (((NXT(0) != limit) && /* checked */ (IS_CHAR(c)) && (c != '<')) && (ctxt->instate != XML_PARSER_EOF)) { if (c == '&') { in_space = 0; if (NXT(1) == '#') { int val = xmlParseCharRef(ctxt); if (val == '&') { if (ctxt->replaceEntities) { if (len + 10 > buf_size) { growBuffer(buf, 10); } buf[len++] = '&'; } else { /* * The reparsing will be done in xmlStringGetNodeList() * called by the attribute() function in SAX.c */ if (len + 10 > buf_size) { growBuffer(buf, 10); } buf[len++] = '&'; buf[len++] = '#'; buf[len++] = '3'; buf[len++] = '8'; buf[len++] = ';'; } } else if (val != 0) { if (len + 10 > buf_size) { growBuffer(buf, 10); } len += xmlCopyChar(0, &buf[len], val); } } else { ent = xmlParseEntityRef(ctxt); if ((ent != NULL) && (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { if (len + 10 > buf_size) { growBuffer(buf, 10); } if ((ctxt->replaceEntities == 0) && (ent->content[0] == '&')) { buf[len++] = '&'; buf[len++] = '#'; buf[len++] = '3'; buf[len++] = '8'; buf[len++] = ';'; } else { buf[len++] = ent->content[0]; } } else if ((ent != NULL) && (ctxt->replaceEntities != 0)) { if (ent->etype != XML_INTERNAL_PREDEFINED_ENTITY) { if (xmlParserEntityCheck(ctxt, ent->length)) goto error; ++ctxt->depth; rep = xmlStringDecodeEntitiesInt(ctxt, ent->content, ent->length, XML_SUBSTITUTE_REF, 0, 0, 0, /* check */ 1); --ctxt->depth; if (rep != NULL) { current = rep; while (*current != 0) { /* non input consuming */ if ((*current == 0xD) || (*current == 0xA) || (*current == 0x9)) { buf[len++] = 0x20; current++; } else buf[len++] = *current++; if (len + 10 > buf_size) { growBuffer(buf, 10); } } xmlFree(rep); rep = NULL; } } else { if (len + 10 > buf_size) { growBuffer(buf, 10); } if (ent->content != NULL) buf[len++] = ent->content[0]; } } else if (ent != NULL) { int i = xmlStrlen(ent->name); const xmlChar *cur = ent->name; /* * We also check for recursion and amplification * when entities are not substituted. They're * often expanded later. */ if ((ent->etype != XML_INTERNAL_PREDEFINED_ENTITY) && (ent->content != NULL)) { if ((ent->flags & XML_ENT_CHECKED) == 0) { unsigned long oldCopy = ctxt->sizeentcopy; ctxt->sizeentcopy = ent->length; ++ctxt->depth; rep = xmlStringDecodeEntitiesInt(ctxt, ent->content, ent->length, XML_SUBSTITUTE_REF, 0, 0, 0, /* check */ 1); --ctxt->depth; /* * If we're parsing DTD content, the entity * might reference other entities which * weren't defined yet, so the check isn't * reliable. */ if (ctxt->inSubset == 0) { ent->flags |= XML_ENT_CHECKED; ent->expandedSize = ctxt->sizeentcopy; } if (rep != NULL) { xmlFree(rep); rep = NULL; } else { ent->content[0] = 0; } if (xmlParserEntityCheck(ctxt, oldCopy)) goto error; } else { if (xmlParserEntityCheck(ctxt, ent->expandedSize)) goto error; } } /* * Just output the reference */ buf[len++] = '&'; while (len + i + 10 > buf_size) { growBuffer(buf, i + 10); } for (;i > 0;i--) buf[len++] = *cur++; buf[len++] = ';'; } } } else { if ((c == 0x20) || (c == 0xD) || (c == 0xA) || (c == 0x9)) { if ((len != 0) || (!normalize)) { if ((!normalize) || (!in_space)) { COPY_BUF(l,buf,len,0x20); while (len + 10 > buf_size) { growBuffer(buf, 10); } } in_space = 1; } } else { in_space = 0; COPY_BUF(l,buf,len,c); if (len + 10 > buf_size) { growBuffer(buf, 10); } } NEXTL(l); } GROW; c = CUR_CHAR(l); if (len > maxLength) { xmlFatalErrMsg(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED, "AttValue length too long\n"); goto mem_error; } } if (ctxt->instate == XML_PARSER_EOF) goto error; if ((in_space) && (normalize)) { while ((len > 0) && (buf[len - 1] == 0x20)) len--; } buf[len] = 0; if (RAW == '<') { xmlFatalErr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, NULL); } else if (RAW != limit) { if ((c != 0) && (!IS_CHAR(c))) { xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, "invalid character in attribute value\n"); } else { xmlFatalErrMsg(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED, "AttValue: ' expected\n"); } } else NEXT; if (attlen != NULL) *attlen = len; return(buf); mem_error: xmlErrMemory(ctxt, NULL); error: if (buf != NULL) xmlFree(buf); if (rep != NULL) xmlFree(rep); return(NULL); } /** * xmlParseAttValue: * @ctxt: an XML parser context * * DEPRECATED: Internal function, don't use. * * parse a value for an attribute * Note: the parser won't do substitution of entities here, this * will be handled later in xmlStringGetNodeList * * [10] AttValue ::= '"' ([^<&"] | Reference)* '"' | * "'" ([^<&'] | Reference)* "'" * * 3.3.3 Attribute-Value Normalization: * Before the value of an attribute is passed to the application or * checked for validity, the XML processor must normalize it as follows: * - a character reference is processed by appending the referenced * character to the attribute value * - an entity reference is processed by recursively processing the * replacement text of the entity * - a whitespace character (#x20, #xD, #xA, #x9) is processed by * appending #x20 to the normalized value, except that only a single * #x20 is appended for a "#xD#xA" sequence that is part of an external * parsed entity or the literal entity value of an internal parsed entity * - other characters are processed by appending them to the normalized value * If the declared value is not CDATA, then the XML processor must further * process the normalized attribute value by discarding any leading and * trailing space (#x20) characters, and by replacing sequences of space * (#x20) characters by a single space (#x20) character. * All attributes for which no declaration has been read should be treated * by a non-validating parser as if declared CDATA. * * Returns the AttValue parsed or NULL. The value has to be freed by the caller. */ xmlChar * xmlParseAttValue(xmlParserCtxtPtr ctxt) { if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL); return(xmlParseAttValueInternal(ctxt, NULL, NULL, 0)); } /** * xmlParseSystemLiteral: * @ctxt: an XML parser context * * DEPRECATED: Internal function, don't use. * * parse an XML Literal * * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'") * * Returns the SystemLiteral parsed or NULL */ xmlChar * xmlParseSystemLiteral(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; int cur, l; int maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; xmlChar stop; int state = ctxt->instate; if (RAW == '"') { NEXT; stop = '"'; } else if (RAW == '\'') { NEXT; stop = '\''; } else { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); return(NULL); } buf = (xmlChar *) xmlMallocAtomic(size); if (buf == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } ctxt->instate = XML_PARSER_SYSTEM_LITERAL; cur = CUR_CHAR(l); while ((IS_CHAR(cur)) && (cur != stop)) { /* checked */ if (len + 5 >= size) { xmlChar *tmp; size *= 2; tmp = (xmlChar *) xmlRealloc(buf, size); if (tmp == NULL) { xmlFree(buf); xmlErrMemory(ctxt, NULL); ctxt->instate = (xmlParserInputState) state; return(NULL); } buf = tmp; } COPY_BUF(l,buf,len,cur); if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "SystemLiteral"); xmlFree(buf); ctxt->instate = (xmlParserInputState) state; return(NULL); } NEXTL(l); cur = CUR_CHAR(l); } buf[len] = 0; if (ctxt->instate == XML_PARSER_EOF) { xmlFree(buf); return(NULL); } ctxt->instate = (xmlParserInputState) state; if (!IS_CHAR(cur)) { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); } else { NEXT; } return(buf); } /** * xmlParsePubidLiteral: * @ctxt: an XML parser context * * DEPRECATED: Internal function, don't use. * * parse an XML public literal * * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" * * Returns the PubidLiteral parsed or NULL. */ xmlChar * xmlParsePubidLiteral(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; int maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_TEXT_LENGTH : XML_MAX_NAME_LENGTH; xmlChar cur; xmlChar stop; xmlParserInputState oldstate = ctxt->instate; if (RAW == '"') { NEXT; stop = '"'; } else if (RAW == '\'') { NEXT; stop = '\''; } else { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); return(NULL); } buf = (xmlChar *) xmlMallocAtomic(size); if (buf == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } ctxt->instate = XML_PARSER_PUBLIC_LITERAL; cur = CUR; while ((IS_PUBIDCHAR_CH(cur)) && (cur != stop)) { /* checked */ if (len + 1 >= size) { xmlChar *tmp; size *= 2; tmp = (xmlChar *) xmlRealloc(buf, size); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buf); return(NULL); } buf = tmp; } buf[len++] = cur; if (len > maxLength) { xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Public ID"); xmlFree(buf); return(NULL); } NEXT; cur = CUR; } buf[len] = 0; if (ctxt->instate == XML_PARSER_EOF) { xmlFree(buf); return(NULL); } if (cur != stop) { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); } else { NEXTL(1); } ctxt->instate = oldstate; return(buf); } static void xmlParseCharDataComplex(xmlParserCtxtPtr ctxt); /* * used for the test in the inner loop of the char data testing */ static const unsigned char test_char_data[256] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x09, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x9, CR/LF separated */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x00, 0x27, /* & */ 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F, 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3A, 0x3B, 0x00, 0x3D, 0x3E, 0x3F, /* < */ 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0x5B, 0x5C, 0x00, 0x5E, 0x5F, /* ] */ 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* non-ascii */ 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, 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, 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 }; /** * xmlParseCharData: * @ctxt: an XML parser context * @cdata: unused * * DEPRECATED: Internal function, don't use. * * Parse character data. Always makes progress if the first char isn't * '<' or '&'. * * if we are within a CDATA section ']]>' marks an end of section. * * The right angle bracket (>) may be represented using the string ">", * and must, for compatibility, be escaped using ">" or a character * reference when it appears in the string "]]>" in content, when that * string is not marking the end of a CDATA section. * * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*) */ void xmlParseCharData(xmlParserCtxtPtr ctxt, ATTRIBUTE_UNUSED int cdata) { const xmlChar *in; int nbchar = 0; int line = ctxt->input->line; int col = ctxt->input->col; int ccol; GROW; /* * Accelerated common case where input don't need to be * modified before passing it to the handler. */ in = ctxt->input->cur; do { get_more_space: while (*in == 0x20) { in++; ctxt->input->col++; } if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); goto get_more_space; } if (*in == '<') { nbchar = in - ctxt->input->cur; if (nbchar > 0) { const xmlChar *tmp = ctxt->input->cur; ctxt->input->cur = in; if ((ctxt->sax != NULL) && (ctxt->sax->ignorableWhitespace != ctxt->sax->characters)) { if (areBlanks(ctxt, tmp, nbchar, 1)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, tmp, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, tmp, nbchar); if (*ctxt->space == -1) *ctxt->space = -2; } } else if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) { ctxt->sax->characters(ctxt->userData, tmp, nbchar); } } return; } get_more: ccol = ctxt->input->col; while (test_char_data[*in]) { in++; ccol++; } ctxt->input->col = ccol; if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); goto get_more; } if (*in == ']') { if ((in[1] == ']') && (in[2] == '>')) { xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); if (ctxt->instate != XML_PARSER_EOF) ctxt->input->cur = in + 1; return; } in++; ctxt->input->col++; goto get_more; } nbchar = in - ctxt->input->cur; if (nbchar > 0) { if ((ctxt->sax != NULL) && (ctxt->sax->ignorableWhitespace != ctxt->sax->characters) && (IS_BLANK_CH(*ctxt->input->cur))) { const xmlChar *tmp = ctxt->input->cur; ctxt->input->cur = in; if (areBlanks(ctxt, tmp, nbchar, 0)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, tmp, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, tmp, nbchar); if (*ctxt->space == -1) *ctxt->space = -2; } line = ctxt->input->line; col = ctxt->input->col; } else if (ctxt->sax != NULL) { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, ctxt->input->cur, nbchar); line = ctxt->input->line; col = ctxt->input->col; } } ctxt->input->cur = in; if (*in == 0xD) { in++; if (*in == 0xA) { ctxt->input->cur = in; in++; ctxt->input->line++; ctxt->input->col = 1; continue; /* while */ } in--; } if (*in == '<') { return; } if (*in == '&') { return; } SHRINK; GROW; if (ctxt->instate == XML_PARSER_EOF) return; in = ctxt->input->cur; } while (((*in >= 0x20) && (*in <= 0x7F)) || (*in == 0x09) || (*in == 0x0a)); ctxt->input->line = line; ctxt->input->col = col; xmlParseCharDataComplex(ctxt); } /** * xmlParseCharDataComplex: * @ctxt: an XML parser context * @cdata: int indicating whether we are within a CDATA section * * Always makes progress if the first char isn't '<' or '&'. * * parse a CharData section.this is the fallback function * of xmlParseCharData() when the parsing requires handling * of non-ASCII characters. */ static void xmlParseCharDataComplex(xmlParserCtxtPtr ctxt) { xmlChar buf[XML_PARSER_BIG_BUFFER_SIZE + 5]; int nbchar = 0; int cur, l; cur = CUR_CHAR(l); while ((cur != '<') && /* checked */ (cur != '&') && (IS_CHAR(cur))) /* test also done in xmlCurrentChar() */ { if ((cur == ']') && (NXT(1) == ']') && (NXT(2) == '>')) { xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); } COPY_BUF(l,buf,nbchar,cur); /* move current position before possible calling of ctxt->sax->characters */ NEXTL(l); if (nbchar >= XML_PARSER_BIG_BUFFER_SIZE) { buf[nbchar] = 0; /* * OK the segment is to be consumed as chars. */ if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { if (areBlanks(ctxt, buf, nbchar, 0)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, buf, nbchar); if ((ctxt->sax->characters != ctxt->sax->ignorableWhitespace) && (*ctxt->space == -1)) *ctxt->space = -2; } } nbchar = 0; /* something really bad happened in the SAX callback */ if (ctxt->instate != XML_PARSER_CONTENT) return; SHRINK; } cur = CUR_CHAR(l); } if (ctxt->instate == XML_PARSER_EOF) return; if (nbchar != 0) { buf[nbchar] = 0; /* * OK the segment is to be consumed as chars. */ if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { if (areBlanks(ctxt, buf, nbchar, 0)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, buf, nbchar); if ((ctxt->sax->characters != ctxt->sax->ignorableWhitespace) && (*ctxt->space == -1)) *ctxt->space = -2; } } } if ((ctxt->input->cur < ctxt->input->end) && (!IS_CHAR(cur))) { /* Generate the error and skip the offending character */ xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "PCDATA invalid Char value %d\n", cur ? cur : CUR); NEXT; } } /** * xmlParseExternalID: * @ctxt: an XML parser context * @publicID: a xmlChar** receiving PubidLiteral * @strict: indicate whether we should restrict parsing to only * production [75], see NOTE below * * DEPRECATED: Internal function, don't use. * * Parse an External ID or a Public ID * * NOTE: Productions [75] and [83] interact badly since [75] can generate * 'PUBLIC' S PubidLiteral S SystemLiteral * * [75] ExternalID ::= 'SYSTEM' S SystemLiteral * | 'PUBLIC' S PubidLiteral S SystemLiteral * * [83] PublicID ::= 'PUBLIC' S PubidLiteral * * Returns the function returns SystemLiteral and in the second * case publicID receives PubidLiteral, is strict is off * it is possible to return NULL and have publicID set. */ xmlChar * xmlParseExternalID(xmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) { xmlChar *URI = NULL; *publicID = NULL; if (CMP6(CUR_PTR, 'S', 'Y', 'S', 'T', 'E', 'M')) { SKIP(6); if (SKIP_BLANKS == 0) { xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Space required after 'SYSTEM'\n"); } URI = xmlParseSystemLiteral(ctxt); if (URI == NULL) { xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); } } else if (CMP6(CUR_PTR, 'P', 'U', 'B', 'L', 'I', 'C')) { SKIP(6); if (SKIP_BLANKS == 0) { xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Space required after 'PUBLIC'\n"); } *publicID = xmlParsePubidLiteral(ctxt); if (*publicID == NULL) { xmlFatalErr(ctxt, XML_ERR_PUBID_REQUIRED, NULL); } if (strict) { /* * We don't handle [83] so "S SystemLiteral" is required. */ if (SKIP_BLANKS == 0) { xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Space required after the Public Identifier\n"); } } else { /* * We handle [83] so we return immediately, if * "S SystemLiteral" is not detected. We skip blanks if no * system literal was found, but this is harmless since we must * be at the end of a NotationDecl. */ if (SKIP_BLANKS == 0) return(NULL); if ((CUR != '\'') && (CUR != '"')) return(NULL); } URI = xmlParseSystemLiteral(ctxt); if (URI == NULL) { xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); } } return(URI); } /** * xmlParseCommentComplex: * @ctxt: an XML parser context * @buf: the already parsed part of the buffer * @len: number of bytes in the buffer * @size: allocated size of the buffer * * Skip an XML (SGML) comment * The spec says that "For compatibility, the string "--" (double-hyphen) * must not occur within comments. " * This is the slow routine in case the accelerator for ascii didn't work * * [15] Comment ::= '' */ static void xmlParseCommentComplex(xmlParserCtxtPtr ctxt, xmlChar *buf, size_t len, size_t size) { int q, ql; int r, rl; int cur, l; size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_HUGE_LENGTH : XML_MAX_TEXT_LENGTH; int inputid; inputid = ctxt->input->id; if (buf == NULL) { len = 0; size = XML_PARSER_BUFFER_SIZE; buf = (xmlChar *) xmlMallocAtomic(size); if (buf == NULL) { xmlErrMemory(ctxt, NULL); return; } } q = CUR_CHAR(ql); if (q == 0) goto not_terminated; if (!IS_CHAR(q)) { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseComment: invalid xmlChar value %d\n", q); xmlFree (buf); return; } NEXTL(ql); r = CUR_CHAR(rl); if (r == 0) goto not_terminated; if (!IS_CHAR(r)) { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseComment: invalid xmlChar value %d\n", r); xmlFree (buf); return; } NEXTL(rl); cur = CUR_CHAR(l); if (cur == 0) goto not_terminated; while (IS_CHAR(cur) && /* checked */ ((cur != '>') || (r != '-') || (q != '-'))) { if ((r == '-') && (q == '-')) { xmlFatalErr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, NULL); } if (len + 5 >= size) { xmlChar *new_buf; size_t new_size; new_size = size * 2; new_buf = (xmlChar *) xmlRealloc(buf, new_size); if (new_buf == NULL) { xmlFree (buf); xmlErrMemory(ctxt, NULL); return; } buf = new_buf; size = new_size; } COPY_BUF(ql,buf,len,q); if (len > maxLength) { xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, "Comment too big found", NULL); xmlFree (buf); return; } q = r; ql = rl; r = cur; rl = l; NEXTL(l); cur = CUR_CHAR(l); } buf[len] = 0; if (ctxt->instate == XML_PARSER_EOF) { xmlFree(buf); return; } if (cur == 0) { xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, "Comment not terminated \n' */ void xmlParseComment(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; size_t size = XML_PARSER_BUFFER_SIZE; size_t len = 0; size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? XML_MAX_HUGE_LENGTH : XML_MAX_TEXT_LENGTH; xmlParserInputState state; const xmlChar *in; size_t nbchar = 0; int ccol; int inputid; /* * Check that there is a comment right here. */ if ((RAW != '<') || (NXT(1) != '!')) return; SKIP(2); if ((RAW != '-') || (NXT(1) != '-')) return; state = ctxt->instate; ctxt->instate = XML_PARSER_COMMENT; inputid = ctxt->input->id; SKIP(2); GROW; /* * Accelerated common case where input don't need to be * modified before passing it to the handler. */ in = ctxt->input->cur; do { if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); } get_more: ccol = ctxt->input->col; while (((*in > '-') && (*in <= 0x7F)) || ((*in >= 0x20) && (*in < '-')) || (*in == 0x09)) { in++; ccol++; } ctxt->input->col = ccol; if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); goto get_more; } nbchar = in - ctxt->input->cur; /* * save current set of data */ if (nbchar > 0) { if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL)) { if (buf == NULL) { if ((*in == '-') && (in[1] == '-')) size = nbchar + 1; else size = XML_PARSER_BUFFER_SIZE + nbchar; buf = (xmlChar *) xmlMallocAtomic(size); if (buf == NULL) { xmlErrMemory(ctxt, NULL); ctxt->instate = state; return; } len = 0; } else if (len + nbchar + 1 >= size) { xmlChar *new_buf; size += len + nbchar + XML_PARSER_BUFFER_SIZE; new_buf = (xmlChar *) xmlRealloc(buf, size); if (new_buf == NULL) { xmlFree (buf); xmlErrMemory(ctxt, NULL); ctxt->instate = state; return; } buf = new_buf; } memcpy(&buf[len], ctxt->input->cur, nbchar); len += nbchar; buf[len] = 0; } } if (len > maxLength) { xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, "Comment too big found", NULL); xmlFree (buf); return; } ctxt->input->cur = in; if (*in == 0xA) { in++; ctxt->input->line++; ctxt->input->col = 1; } if (*in == 0xD) { in++; if (*in == 0xA) { ctxt->input->cur = in; in++; ctxt->input->line++; ctxt->input->col = 1; goto get_more; } in--; } SHRINK; GROW; if (ctxt->instate == XML_PARSER_EOF) { xmlFree(buf); return; } in = ctxt->input->cur; if (*in == '-') { if (in[1] == '-') { if (in[2] == '>') { if (ctxt->input->id != inputid) { xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_BOUNDARY, "comment doesn't start and stop in the" " same entity\n"); } SKIP(3); if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) && (!ctxt->disableSAX)) { if (buf != NULL) ctxt->sax->comment(ctxt->userData, buf); else ctxt->sax->comment(ctxt->userData, BAD_CAST ""); } if (buf != NULL) xmlFree(buf); if (ctxt->instate != XML_PARSER_EOF) ctxt->instate = state; return; } if (buf != NULL) { xmlFatalErrMsgStr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, "Double hyphen within comment: " " as comment */ start = cur; continue; } } else if ((*cur == '"') || (*cur == '\'') || (*cur == ']')) { state = *cur; } cur++; } /* * Rescan the three last characters to detect "" * split across chunks. */ if ((state == 0) || (state == '-')) { if (cur - start < 3) cur = start; else cur -= 3; } index = cur - ctxt->input->cur; if (index > LONG_MAX) { ctxt->checkIndex = 0; ctxt->endCheckState = 0; return(1); } ctxt->checkIndex = index; ctxt->endCheckState = state; return(0); } /** * xmlCheckCdataPush: * @cur: pointer to the block of characters * @len: length of the block in bytes * @complete: 1 if complete CDATA block is passed in, 0 if partial block * * Check that the block of characters is okay as SCdata content [20] * * Returns the number of bytes to pass if okay, a negative index where an * UTF-8 error occurred otherwise */ static int xmlCheckCdataPush(const xmlChar *utf, int len, int complete) { int ix; unsigned char c; int codepoint; if ((utf == NULL) || (len <= 0)) return(0); for (ix = 0; ix < len;) { /* string is 0-terminated */ c = utf[ix]; if ((c & 0x80) == 0x00) { /* 1-byte code, starts with 10 */ if (c >= 0x20) ix++; else if ((c == 0xA) || (c == 0xD) || (c == 0x9)) ix++; else return(-ix); } else if ((c & 0xe0) == 0xc0) {/* 2-byte code, starts with 110 */ if (ix + 2 > len) return(complete ? -ix : ix); if ((utf[ix+1] & 0xc0 ) != 0x80) return(-ix); codepoint = (utf[ix] & 0x1f) << 6; codepoint |= utf[ix+1] & 0x3f; if (!xmlIsCharQ(codepoint)) return(-ix); ix += 2; } else if ((c & 0xf0) == 0xe0) {/* 3-byte code, starts with 1110 */ if (ix + 3 > len) return(complete ? -ix : ix); if (((utf[ix+1] & 0xc0) != 0x80) || ((utf[ix+2] & 0xc0) != 0x80)) return(-ix); codepoint = (utf[ix] & 0xf) << 12; codepoint |= (utf[ix+1] & 0x3f) << 6; codepoint |= utf[ix+2] & 0x3f; if (!xmlIsCharQ(codepoint)) return(-ix); ix += 3; } else if ((c & 0xf8) == 0xf0) {/* 4-byte code, starts with 11110 */ if (ix + 4 > len) return(complete ? -ix : ix); if (((utf[ix+1] & 0xc0) != 0x80) || ((utf[ix+2] & 0xc0) != 0x80) || ((utf[ix+3] & 0xc0) != 0x80)) return(-ix); codepoint = (utf[ix] & 0x7) << 18; codepoint |= (utf[ix+1] & 0x3f) << 12; codepoint |= (utf[ix+2] & 0x3f) << 6; codepoint |= utf[ix+3] & 0x3f; if (!xmlIsCharQ(codepoint)) return(-ix); ix += 4; } else /* unknown encoding */ return(-ix); } return(ix); } /** * xmlParseTryOrFinish: * @ctxt: an XML parser context * @terminate: last chunk indicator * * Try to progress on parsing * * Returns zero if no parsing was possible */ static int xmlParseTryOrFinish(xmlParserCtxtPtr ctxt, int terminate) { int ret = 0; int tlen; size_t avail; xmlChar cur, next; if (ctxt->input == NULL) return(0); #ifdef DEBUG_PUSH switch (ctxt->instate) { case XML_PARSER_EOF: xmlGenericError(xmlGenericErrorContext, "PP: try EOF\n"); break; case XML_PARSER_START: xmlGenericError(xmlGenericErrorContext, "PP: try START\n"); break; case XML_PARSER_MISC: xmlGenericError(xmlGenericErrorContext, "PP: try MISC\n");break; case XML_PARSER_COMMENT: xmlGenericError(xmlGenericErrorContext, "PP: try COMMENT\n");break; case XML_PARSER_PROLOG: xmlGenericError(xmlGenericErrorContext, "PP: try PROLOG\n");break; case XML_PARSER_START_TAG: xmlGenericError(xmlGenericErrorContext, "PP: try START_TAG\n");break; case XML_PARSER_CONTENT: xmlGenericError(xmlGenericErrorContext, "PP: try CONTENT\n");break; case XML_PARSER_CDATA_SECTION: xmlGenericError(xmlGenericErrorContext, "PP: try CDATA_SECTION\n");break; case XML_PARSER_END_TAG: xmlGenericError(xmlGenericErrorContext, "PP: try END_TAG\n");break; case XML_PARSER_ENTITY_DECL: xmlGenericError(xmlGenericErrorContext, "PP: try ENTITY_DECL\n");break; case XML_PARSER_ENTITY_VALUE: xmlGenericError(xmlGenericErrorContext, "PP: try ENTITY_VALUE\n");break; case XML_PARSER_ATTRIBUTE_VALUE: xmlGenericError(xmlGenericErrorContext, "PP: try ATTRIBUTE_VALUE\n");break; case XML_PARSER_DTD: xmlGenericError(xmlGenericErrorContext, "PP: try DTD\n");break; case XML_PARSER_EPILOG: xmlGenericError(xmlGenericErrorContext, "PP: try EPILOG\n");break; case XML_PARSER_PI: xmlGenericError(xmlGenericErrorContext, "PP: try PI\n");break; case XML_PARSER_IGNORE: xmlGenericError(xmlGenericErrorContext, "PP: try IGNORE\n");break; } #endif if ((ctxt->input != NULL) && (ctxt->input->cur - ctxt->input->base > 4096)) { xmlParserShrink(ctxt); } while (ctxt->instate != XML_PARSER_EOF) { if ((ctxt->errNo != XML_ERR_OK) && (ctxt->disableSAX == 1)) return(0); if (ctxt->input == NULL) break; if (ctxt->input->buf != NULL) { /* * If we are operating on converted input, try to flush * remaining chars to avoid them stalling in the non-converted * buffer. */ if ((ctxt->input->buf->raw != NULL) && (xmlBufIsEmpty(ctxt->input->buf->raw) == 0)) { size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input); size_t current = ctxt->input->cur - ctxt->input->base; int res; res = xmlParserInputBufferPush(ctxt->input->buf, 0, ""); xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, current); if (res < 0) { xmlFatalErr(ctxt, ctxt->input->buf->error, NULL); xmlHaltParser(ctxt); return(0); } } } avail = ctxt->input->end - ctxt->input->cur; if (avail < 1) goto done; switch (ctxt->instate) { case XML_PARSER_EOF: /* * Document parsing is done ! */ goto done; case XML_PARSER_START: if (ctxt->charset == XML_CHAR_ENCODING_NONE) { xmlChar start[4]; xmlCharEncoding enc; /* * Very first chars read from the document flow. */ if (avail < 4) goto done; /* * Get the 4 first bytes and decode the charset * if enc != XML_CHAR_ENCODING_NONE * plug some encoding conversion routines, * else xmlSwitchEncoding will set to (default) * UTF8. */ start[0] = RAW; start[1] = NXT(1); start[2] = NXT(2); start[3] = NXT(3); enc = xmlDetectCharEncoding(start, 4); /* * We need more bytes to detect EBCDIC code pages. * See xmlDetectEBCDIC. */ if ((enc == XML_CHAR_ENCODING_EBCDIC) && (!terminate) && (avail < 200)) goto done; xmlSwitchEncoding(ctxt, enc); break; } if (avail < 2) goto done; cur = ctxt->input->cur[0]; next = ctxt->input->cur[1]; if (cur == 0) { if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator); xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); xmlHaltParser(ctxt); #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering EOF\n"); #endif if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) ctxt->sax->endDocument(ctxt->userData); goto done; } if ((cur == '<') && (next == '?')) { /* PI or XML decl */ if (avail < 5) goto done; if ((!terminate) && (!xmlParseLookupString(ctxt, 2, "?>", 2))) goto done; if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator); if ((ctxt->input->cur[2] == 'x') && (ctxt->input->cur[3] == 'm') && (ctxt->input->cur[4] == 'l') && (IS_BLANK_CH(ctxt->input->cur[5]))) { ret += 5; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: Parsing XML Decl\n"); #endif xmlParseXMLDecl(ctxt); if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) { /* * The XML REC instructs us to stop parsing right * here */ xmlHaltParser(ctxt); return(0); } ctxt->standalone = ctxt->input->standalone; if ((ctxt->encoding == NULL) && (ctxt->input->encoding != NULL)) ctxt->encoding = xmlStrdup(ctxt->input->encoding); if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) ctxt->sax->startDocument(ctxt->userData); ctxt->instate = XML_PARSER_MISC; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering MISC\n"); #endif } else { ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) ctxt->sax->startDocument(ctxt->userData); ctxt->instate = XML_PARSER_MISC; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering MISC\n"); #endif } } else { if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator); ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); if (ctxt->version == NULL) { xmlErrMemory(ctxt, NULL); break; } if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) ctxt->sax->startDocument(ctxt->userData); ctxt->instate = XML_PARSER_MISC; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering MISC\n"); #endif } break; case XML_PARSER_START_TAG: { const xmlChar *name; const xmlChar *prefix = NULL; const xmlChar *URI = NULL; int line = ctxt->input->line; int nsNr = ctxt->nsNr; if ((avail < 2) && (ctxt->inputNr == 1)) goto done; cur = ctxt->input->cur[0]; if (cur != '<') { xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); xmlHaltParser(ctxt); if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) ctxt->sax->endDocument(ctxt->userData); goto done; } if ((!terminate) && (!xmlParseLookupGt(ctxt))) goto done; if (ctxt->spaceNr == 0) spacePush(ctxt, -1); else if (*ctxt->space == -2) spacePush(ctxt, -1); else spacePush(ctxt, *ctxt->space); #ifdef LIBXML_SAX1_ENABLED if (ctxt->sax2) #endif /* LIBXML_SAX1_ENABLED */ name = xmlParseStartTag2(ctxt, &prefix, &URI, &tlen); #ifdef LIBXML_SAX1_ENABLED else name = xmlParseStartTag(ctxt); #endif /* LIBXML_SAX1_ENABLED */ if (ctxt->instate == XML_PARSER_EOF) goto done; if (name == NULL) { spacePop(ctxt); xmlHaltParser(ctxt); if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) ctxt->sax->endDocument(ctxt->userData); goto done; } #ifdef LIBXML_VALID_ENABLED /* * [ VC: Root Element Type ] * The Name in the document type declaration must match * the element type of the root element. */ if (ctxt->validate && ctxt->wellFormed && ctxt->myDoc && ctxt->node && (ctxt->node == ctxt->myDoc->children)) ctxt->valid &= xmlValidateRoot(&ctxt->vctxt, ctxt->myDoc); #endif /* LIBXML_VALID_ENABLED */ /* * Check for an Empty Element. */ if ((RAW == '/') && (NXT(1) == '>')) { SKIP(2); if (ctxt->sax2) { if ((ctxt->sax != NULL) && (ctxt->sax->endElementNs != NULL) && (!ctxt->disableSAX)) ctxt->sax->endElementNs(ctxt->userData, name, prefix, URI); if (ctxt->nsNr - nsNr > 0) nsPop(ctxt, ctxt->nsNr - nsNr); #ifdef LIBXML_SAX1_ENABLED } else { if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL) && (!ctxt->disableSAX)) ctxt->sax->endElement(ctxt->userData, name); #endif /* LIBXML_SAX1_ENABLED */ } if (ctxt->instate == XML_PARSER_EOF) goto done; spacePop(ctxt); if (ctxt->nameNr == 0) { ctxt->instate = XML_PARSER_EPILOG; } else { ctxt->instate = XML_PARSER_CONTENT; } break; } if (RAW == '>') { NEXT; } else { xmlFatalErrMsgStr(ctxt, XML_ERR_GT_REQUIRED, "Couldn't find end of Start Tag %s\n", name); nodePop(ctxt); spacePop(ctxt); } nameNsPush(ctxt, name, prefix, URI, line, ctxt->nsNr - nsNr); ctxt->instate = XML_PARSER_CONTENT; break; } case XML_PARSER_CONTENT: { if ((avail < 2) && (ctxt->inputNr == 1)) goto done; cur = ctxt->input->cur[0]; next = ctxt->input->cur[1]; if ((cur == '<') && (next == '/')) { ctxt->instate = XML_PARSER_END_TAG; break; } else if ((cur == '<') && (next == '?')) { if ((!terminate) && (!xmlParseLookupString(ctxt, 2, "?>", 2))) goto done; xmlParsePI(ctxt); ctxt->instate = XML_PARSER_CONTENT; } else if ((cur == '<') && (next != '!')) { ctxt->instate = XML_PARSER_START_TAG; break; } else if ((cur == '<') && (next == '!') && (ctxt->input->cur[2] == '-') && (ctxt->input->cur[3] == '-')) { if ((!terminate) && (!xmlParseLookupString(ctxt, 4, "-->", 3))) goto done; xmlParseComment(ctxt); ctxt->instate = XML_PARSER_CONTENT; } else if ((cur == '<') && (ctxt->input->cur[1] == '!') && (ctxt->input->cur[2] == '[') && (ctxt->input->cur[3] == 'C') && (ctxt->input->cur[4] == 'D') && (ctxt->input->cur[5] == 'A') && (ctxt->input->cur[6] == 'T') && (ctxt->input->cur[7] == 'A') && (ctxt->input->cur[8] == '[')) { SKIP(9); ctxt->instate = XML_PARSER_CDATA_SECTION; break; } else if ((cur == '<') && (next == '!') && (avail < 9)) { goto done; } else if (cur == '<') { xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "detected an error in element content\n"); SKIP(1); } else if (cur == '&') { if ((!terminate) && (!xmlParseLookupChar(ctxt, ';'))) goto done; xmlParseReference(ctxt); } else { /* TODO Avoid the extra copy, handle directly !!! */ /* * Goal of the following test is: * - minimize calls to the SAX 'character' callback * when they are mergeable * - handle an problem for isBlank when we only parse * a sequence of blank chars and the next one is * not available to check against '<' presence. * - tries to homogenize the differences in SAX * callbacks between the push and pull versions * of the parser. */ if ((ctxt->inputNr == 1) && (avail < XML_PARSER_BIG_BUFFER_SIZE)) { if ((!terminate) && (!xmlParseLookupCharData(ctxt))) goto done; } ctxt->checkIndex = 0; xmlParseCharData(ctxt, 0); } break; } case XML_PARSER_END_TAG: if (avail < 2) goto done; if ((!terminate) && (!xmlParseLookupChar(ctxt, '>'))) goto done; if (ctxt->sax2) { xmlParseEndTag2(ctxt, &ctxt->pushTab[ctxt->nameNr - 1]); nameNsPop(ctxt); } #ifdef LIBXML_SAX1_ENABLED else xmlParseEndTag1(ctxt, 0); #endif /* LIBXML_SAX1_ENABLED */ if (ctxt->instate == XML_PARSER_EOF) { /* Nothing */ } else if (ctxt->nameNr == 0) { ctxt->instate = XML_PARSER_EPILOG; } else { ctxt->instate = XML_PARSER_CONTENT; } break; case XML_PARSER_CDATA_SECTION: { /* * The Push mode need to have the SAX callback for * cdataBlock merge back contiguous callbacks. */ const xmlChar *term; if (terminate) { /* * Don't call xmlParseLookupString. If 'terminate' * is set, checkIndex is invalid. */ term = BAD_CAST strstr((const char *) ctxt->input->cur, "]]>"); } else { term = xmlParseLookupString(ctxt, 0, "]]>", 3); } if (term == NULL) { int tmp, size; if (terminate) { /* Unfinished CDATA section */ size = ctxt->input->end - ctxt->input->cur; } else { if (avail < XML_PARSER_BIG_BUFFER_SIZE + 2) goto done; ctxt->checkIndex = 0; /* XXX: Why don't we pass the full buffer? */ size = XML_PARSER_BIG_BUFFER_SIZE; } tmp = xmlCheckCdataPush(ctxt->input->cur, size, 0); if (tmp <= 0) { tmp = -tmp; ctxt->input->cur += tmp; goto encoding_error; } if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { if (ctxt->sax->cdataBlock != NULL) ctxt->sax->cdataBlock(ctxt->userData, ctxt->input->cur, tmp); else if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, ctxt->input->cur, tmp); } if (ctxt->instate == XML_PARSER_EOF) goto done; SKIPL(tmp); } else { int base = term - CUR_PTR; int tmp; tmp = xmlCheckCdataPush(ctxt->input->cur, base, 1); if ((tmp < 0) || (tmp != base)) { tmp = -tmp; ctxt->input->cur += tmp; goto encoding_error; } if ((ctxt->sax != NULL) && (base == 0) && (ctxt->sax->cdataBlock != NULL) && (!ctxt->disableSAX)) { /* * Special case to provide identical behaviour * between pull and push parsers on enpty CDATA * sections */ if ((ctxt->input->cur - ctxt->input->base >= 9) && (!strncmp((const char *)&ctxt->input->cur[-9], "sax->cdataBlock(ctxt->userData, BAD_CAST "", 0); } else if ((ctxt->sax != NULL) && (base > 0) && (!ctxt->disableSAX)) { if (ctxt->sax->cdataBlock != NULL) ctxt->sax->cdataBlock(ctxt->userData, ctxt->input->cur, base); else if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, ctxt->input->cur, base); } if (ctxt->instate == XML_PARSER_EOF) goto done; SKIPL(base + 3); ctxt->instate = XML_PARSER_CONTENT; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering CONTENT\n"); #endif } break; } case XML_PARSER_MISC: case XML_PARSER_PROLOG: case XML_PARSER_EPILOG: SKIP_BLANKS; avail = ctxt->input->end - ctxt->input->cur; if (avail < 2) goto done; cur = ctxt->input->cur[0]; next = ctxt->input->cur[1]; if ((cur == '<') && (next == '?')) { if ((!terminate) && (!xmlParseLookupString(ctxt, 2, "?>", 2))) goto done; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: Parsing PI\n"); #endif xmlParsePI(ctxt); if (ctxt->instate == XML_PARSER_EOF) goto done; } else if ((cur == '<') && (next == '!') && (ctxt->input->cur[2] == '-') && (ctxt->input->cur[3] == '-')) { if ((!terminate) && (!xmlParseLookupString(ctxt, 4, "-->", 3))) goto done; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: Parsing Comment\n"); #endif xmlParseComment(ctxt); if (ctxt->instate == XML_PARSER_EOF) goto done; } else if ((ctxt->instate == XML_PARSER_MISC) && (cur == '<') && (next == '!') && (ctxt->input->cur[2] == 'D') && (ctxt->input->cur[3] == 'O') && (ctxt->input->cur[4] == 'C') && (ctxt->input->cur[5] == 'T') && (ctxt->input->cur[6] == 'Y') && (ctxt->input->cur[7] == 'P') && (ctxt->input->cur[8] == 'E')) { if ((!terminate) && (!xmlParseLookupGt(ctxt))) goto done; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: Parsing internal subset\n"); #endif ctxt->inSubset = 1; xmlParseDocTypeDecl(ctxt); if (ctxt->instate == XML_PARSER_EOF) goto done; if (RAW == '[') { ctxt->instate = XML_PARSER_DTD; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering DTD\n"); #endif } else { /* * Create and update the external subset. */ ctxt->inSubset = 2; if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && (ctxt->sax->externalSubset != NULL)) ctxt->sax->externalSubset(ctxt->userData, ctxt->intSubName, ctxt->extSubSystem, ctxt->extSubURI); ctxt->inSubset = 0; xmlCleanSpecialAttr(ctxt); ctxt->instate = XML_PARSER_PROLOG; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering PROLOG\n"); #endif } } else if ((cur == '<') && (next == '!') && (avail < (ctxt->instate == XML_PARSER_MISC ? 9 : 4))) { goto done; } else if (ctxt->instate == XML_PARSER_EPILOG) { xmlFatalErr(ctxt, XML_ERR_DOCUMENT_END, NULL); xmlHaltParser(ctxt); #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering EOF\n"); #endif if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) ctxt->sax->endDocument(ctxt->userData); goto done; } else { ctxt->instate = XML_PARSER_START_TAG; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering START_TAG\n"); #endif } break; case XML_PARSER_DTD: { if ((!terminate) && (!xmlParseLookupInternalSubset(ctxt))) goto done; xmlParseInternalSubset(ctxt); if (ctxt->instate == XML_PARSER_EOF) goto done; ctxt->inSubset = 2; if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && (ctxt->sax->externalSubset != NULL)) ctxt->sax->externalSubset(ctxt->userData, ctxt->intSubName, ctxt->extSubSystem, ctxt->extSubURI); ctxt->inSubset = 0; xmlCleanSpecialAttr(ctxt); if (ctxt->instate == XML_PARSER_EOF) goto done; ctxt->instate = XML_PARSER_PROLOG; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering PROLOG\n"); #endif break; } case XML_PARSER_COMMENT: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == COMMENT\n"); ctxt->instate = XML_PARSER_CONTENT; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering CONTENT\n"); #endif break; case XML_PARSER_IGNORE: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == IGNORE"); ctxt->instate = XML_PARSER_DTD; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering DTD\n"); #endif break; case XML_PARSER_PI: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == PI\n"); ctxt->instate = XML_PARSER_CONTENT; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering CONTENT\n"); #endif break; case XML_PARSER_ENTITY_DECL: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == ENTITY_DECL\n"); ctxt->instate = XML_PARSER_DTD; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering DTD\n"); #endif break; case XML_PARSER_ENTITY_VALUE: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == ENTITY_VALUE\n"); ctxt->instate = XML_PARSER_CONTENT; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering DTD\n"); #endif break; case XML_PARSER_ATTRIBUTE_VALUE: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == ATTRIBUTE_VALUE\n"); ctxt->instate = XML_PARSER_START_TAG; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering START_TAG\n"); #endif break; case XML_PARSER_SYSTEM_LITERAL: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == SYSTEM_LITERAL\n"); ctxt->instate = XML_PARSER_START_TAG; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering START_TAG\n"); #endif break; case XML_PARSER_PUBLIC_LITERAL: xmlGenericError(xmlGenericErrorContext, "PP: internal error, state == PUBLIC_LITERAL\n"); ctxt->instate = XML_PARSER_START_TAG; #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: entering START_TAG\n"); #endif break; } } done: #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: done %d\n", ret); #endif return(ret); encoding_error: if (ctxt->input->end - ctxt->input->cur < 4) { __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR, "Input is not proper UTF-8, indicate encoding !\n", NULL, NULL); } else { char buffer[150]; snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n", ctxt->input->cur[0], ctxt->input->cur[1], ctxt->input->cur[2], ctxt->input->cur[3]); __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR, "Input is not proper UTF-8, indicate encoding !\n%s", BAD_CAST buffer, NULL); } return(0); } /** * xmlParseChunk: * @ctxt: an XML parser context * @chunk: an char array * @size: the size in byte of the chunk * @terminate: last chunk indicator * * Parse a Chunk of memory * * Returns zero if no error, the xmlParserErrors otherwise. */ int xmlParseChunk(xmlParserCtxtPtr ctxt, const char *chunk, int size, int terminate) { int end_in_lf = 0; if (ctxt == NULL) return(XML_ERR_INTERNAL_ERROR); if ((ctxt->errNo != XML_ERR_OK) && (ctxt->disableSAX == 1)) return(ctxt->errNo); if (ctxt->instate == XML_PARSER_EOF) return(-1); if (ctxt->input == NULL) return(-1); ctxt->progressive = 1; if (ctxt->instate == XML_PARSER_START) xmlDetectSAX2(ctxt); if ((size > 0) && (chunk != NULL) && (!terminate) && (chunk[size - 1] == '\r')) { end_in_lf = 1; size--; } if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) && (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) { size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input); size_t cur = ctxt->input->cur - ctxt->input->base; int res; res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk); xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur); if (res < 0) { xmlFatalErr(ctxt, ctxt->input->buf->error, NULL); xmlHaltParser(ctxt); return(ctxt->errNo); } #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: pushed %d\n", size); #endif } else if (ctxt->instate != XML_PARSER_EOF) { if ((ctxt->input != NULL) && ctxt->input->buf != NULL) { xmlParserInputBufferPtr in = ctxt->input->buf; if ((in->encoder != NULL) && (in->buffer != NULL) && (in->raw != NULL)) { int nbchars; size_t base = xmlBufGetInputBase(in->buffer, ctxt->input); size_t current = ctxt->input->cur - ctxt->input->base; nbchars = xmlCharEncInput(in, terminate); xmlBufSetInputBaseCur(in->buffer, ctxt->input, base, current); if (nbchars < 0) { xmlFatalErr(ctxt, in->error, NULL); xmlHaltParser(ctxt); return(ctxt->errNo); } } } } xmlParseTryOrFinish(ctxt, terminate); if (ctxt->instate == XML_PARSER_EOF) return(ctxt->errNo); if ((ctxt->input != NULL) && (((ctxt->input->end - ctxt->input->cur) > XML_MAX_LOOKUP_LIMIT) || ((ctxt->input->cur - ctxt->input->base) > XML_MAX_LOOKUP_LIMIT)) && ((ctxt->options & XML_PARSE_HUGE) == 0)) { xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "Huge input lookup"); xmlHaltParser(ctxt); } if ((ctxt->errNo != XML_ERR_OK) && (ctxt->disableSAX == 1)) return(ctxt->errNo); if ((end_in_lf == 1) && (ctxt->input != NULL) && (ctxt->input->buf != NULL)) { size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input); size_t current = ctxt->input->cur - ctxt->input->base; int res; res = xmlParserInputBufferPush(ctxt->input->buf, 1, "\r"); xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, current); if (res < 0) { xmlFatalErr(ctxt, ctxt->input->buf->error, NULL); xmlHaltParser(ctxt); return(ctxt->errNo); } } if (terminate) { /* * Check for termination */ if ((ctxt->instate != XML_PARSER_EOF) && (ctxt->instate != XML_PARSER_EPILOG)) { xmlFatalErr(ctxt, XML_ERR_DOCUMENT_END, NULL); } if ((ctxt->instate == XML_PARSER_EPILOG) && (ctxt->input->cur < ctxt->input->end)) { xmlFatalErr(ctxt, XML_ERR_DOCUMENT_END, NULL); } if (ctxt->instate != XML_PARSER_EOF) { if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) ctxt->sax->endDocument(ctxt->userData); } ctxt->instate = XML_PARSER_EOF; } if (ctxt->wellFormed == 0) return((xmlParserErrors) ctxt->errNo); else return(0); } /************************************************************************ * * * I/O front end functions to the parser * * * ************************************************************************/ /** * xmlCreatePushParserCtxt: * @sax: a SAX handler * @user_data: The user data returned on SAX callbacks * @chunk: a pointer to an array of chars * @size: number of chars in the array * @filename: an optional file name or URI * * Create a parser context for using the XML parser in push mode. * If @buffer and @size are non-NULL, the data is used to detect * the encoding. The remaining characters will be parsed so they * don't need to be fed in again through xmlParseChunk. * To allow content encoding detection, @size should be >= 4 * The value of @filename is used for fetching external entities * and error/warning reports. * * Returns the new parser context or NULL */ xmlParserCtxtPtr xmlCreatePushParserCtxt(xmlSAXHandlerPtr sax, void *user_data, const char *chunk, int size, const char *filename) { xmlParserCtxtPtr ctxt; xmlParserInputPtr inputStream; xmlParserInputBufferPtr buf; buf = xmlAllocParserInputBuffer(XML_CHAR_ENCODING_NONE); if (buf == NULL) return(NULL); ctxt = xmlNewSAXParserCtxt(sax, user_data); if (ctxt == NULL) { xmlErrMemory(NULL, "creating parser: out of memory\n"); xmlFreeParserInputBuffer(buf); return(NULL); } ctxt->dictNames = 1; if (filename == NULL) { ctxt->directory = NULL; } else { ctxt->directory = xmlParserGetDirectory(filename); } inputStream = xmlNewInputStream(ctxt); if (inputStream == NULL) { xmlFreeParserCtxt(ctxt); xmlFreeParserInputBuffer(buf); return(NULL); } if (filename == NULL) inputStream->filename = NULL; else { inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) filename); if (inputStream->filename == NULL) { xmlFreeInputStream(inputStream); xmlFreeParserCtxt(ctxt); xmlFreeParserInputBuffer(buf); return(NULL); } } inputStream->buf = buf; xmlBufResetInput(inputStream->buf->buffer, inputStream); inputPush(ctxt, inputStream); /* * If the caller didn't provide an initial 'chunk' for determining * the encoding, we set the context to XML_CHAR_ENCODING_NONE so * that it can be automatically determined later */ ctxt->charset = XML_CHAR_ENCODING_NONE; if ((size != 0) && (chunk != NULL) && (ctxt->input != NULL) && (ctxt->input->buf != NULL)) { size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input); size_t cur = ctxt->input->cur - ctxt->input->base; int res; res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk); xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur); if (res < 0) { xmlFatalErr(ctxt, ctxt->input->buf->error, NULL); xmlHaltParser(ctxt); } #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: pushed %d\n", size); #endif } return(ctxt); } #endif /* LIBXML_PUSH_ENABLED */ /** * xmlStopParser: * @ctxt: an XML parser context * * Blocks further parser processing */ void xmlStopParser(xmlParserCtxtPtr ctxt) { if (ctxt == NULL) return; xmlHaltParser(ctxt); ctxt->errNo = XML_ERR_USER_STOP; } /** * xmlCreateIOParserCtxt: * @sax: a SAX handler * @user_data: The user data returned on SAX callbacks * @ioread: an I/O read function * @ioclose: an I/O close function * @ioctx: an I/O handler * @enc: the charset encoding if known * * Create a parser context for using the XML parser with an existing * I/O stream * * Returns the new parser context or NULL */ xmlParserCtxtPtr xmlCreateIOParserCtxt(xmlSAXHandlerPtr sax, void *user_data, xmlInputReadCallback ioread, xmlInputCloseCallback ioclose, void *ioctx, xmlCharEncoding enc) { xmlParserCtxtPtr ctxt; xmlParserInputPtr inputStream; xmlParserInputBufferPtr buf; if (ioread == NULL) return(NULL); buf = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx, enc); if (buf == NULL) { if (ioclose != NULL) ioclose(ioctx); return (NULL); } ctxt = xmlNewSAXParserCtxt(sax, user_data); if (ctxt == NULL) { xmlFreeParserInputBuffer(buf); return(NULL); } inputStream = xmlNewIOInputStream(ctxt, buf, enc); if (inputStream == NULL) { xmlFreeParserCtxt(ctxt); return(NULL); } inputPush(ctxt, inputStream); return(ctxt); } #ifdef LIBXML_VALID_ENABLED /************************************************************************ * * * Front ends when parsing a DTD * * * ************************************************************************/ /** * xmlIOParseDTD: * @sax: the SAX handler block or NULL * @input: an Input Buffer * @enc: the charset encoding if known * * Load and parse a DTD * * Returns the resulting xmlDtdPtr or NULL in case of error. * @input will be freed by the function in any case. */ xmlDtdPtr xmlIOParseDTD(xmlSAXHandlerPtr sax, xmlParserInputBufferPtr input, xmlCharEncoding enc) { xmlDtdPtr ret = NULL; xmlParserCtxtPtr ctxt; xmlParserInputPtr pinput = NULL; xmlChar start[4]; if (input == NULL) return(NULL); ctxt = xmlNewSAXParserCtxt(sax, NULL); if (ctxt == NULL) { xmlFreeParserInputBuffer(input); return(NULL); } /* We are loading a DTD */ ctxt->options |= XML_PARSE_DTDLOAD; xmlDetectSAX2(ctxt); /* * generate a parser input from the I/O handler */ pinput = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); if (pinput == NULL) { xmlFreeParserInputBuffer(input); xmlFreeParserCtxt(ctxt); return(NULL); } /* * plug some encoding conversion routines here. */ if (xmlPushInput(ctxt, pinput) < 0) { xmlFreeParserCtxt(ctxt); return(NULL); } if (enc != XML_CHAR_ENCODING_NONE) { xmlSwitchEncoding(ctxt, enc); } pinput->filename = NULL; pinput->line = 1; pinput->col = 1; pinput->base = ctxt->input->cur; pinput->cur = ctxt->input->cur; pinput->free = NULL; /* * let's parse that entity knowing it's an external subset. */ ctxt->inSubset = 2; ctxt->myDoc = xmlNewDoc(BAD_CAST "1.0"); if (ctxt->myDoc == NULL) { xmlErrMemory(ctxt, "New Doc failed"); return(NULL); } ctxt->myDoc->properties = XML_DOC_INTERNAL; ctxt->myDoc->extSubset = xmlNewDtd(ctxt->myDoc, BAD_CAST "none", BAD_CAST "none", BAD_CAST "none"); if ((enc == XML_CHAR_ENCODING_NONE) && ((ctxt->input->end - ctxt->input->cur) >= 4)) { /* * Get the 4 first bytes and decode the charset * if enc != XML_CHAR_ENCODING_NONE * plug some encoding conversion routines. */ start[0] = RAW; start[1] = NXT(1); start[2] = NXT(2); start[3] = NXT(3); enc = xmlDetectCharEncoding(start, 4); if (enc != XML_CHAR_ENCODING_NONE) { xmlSwitchEncoding(ctxt, enc); } } xmlParseExternalSubset(ctxt, BAD_CAST "none", BAD_CAST "none"); if (ctxt->myDoc != NULL) { if (ctxt->wellFormed) { ret = ctxt->myDoc->extSubset; ctxt->myDoc->extSubset = NULL; if (ret != NULL) { xmlNodePtr tmp; ret->doc = NULL; tmp = ret->children; while (tmp != NULL) { tmp->doc = NULL; tmp = tmp->next; } } } else { ret = NULL; } xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } xmlFreeParserCtxt(ctxt); return(ret); } /** * xmlSAXParseDTD: * @sax: the SAX handler block * @ExternalID: a NAME* containing the External ID of the DTD * @SystemID: a NAME* containing the URL to the DTD * * DEPRECATED: Don't use. * * Load and parse an external subset. * * Returns the resulting xmlDtdPtr or NULL in case of error. */ xmlDtdPtr xmlSAXParseDTD(xmlSAXHandlerPtr sax, const xmlChar *ExternalID, const xmlChar *SystemID) { xmlDtdPtr ret = NULL; xmlParserCtxtPtr ctxt; xmlParserInputPtr input = NULL; xmlCharEncoding enc; xmlChar* systemIdCanonic; if ((ExternalID == NULL) && (SystemID == NULL)) return(NULL); ctxt = xmlNewSAXParserCtxt(sax, NULL); if (ctxt == NULL) { return(NULL); } /* We are loading a DTD */ ctxt->options |= XML_PARSE_DTDLOAD; /* * Canonicalise the system ID */ systemIdCanonic = xmlCanonicPath(SystemID); if ((SystemID != NULL) && (systemIdCanonic == NULL)) { xmlFreeParserCtxt(ctxt); return(NULL); } /* * Ask the Entity resolver to load the damn thing */ if ((ctxt->sax != NULL) && (ctxt->sax->resolveEntity != NULL)) input = ctxt->sax->resolveEntity(ctxt->userData, ExternalID, systemIdCanonic); if (input == NULL) { xmlFreeParserCtxt(ctxt); if (systemIdCanonic != NULL) xmlFree(systemIdCanonic); return(NULL); } /* * plug some encoding conversion routines here. */ if (xmlPushInput(ctxt, input) < 0) { xmlFreeParserCtxt(ctxt); if (systemIdCanonic != NULL) xmlFree(systemIdCanonic); return(NULL); } if ((ctxt->input->end - ctxt->input->cur) >= 4) { enc = xmlDetectCharEncoding(ctxt->input->cur, 4); xmlSwitchEncoding(ctxt, enc); } if (input->filename == NULL) input->filename = (char *) systemIdCanonic; else xmlFree(systemIdCanonic); input->line = 1; input->col = 1; input->base = ctxt->input->cur; input->cur = ctxt->input->cur; input->free = NULL; /* * let's parse that entity knowing it's an external subset. */ ctxt->inSubset = 2; ctxt->myDoc = xmlNewDoc(BAD_CAST "1.0"); if (ctxt->myDoc == NULL) { xmlErrMemory(ctxt, "New Doc failed"); xmlFreeParserCtxt(ctxt); return(NULL); } ctxt->myDoc->properties = XML_DOC_INTERNAL; ctxt->myDoc->extSubset = xmlNewDtd(ctxt->myDoc, BAD_CAST "none", ExternalID, SystemID); xmlParseExternalSubset(ctxt, ExternalID, SystemID); if (ctxt->myDoc != NULL) { if (ctxt->wellFormed) { ret = ctxt->myDoc->extSubset; ctxt->myDoc->extSubset = NULL; if (ret != NULL) { xmlNodePtr tmp; ret->doc = NULL; tmp = ret->children; while (tmp != NULL) { tmp->doc = NULL; tmp = tmp->next; } } } else { ret = NULL; } xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } xmlFreeParserCtxt(ctxt); return(ret); } /** * xmlParseDTD: * @ExternalID: a NAME* containing the External ID of the DTD * @SystemID: a NAME* containing the URL to the DTD * * Load and parse an external subset. * * Returns the resulting xmlDtdPtr or NULL in case of error. */ xmlDtdPtr xmlParseDTD(const xmlChar *ExternalID, const xmlChar *SystemID) { return(xmlSAXParseDTD(NULL, ExternalID, SystemID)); } #endif /* LIBXML_VALID_ENABLED */ /************************************************************************ * * * Front ends when parsing an Entity * * * ************************************************************************/ /** * xmlParseCtxtExternalEntity: * @ctx: the existing parsing context * @URL: the URL for the entity to load * @ID: the System ID for the entity to load * @lst: the return value for the set of parsed nodes * * Parse an external general entity within an existing parsing context * An external general parsed entity is well-formed if it matches the * production labeled extParsedEnt. * * [78] extParsedEnt ::= TextDecl? content * * Returns 0 if the entity is well formed, -1 in case of args problem and * the parser error code otherwise */ int xmlParseCtxtExternalEntity(xmlParserCtxtPtr ctx, const xmlChar *URL, const xmlChar *ID, xmlNodePtr *lst) { void *userData; if (ctx == NULL) return(-1); /* * If the user provided their own SAX callbacks, then reuse the * userData callback field, otherwise the expected setup in a * DOM builder is to have userData == ctxt */ if (ctx->userData == ctx) userData = NULL; else userData = ctx->userData; return xmlParseExternalEntityPrivate(ctx->myDoc, ctx, ctx->sax, userData, ctx->depth + 1, URL, ID, lst); } /** * xmlParseExternalEntityPrivate: * @doc: the document the chunk pertains to * @oldctxt: the previous parser context if available * @sax: the SAX handler block (possibly NULL) * @user_data: The user data returned on SAX callbacks (possibly NULL) * @depth: Used for loop detection, use 0 * @URL: the URL for the entity to load * @ID: the System ID for the entity to load * @list: the return value for the set of parsed nodes * * Private version of xmlParseExternalEntity() * * Returns 0 if the entity is well formed, -1 in case of args problem and * the parser error code otherwise */ static xmlParserErrors xmlParseExternalEntityPrivate(xmlDocPtr doc, xmlParserCtxtPtr oldctxt, xmlSAXHandlerPtr sax, void *user_data, int depth, const xmlChar *URL, const xmlChar *ID, xmlNodePtr *list) { xmlParserCtxtPtr ctxt; xmlDocPtr newDoc; xmlNodePtr newRoot; xmlParserErrors ret = XML_ERR_OK; xmlChar start[4]; xmlCharEncoding enc; if (((depth > 40) && ((oldctxt == NULL) || (oldctxt->options & XML_PARSE_HUGE) == 0)) || (depth > 100)) { xmlFatalErrMsg(oldctxt, XML_ERR_ENTITY_LOOP, "Maximum entity nesting depth exceeded"); return(XML_ERR_ENTITY_LOOP); } if (list != NULL) *list = NULL; if ((URL == NULL) && (ID == NULL)) return(XML_ERR_INTERNAL_ERROR); if (doc == NULL) return(XML_ERR_INTERNAL_ERROR); ctxt = xmlCreateEntityParserCtxtInternal(sax, user_data, URL, ID, NULL, oldctxt); if (ctxt == NULL) return(XML_WAR_UNDECLARED_ENTITY); if (oldctxt != NULL) { ctxt->nbErrors = oldctxt->nbErrors; ctxt->nbWarnings = oldctxt->nbWarnings; } xmlDetectSAX2(ctxt); newDoc = xmlNewDoc(BAD_CAST "1.0"); if (newDoc == NULL) { xmlFreeParserCtxt(ctxt); return(XML_ERR_INTERNAL_ERROR); } newDoc->properties = XML_DOC_INTERNAL; if (doc) { newDoc->intSubset = doc->intSubset; newDoc->extSubset = doc->extSubset; if (doc->dict) { newDoc->dict = doc->dict; xmlDictReference(newDoc->dict); } if (doc->URL != NULL) { newDoc->URL = xmlStrdup(doc->URL); } } newRoot = xmlNewDocNode(newDoc, NULL, BAD_CAST "pseudoroot", NULL); if (newRoot == NULL) { if (sax != NULL) xmlFreeParserCtxt(ctxt); newDoc->intSubset = NULL; newDoc->extSubset = NULL; xmlFreeDoc(newDoc); return(XML_ERR_INTERNAL_ERROR); } xmlAddChild((xmlNodePtr) newDoc, newRoot); nodePush(ctxt, newDoc->children); if (doc == NULL) { ctxt->myDoc = newDoc; } else { ctxt->myDoc = doc; newRoot->doc = doc; } /* * Get the 4 first bytes and decode the charset * if enc != XML_CHAR_ENCODING_NONE * plug some encoding conversion routines. */ GROW; if ((ctxt->input->end - ctxt->input->cur) >= 4) { start[0] = RAW; start[1] = NXT(1); start[2] = NXT(2); start[3] = NXT(3); enc = xmlDetectCharEncoding(start, 4); if (enc != XML_CHAR_ENCODING_NONE) { xmlSwitchEncoding(ctxt, enc); } } /* * Parse a possible text declaration first */ if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { xmlParseTextDecl(ctxt); /* * An XML-1.0 document can't reference an entity not XML-1.0 */ if ((xmlStrEqual(oldctxt->version, BAD_CAST "1.0")) && (!xmlStrEqual(ctxt->input->version, BAD_CAST "1.0"))) { xmlFatalErrMsg(ctxt, XML_ERR_VERSION_MISMATCH, "Version mismatch between document and entity\n"); } } ctxt->instate = XML_PARSER_CONTENT; ctxt->depth = depth; if (oldctxt != NULL) { ctxt->_private = oldctxt->_private; ctxt->loadsubset = oldctxt->loadsubset; ctxt->validate = oldctxt->validate; ctxt->valid = oldctxt->valid; ctxt->replaceEntities = oldctxt->replaceEntities; if (oldctxt->validate) { ctxt->vctxt.error = oldctxt->vctxt.error; ctxt->vctxt.warning = oldctxt->vctxt.warning; ctxt->vctxt.userData = oldctxt->vctxt.userData; ctxt->vctxt.flags = oldctxt->vctxt.flags; } ctxt->external = oldctxt->external; if (ctxt->dict) xmlDictFree(ctxt->dict); ctxt->dict = oldctxt->dict; ctxt->str_xml = xmlDictLookup(ctxt->dict, BAD_CAST "xml", 3); ctxt->str_xmlns = xmlDictLookup(ctxt->dict, BAD_CAST "xmlns", 5); ctxt->str_xml_ns = xmlDictLookup(ctxt->dict, XML_XML_NAMESPACE, 36); ctxt->dictNames = oldctxt->dictNames; ctxt->attsDefault = oldctxt->attsDefault; ctxt->attsSpecial = oldctxt->attsSpecial; ctxt->linenumbers = oldctxt->linenumbers; ctxt->record_info = oldctxt->record_info; ctxt->node_seq.maximum = oldctxt->node_seq.maximum; ctxt->node_seq.length = oldctxt->node_seq.length; ctxt->node_seq.buffer = oldctxt->node_seq.buffer; } else { /* * Doing validity checking on chunk without context * doesn't make sense */ ctxt->_private = NULL; ctxt->validate = 0; ctxt->external = 2; ctxt->loadsubset = 0; } xmlParseContent(ctxt); if ((RAW == '<') && (NXT(1) == '/')) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); } else if (RAW != 0) { xmlFatalErr(ctxt, XML_ERR_EXTRA_CONTENT, NULL); } if (ctxt->node != newDoc->children) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); } if (!ctxt->wellFormed) { ret = (xmlParserErrors)ctxt->errNo; if (oldctxt != NULL) { oldctxt->errNo = ctxt->errNo; oldctxt->wellFormed = 0; xmlCopyError(&ctxt->lastError, &oldctxt->lastError); } } else { if (list != NULL) { xmlNodePtr cur; /* * Return the newly created nodeset after unlinking it from * they pseudo parent. */ cur = newDoc->children->children; *list = cur; while (cur != NULL) { cur->parent = NULL; cur = cur->next; } newDoc->children->children = NULL; } ret = XML_ERR_OK; } /* * Also record the size of the entity parsed */ if (ctxt->input != NULL && oldctxt != NULL) { unsigned long consumed = ctxt->input->consumed; xmlSaturatedAddSizeT(&consumed, ctxt->input->cur - ctxt->input->base); xmlSaturatedAdd(&oldctxt->sizeentities, consumed); xmlSaturatedAdd(&oldctxt->sizeentities, ctxt->sizeentities); xmlSaturatedAdd(&oldctxt->sizeentcopy, consumed); xmlSaturatedAdd(&oldctxt->sizeentcopy, ctxt->sizeentcopy); } if (oldctxt != NULL) { ctxt->dict = NULL; ctxt->attsDefault = NULL; ctxt->attsSpecial = NULL; oldctxt->nbErrors = ctxt->nbErrors; oldctxt->nbWarnings = ctxt->nbWarnings; oldctxt->validate = ctxt->validate; oldctxt->valid = ctxt->valid; oldctxt->node_seq.maximum = ctxt->node_seq.maximum; oldctxt->node_seq.length = ctxt->node_seq.length; oldctxt->node_seq.buffer = ctxt->node_seq.buffer; } ctxt->node_seq.maximum = 0; ctxt->node_seq.length = 0; ctxt->node_seq.buffer = NULL; xmlFreeParserCtxt(ctxt); newDoc->intSubset = NULL; newDoc->extSubset = NULL; xmlFreeDoc(newDoc); return(ret); } #ifdef LIBXML_SAX1_ENABLED /** * xmlParseExternalEntity: * @doc: the document the chunk pertains to * @sax: the SAX handler block (possibly NULL) * @user_data: The user data returned on SAX callbacks (possibly NULL) * @depth: Used for loop detection, use 0 * @URL: the URL for the entity to load * @ID: the System ID for the entity to load * @lst: the return value for the set of parsed nodes * * Parse an external general entity * An external general parsed entity is well-formed if it matches the * production labeled extParsedEnt. * * [78] extParsedEnt ::= TextDecl? content * * Returns 0 if the entity is well formed, -1 in case of args problem and * the parser error code otherwise */ int xmlParseExternalEntity(xmlDocPtr doc, xmlSAXHandlerPtr sax, void *user_data, int depth, const xmlChar *URL, const xmlChar *ID, xmlNodePtr *lst) { return(xmlParseExternalEntityPrivate(doc, NULL, sax, user_data, depth, URL, ID, lst)); } /** * xmlParseBalancedChunkMemory: * @doc: the document the chunk pertains to (must not be NULL) * @sax: the SAX handler block (possibly NULL) * @user_data: The user data returned on SAX callbacks (possibly NULL) * @depth: Used for loop detection, use 0 * @string: the input string in UTF8 or ISO-Latin (zero terminated) * @lst: the return value for the set of parsed nodes * * Parse a well-balanced chunk of an XML document * called by the parser * The allowed sequence for the Well Balanced Chunk is the one defined by * the content production in the XML grammar: * * [43] content ::= (element | CharData | Reference | CDSect | PI | Comment)* * * Returns 0 if the chunk is well balanced, -1 in case of args problem and * the parser error code otherwise */ int xmlParseBalancedChunkMemory(xmlDocPtr doc, xmlSAXHandlerPtr sax, void *user_data, int depth, const xmlChar *string, xmlNodePtr *lst) { return xmlParseBalancedChunkMemoryRecover( doc, sax, user_data, depth, string, lst, 0 ); } #endif /* LIBXML_SAX1_ENABLED */ /** * xmlParseBalancedChunkMemoryInternal: * @oldctxt: the existing parsing context * @string: the input string in UTF8 or ISO-Latin (zero terminated) * @user_data: the user data field for the parser context * @lst: the return value for the set of parsed nodes * * * Parse a well-balanced chunk of an XML document * called by the parser * The allowed sequence for the Well Balanced Chunk is the one defined by * the content production in the XML grammar: * * [43] content ::= (element | CharData | Reference | CDSect | PI | Comment)* * * Returns XML_ERR_OK if the chunk is well balanced, and the parser * error code otherwise * * In case recover is set to 1, the nodelist will not be empty even if * the parsed chunk is not well balanced. */ static xmlParserErrors xmlParseBalancedChunkMemoryInternal(xmlParserCtxtPtr oldctxt, const xmlChar *string, void *user_data, xmlNodePtr *lst) { xmlParserCtxtPtr ctxt; xmlDocPtr newDoc = NULL; xmlNodePtr newRoot; xmlSAXHandlerPtr oldsax = NULL; xmlNodePtr content = NULL; xmlNodePtr last = NULL; int size; xmlParserErrors ret = XML_ERR_OK; #ifdef SAX2 int i; #endif if (((oldctxt->depth > 40) && ((oldctxt->options & XML_PARSE_HUGE) == 0)) || (oldctxt->depth > 100)) { xmlFatalErrMsg(oldctxt, XML_ERR_ENTITY_LOOP, "Maximum entity nesting depth exceeded"); return(XML_ERR_ENTITY_LOOP); } if (lst != NULL) *lst = NULL; if (string == NULL) return(XML_ERR_INTERNAL_ERROR); size = xmlStrlen(string); ctxt = xmlCreateMemoryParserCtxt((char *) string, size); if (ctxt == NULL) return(XML_WAR_UNDECLARED_ENTITY); ctxt->nbErrors = oldctxt->nbErrors; ctxt->nbWarnings = oldctxt->nbWarnings; if (user_data != NULL) ctxt->userData = user_data; else ctxt->userData = ctxt; if (ctxt->dict != NULL) xmlDictFree(ctxt->dict); ctxt->dict = oldctxt->dict; ctxt->input_id = oldctxt->input_id; ctxt->str_xml = xmlDictLookup(ctxt->dict, BAD_CAST "xml", 3); ctxt->str_xmlns = xmlDictLookup(ctxt->dict, BAD_CAST "xmlns", 5); ctxt->str_xml_ns = xmlDictLookup(ctxt->dict, XML_XML_NAMESPACE, 36); #ifdef SAX2 /* propagate namespaces down the entity */ for (i = 0;i < oldctxt->nsNr;i += 2) { nsPush(ctxt, oldctxt->nsTab[i], oldctxt->nsTab[i+1]); } #endif oldsax = ctxt->sax; ctxt->sax = oldctxt->sax; xmlDetectSAX2(ctxt); ctxt->replaceEntities = oldctxt->replaceEntities; ctxt->options = oldctxt->options; ctxt->_private = oldctxt->_private; if (oldctxt->myDoc == NULL) { newDoc = xmlNewDoc(BAD_CAST "1.0"); if (newDoc == NULL) { ctxt->sax = oldsax; ctxt->dict = NULL; xmlFreeParserCtxt(ctxt); return(XML_ERR_INTERNAL_ERROR); } newDoc->properties = XML_DOC_INTERNAL; newDoc->dict = ctxt->dict; xmlDictReference(newDoc->dict); ctxt->myDoc = newDoc; } else { ctxt->myDoc = oldctxt->myDoc; content = ctxt->myDoc->children; last = ctxt->myDoc->last; } newRoot = xmlNewDocNode(ctxt->myDoc, NULL, BAD_CAST "pseudoroot", NULL); if (newRoot == NULL) { ctxt->sax = oldsax; ctxt->dict = NULL; xmlFreeParserCtxt(ctxt); if (newDoc != NULL) { xmlFreeDoc(newDoc); } return(XML_ERR_INTERNAL_ERROR); } ctxt->myDoc->children = NULL; ctxt->myDoc->last = NULL; xmlAddChild((xmlNodePtr) ctxt->myDoc, newRoot); nodePush(ctxt, ctxt->myDoc->children); ctxt->instate = XML_PARSER_CONTENT; ctxt->depth = oldctxt->depth; ctxt->validate = 0; ctxt->loadsubset = oldctxt->loadsubset; if ((oldctxt->validate) || (oldctxt->replaceEntities != 0)) { /* * ID/IDREF registration will be done in xmlValidateElement below */ ctxt->loadsubset |= XML_SKIP_IDS; } ctxt->dictNames = oldctxt->dictNames; ctxt->attsDefault = oldctxt->attsDefault; ctxt->attsSpecial = oldctxt->attsSpecial; xmlParseContent(ctxt); if ((RAW == '<') && (NXT(1) == '/')) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); } else if (RAW != 0) { xmlFatalErr(ctxt, XML_ERR_EXTRA_CONTENT, NULL); } if (ctxt->node != ctxt->myDoc->children) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); } if (!ctxt->wellFormed) { ret = (xmlParserErrors)ctxt->errNo; oldctxt->errNo = ctxt->errNo; oldctxt->wellFormed = 0; xmlCopyError(&ctxt->lastError, &oldctxt->lastError); } else { ret = XML_ERR_OK; } if ((lst != NULL) && (ret == XML_ERR_OK)) { xmlNodePtr cur; /* * Return the newly created nodeset after unlinking it from * they pseudo parent. */ cur = ctxt->myDoc->children->children; *lst = cur; while (cur != NULL) { #ifdef LIBXML_VALID_ENABLED if ((oldctxt->validate) && (oldctxt->wellFormed) && (oldctxt->myDoc) && (oldctxt->myDoc->intSubset) && (cur->type == XML_ELEMENT_NODE)) { oldctxt->valid &= xmlValidateElement(&oldctxt->vctxt, oldctxt->myDoc, cur); } #endif /* LIBXML_VALID_ENABLED */ cur->parent = NULL; cur = cur->next; } ctxt->myDoc->children->children = NULL; } if (ctxt->myDoc != NULL) { xmlFreeNode(ctxt->myDoc->children); ctxt->myDoc->children = content; ctxt->myDoc->last = last; } /* * Also record the size of the entity parsed */ if (ctxt->input != NULL && oldctxt != NULL) { unsigned long consumed = ctxt->input->consumed; xmlSaturatedAddSizeT(&consumed, ctxt->input->cur - ctxt->input->base); xmlSaturatedAdd(&oldctxt->sizeentcopy, consumed); xmlSaturatedAdd(&oldctxt->sizeentcopy, ctxt->sizeentcopy); } oldctxt->nbErrors = ctxt->nbErrors; oldctxt->nbWarnings = ctxt->nbWarnings; ctxt->sax = oldsax; ctxt->dict = NULL; ctxt->attsDefault = NULL; ctxt->attsSpecial = NULL; xmlFreeParserCtxt(ctxt); if (newDoc != NULL) { xmlFreeDoc(newDoc); } return(ret); } /** * xmlParseInNodeContext: * @node: the context node * @data: the input string * @datalen: the input string length in bytes * @options: a combination of xmlParserOption * @lst: the return value for the set of parsed nodes * * Parse a well-balanced chunk of an XML document * within the context (DTD, namespaces, etc ...) of the given node. * * The allowed sequence for the data is a Well Balanced Chunk defined by * the content production in the XML grammar: * * [43] content ::= (element | CharData | Reference | CDSect | PI | Comment)* * * Returns XML_ERR_OK if the chunk is well balanced, and the parser * error code otherwise */ xmlParserErrors xmlParseInNodeContext(xmlNodePtr node, const char *data, int datalen, int options, xmlNodePtr *lst) { #ifdef SAX2 xmlParserCtxtPtr ctxt; xmlDocPtr doc = NULL; xmlNodePtr fake, cur; int nsnr = 0; xmlParserErrors ret = XML_ERR_OK; /* * check all input parameters, grab the document */ if ((lst == NULL) || (node == NULL) || (data == NULL) || (datalen < 0)) return(XML_ERR_INTERNAL_ERROR); switch (node->type) { case XML_ELEMENT_NODE: case XML_ATTRIBUTE_NODE: case XML_TEXT_NODE: case XML_CDATA_SECTION_NODE: case XML_ENTITY_REF_NODE: case XML_PI_NODE: case XML_COMMENT_NODE: case XML_DOCUMENT_NODE: case XML_HTML_DOCUMENT_NODE: break; default: return(XML_ERR_INTERNAL_ERROR); } while ((node != NULL) && (node->type != XML_ELEMENT_NODE) && (node->type != XML_DOCUMENT_NODE) && (node->type != XML_HTML_DOCUMENT_NODE)) node = node->parent; if (node == NULL) return(XML_ERR_INTERNAL_ERROR); if (node->type == XML_ELEMENT_NODE) doc = node->doc; else doc = (xmlDocPtr) node; if (doc == NULL) return(XML_ERR_INTERNAL_ERROR); /* * allocate a context and set-up everything not related to the * node position in the tree */ if (doc->type == XML_DOCUMENT_NODE) ctxt = xmlCreateMemoryParserCtxt((char *) data, datalen); #ifdef LIBXML_HTML_ENABLED else if (doc->type == XML_HTML_DOCUMENT_NODE) { ctxt = htmlCreateMemoryParserCtxt((char *) data, datalen); /* * When parsing in context, it makes no sense to add implied * elements like html/body/etc... */ options |= HTML_PARSE_NOIMPLIED; } #endif else return(XML_ERR_INTERNAL_ERROR); if (ctxt == NULL) return(XML_ERR_NO_MEMORY); /* * Use input doc's dict if present, else assure XML_PARSE_NODICT is set. * We need a dictionary for xmlDetectSAX2, so if there's no doc dict * we must wait until the last moment to free the original one. */ if (doc->dict != NULL) { if (ctxt->dict != NULL) xmlDictFree(ctxt->dict); ctxt->dict = doc->dict; } else options |= XML_PARSE_NODICT; if (doc->encoding != NULL) { xmlCharEncodingHandlerPtr hdlr; if (ctxt->encoding != NULL) xmlFree((xmlChar *) ctxt->encoding); ctxt->encoding = xmlStrdup((const xmlChar *) doc->encoding); hdlr = xmlFindCharEncodingHandler((const char *) doc->encoding); if (hdlr != NULL) { xmlSwitchToEncoding(ctxt, hdlr); } else { return(XML_ERR_UNSUPPORTED_ENCODING); } } xmlCtxtUseOptionsInternal(ctxt, options, NULL); xmlDetectSAX2(ctxt); ctxt->myDoc = doc; /* parsing in context, i.e. as within existing content */ ctxt->input_id = 2; ctxt->instate = XML_PARSER_CONTENT; fake = xmlNewDocComment(node->doc, NULL); if (fake == NULL) { xmlFreeParserCtxt(ctxt); return(XML_ERR_NO_MEMORY); } xmlAddChild(node, fake); if (node->type == XML_ELEMENT_NODE) { nodePush(ctxt, node); /* * initialize the SAX2 namespaces stack */ cur = node; while ((cur != NULL) && (cur->type == XML_ELEMENT_NODE)) { xmlNsPtr ns = cur->nsDef; const xmlChar *iprefix, *ihref; while (ns != NULL) { if (ctxt->dict) { iprefix = xmlDictLookup(ctxt->dict, ns->prefix, -1); ihref = xmlDictLookup(ctxt->dict, ns->href, -1); } else { iprefix = ns->prefix; ihref = ns->href; } if (xmlGetNamespace(ctxt, iprefix) == NULL) { nsPush(ctxt, iprefix, ihref); nsnr++; } ns = ns->next; } cur = cur->parent; } } if ((ctxt->validate) || (ctxt->replaceEntities != 0)) { /* * ID/IDREF registration will be done in xmlValidateElement below */ ctxt->loadsubset |= XML_SKIP_IDS; } #ifdef LIBXML_HTML_ENABLED if (doc->type == XML_HTML_DOCUMENT_NODE) __htmlParseContent(ctxt); else #endif xmlParseContent(ctxt); nsPop(ctxt, nsnr); if ((RAW == '<') && (NXT(1) == '/')) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); } else if (RAW != 0) { xmlFatalErr(ctxt, XML_ERR_EXTRA_CONTENT, NULL); } if ((ctxt->node != NULL) && (ctxt->node != node)) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); ctxt->wellFormed = 0; } if (!ctxt->wellFormed) { if (ctxt->errNo == 0) ret = XML_ERR_INTERNAL_ERROR; else ret = (xmlParserErrors)ctxt->errNo; } else { ret = XML_ERR_OK; } /* * Return the newly created nodeset after unlinking it from * the pseudo sibling. */ cur = fake->next; fake->next = NULL; node->last = fake; if (cur != NULL) { cur->prev = NULL; } *lst = cur; while (cur != NULL) { cur->parent = NULL; cur = cur->next; } xmlUnlinkNode(fake); xmlFreeNode(fake); if (ret != XML_ERR_OK) { xmlFreeNodeList(*lst); *lst = NULL; } if (doc->dict != NULL) ctxt->dict = NULL; xmlFreeParserCtxt(ctxt); return(ret); #else /* !SAX2 */ return(XML_ERR_INTERNAL_ERROR); #endif } #ifdef LIBXML_SAX1_ENABLED /** * xmlParseBalancedChunkMemoryRecover: * @doc: the document the chunk pertains to (must not be NULL) * @sax: the SAX handler block (possibly NULL) * @user_data: The user data returned on SAX callbacks (possibly NULL) * @depth: Used for loop detection, use 0 * @string: the input string in UTF8 or ISO-Latin (zero terminated) * @lst: the return value for the set of parsed nodes * @recover: return nodes even if the data is broken (use 0) * * * Parse a well-balanced chunk of an XML document * called by the parser * The allowed sequence for the Well Balanced Chunk is the one defined by * the content production in the XML grammar: * * [43] content ::= (element | CharData | Reference | CDSect | PI | Comment)* * * Returns 0 if the chunk is well balanced, -1 in case of args problem and * the parser error code otherwise * * In case recover is set to 1, the nodelist will not be empty even if * the parsed chunk is not well balanced, assuming the parsing succeeded to * some extent. */ int xmlParseBalancedChunkMemoryRecover(xmlDocPtr doc, xmlSAXHandlerPtr sax, void *user_data, int depth, const xmlChar *string, xmlNodePtr *lst, int recover) { xmlParserCtxtPtr ctxt; xmlDocPtr newDoc; xmlSAXHandlerPtr oldsax = NULL; xmlNodePtr content, newRoot; int size; int ret = 0; if (depth > 40) { return(XML_ERR_ENTITY_LOOP); } if (lst != NULL) *lst = NULL; if (string == NULL) return(-1); size = xmlStrlen(string); ctxt = xmlCreateMemoryParserCtxt((char *) string, size); if (ctxt == NULL) return(-1); ctxt->userData = ctxt; if (sax != NULL) { oldsax = ctxt->sax; ctxt->sax = sax; if (user_data != NULL) ctxt->userData = user_data; } newDoc = xmlNewDoc(BAD_CAST "1.0"); if (newDoc == NULL) { xmlFreeParserCtxt(ctxt); return(-1); } newDoc->properties = XML_DOC_INTERNAL; if ((doc != NULL) && (doc->dict != NULL)) { xmlDictFree(ctxt->dict); ctxt->dict = doc->dict; xmlDictReference(ctxt->dict); ctxt->str_xml = xmlDictLookup(ctxt->dict, BAD_CAST "xml", 3); ctxt->str_xmlns = xmlDictLookup(ctxt->dict, BAD_CAST "xmlns", 5); ctxt->str_xml_ns = xmlDictLookup(ctxt->dict, XML_XML_NAMESPACE, 36); ctxt->dictNames = 1; } else { xmlCtxtUseOptionsInternal(ctxt, XML_PARSE_NODICT, NULL); } /* doc == NULL is only supported for historic reasons */ if (doc != NULL) { newDoc->intSubset = doc->intSubset; newDoc->extSubset = doc->extSubset; } newRoot = xmlNewDocNode(newDoc, NULL, BAD_CAST "pseudoroot", NULL); if (newRoot == NULL) { if (sax != NULL) ctxt->sax = oldsax; xmlFreeParserCtxt(ctxt); newDoc->intSubset = NULL; newDoc->extSubset = NULL; xmlFreeDoc(newDoc); return(-1); } xmlAddChild((xmlNodePtr) newDoc, newRoot); nodePush(ctxt, newRoot); /* doc == NULL is only supported for historic reasons */ if (doc == NULL) { ctxt->myDoc = newDoc; } else { ctxt->myDoc = newDoc; newDoc->children->doc = doc; /* Ensure that doc has XML spec namespace */ xmlSearchNsByHref(doc, (xmlNodePtr)doc, XML_XML_NAMESPACE); newDoc->oldNs = doc->oldNs; } ctxt->instate = XML_PARSER_CONTENT; ctxt->input_id = 2; ctxt->depth = depth; /* * Doing validity checking on chunk doesn't make sense */ ctxt->validate = 0; ctxt->loadsubset = 0; xmlDetectSAX2(ctxt); if ( doc != NULL ){ content = doc->children; doc->children = NULL; xmlParseContent(ctxt); doc->children = content; } else { xmlParseContent(ctxt); } if ((RAW == '<') && (NXT(1) == '/')) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); } else if (RAW != 0) { xmlFatalErr(ctxt, XML_ERR_EXTRA_CONTENT, NULL); } if (ctxt->node != newDoc->children) { xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); } if (!ctxt->wellFormed) { if (ctxt->errNo == 0) ret = 1; else ret = ctxt->errNo; } else { ret = 0; } if ((lst != NULL) && ((ret == 0) || (recover == 1))) { xmlNodePtr cur; /* * Return the newly created nodeset after unlinking it from * they pseudo parent. */ cur = newDoc->children->children; *lst = cur; while (cur != NULL) { xmlSetTreeDoc(cur, doc); cur->parent = NULL; cur = cur->next; } newDoc->children->children = NULL; } if (sax != NULL) ctxt->sax = oldsax; xmlFreeParserCtxt(ctxt); newDoc->intSubset = NULL; newDoc->extSubset = NULL; /* This leaks the namespace list if doc == NULL */ newDoc->oldNs = NULL; xmlFreeDoc(newDoc); return(ret); } /** * xmlSAXParseEntity: * @sax: the SAX handler block * @filename: the filename * * DEPRECATED: Don't use. * * parse an XML external entity out of context and build a tree. * It use the given SAX function block to handle the parsing callback. * If sax is NULL, fallback to the default DOM tree building routines. * * [78] extParsedEnt ::= TextDecl? content * * This correspond to a "Well Balanced" chunk * * Returns the resulting document tree */ xmlDocPtr xmlSAXParseEntity(xmlSAXHandlerPtr sax, const char *filename) { xmlDocPtr ret; xmlParserCtxtPtr ctxt; ctxt = xmlCreateFileParserCtxt(filename); if (ctxt == NULL) { return(NULL); } if (sax != NULL) { if (ctxt->sax != NULL) xmlFree(ctxt->sax); ctxt->sax = sax; ctxt->userData = NULL; } xmlParseExtParsedEnt(ctxt); if (ctxt->wellFormed) ret = ctxt->myDoc; else { ret = NULL; xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } if (sax != NULL) ctxt->sax = NULL; xmlFreeParserCtxt(ctxt); return(ret); } /** * xmlParseEntity: * @filename: the filename * * parse an XML external entity out of context and build a tree. * * [78] extParsedEnt ::= TextDecl? content * * This correspond to a "Well Balanced" chunk * * Returns the resulting document tree */ xmlDocPtr xmlParseEntity(const char *filename) { return(xmlSAXParseEntity(NULL, filename)); } #endif /* LIBXML_SAX1_ENABLED */ /** * xmlCreateEntityParserCtxtInternal: * @URL: the entity URL * @ID: the entity PUBLIC ID * @base: a possible base for the target URI * @pctx: parser context used to set options on new context * * Create a parser context for an external entity * Automatic support for ZLIB/Compress compressed document is provided * by default if found at compile-time. * * Returns the new parser context or NULL */ static xmlParserCtxtPtr xmlCreateEntityParserCtxtInternal(xmlSAXHandlerPtr sax, void *userData, const xmlChar *URL, const xmlChar *ID, const xmlChar *base, xmlParserCtxtPtr pctx) { xmlParserCtxtPtr ctxt; xmlParserInputPtr inputStream; char *directory = NULL; xmlChar *uri; ctxt = xmlNewSAXParserCtxt(sax, userData); if (ctxt == NULL) { return(NULL); } if (pctx != NULL) { ctxt->options = pctx->options; ctxt->_private = pctx->_private; ctxt->input_id = pctx->input_id; } /* Don't read from stdin. */ if (xmlStrcmp(URL, BAD_CAST "-") == 0) URL = BAD_CAST "./-"; uri = xmlBuildURI(URL, base); if (uri == NULL) { inputStream = xmlLoadExternalEntity((char *)URL, (char *)ID, ctxt); if (inputStream == NULL) { xmlFreeParserCtxt(ctxt); return(NULL); } inputPush(ctxt, inputStream); if ((ctxt->directory == NULL) && (directory == NULL)) directory = xmlParserGetDirectory((char *)URL); if ((ctxt->directory == NULL) && (directory != NULL)) ctxt->directory = directory; } else { inputStream = xmlLoadExternalEntity((char *)uri, (char *)ID, ctxt); if (inputStream == NULL) { xmlFree(uri); xmlFreeParserCtxt(ctxt); return(NULL); } inputPush(ctxt, inputStream); if ((ctxt->directory == NULL) && (directory == NULL)) directory = xmlParserGetDirectory((char *)uri); if ((ctxt->directory == NULL) && (directory != NULL)) ctxt->directory = directory; xmlFree(uri); } return(ctxt); } /** * xmlCreateEntityParserCtxt: * @URL: the entity URL * @ID: the entity PUBLIC ID * @base: a possible base for the target URI * * Create a parser context for an external entity * Automatic support for ZLIB/Compress compressed document is provided * by default if found at compile-time. * * Returns the new parser context or NULL */ xmlParserCtxtPtr xmlCreateEntityParserCtxt(const xmlChar *URL, const xmlChar *ID, const xmlChar *base) { return xmlCreateEntityParserCtxtInternal(NULL, NULL, URL, ID, base, NULL); } /************************************************************************ * * * Front ends when parsing from a file * * * ************************************************************************/ /** * xmlCreateURLParserCtxt: * @filename: the filename or URL * @options: a combination of xmlParserOption * * Create a parser context for a file or URL content. * Automatic support for ZLIB/Compress compressed document is provided * by default if found at compile-time and for file accesses * * Returns the new parser context or NULL */ xmlParserCtxtPtr xmlCreateURLParserCtxt(const char *filename, int options) { xmlParserCtxtPtr ctxt; xmlParserInputPtr inputStream; char *directory = NULL; ctxt = xmlNewParserCtxt(); if (ctxt == NULL) { xmlErrMemory(NULL, "cannot allocate parser context"); return(NULL); } if (options) xmlCtxtUseOptionsInternal(ctxt, options, NULL); ctxt->linenumbers = 1; inputStream = xmlLoadExternalEntity(filename, NULL, ctxt); if (inputStream == NULL) { xmlFreeParserCtxt(ctxt); return(NULL); } inputPush(ctxt, inputStream); if ((ctxt->directory == NULL) && (directory == NULL)) directory = xmlParserGetDirectory(filename); if ((ctxt->directory == NULL) && (directory != NULL)) ctxt->directory = directory; return(ctxt); } /** * xmlCreateFileParserCtxt: * @filename: the filename * * Create a parser context for a file content. * Automatic support for ZLIB/Compress compressed document is provided * by default if found at compile-time. * * Returns the new parser context or NULL */ xmlParserCtxtPtr xmlCreateFileParserCtxt(const char *filename) { return(xmlCreateURLParserCtxt(filename, 0)); } #ifdef LIBXML_SAX1_ENABLED /** * xmlSAXParseFileWithData: * @sax: the SAX handler block * @filename: the filename * @recovery: work in recovery mode, i.e. tries to read no Well Formed * documents * @data: the userdata * * DEPRECATED: Use xmlNewSAXParserCtxt and xmlCtxtReadFile. * * parse an XML file and build a tree. Automatic support for ZLIB/Compress * compressed document is provided by default if found at compile-time. * It use the given SAX function block to handle the parsing callback. * If sax is NULL, fallback to the default DOM tree building routines. * * User data (void *) is stored within the parser context in the * context's _private member, so it is available nearly everywhere in libxml * * Returns the resulting document tree */ xmlDocPtr xmlSAXParseFileWithData(xmlSAXHandlerPtr sax, const char *filename, int recovery, void *data) { xmlDocPtr ret; xmlParserCtxtPtr ctxt; xmlInitParser(); ctxt = xmlCreateFileParserCtxt(filename); if (ctxt == NULL) { return(NULL); } if (sax != NULL) { if (ctxt->sax != NULL) xmlFree(ctxt->sax); ctxt->sax = sax; } xmlDetectSAX2(ctxt); if (data!=NULL) { ctxt->_private = data; } if (ctxt->directory == NULL) ctxt->directory = xmlParserGetDirectory(filename); ctxt->recovery = recovery; xmlParseDocument(ctxt); if ((ctxt->wellFormed) || recovery) { ret = ctxt->myDoc; if ((ret != NULL) && (ctxt->input->buf != NULL)) { if (ctxt->input->buf->compressed > 0) ret->compression = 9; else ret->compression = ctxt->input->buf->compressed; } } else { ret = NULL; xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } if (sax != NULL) ctxt->sax = NULL; xmlFreeParserCtxt(ctxt); return(ret); } /** * xmlSAXParseFile: * @sax: the SAX handler block * @filename: the filename * @recovery: work in recovery mode, i.e. tries to read no Well Formed * documents * * DEPRECATED: Use xmlNewSAXParserCtxt and xmlCtxtReadFile. * * parse an XML file and build a tree. Automatic support for ZLIB/Compress * compressed document is provided by default if found at compile-time. * It use the given SAX function block to handle the parsing callback. * If sax is NULL, fallback to the default DOM tree building routines. * * Returns the resulting document tree */ xmlDocPtr xmlSAXParseFile(xmlSAXHandlerPtr sax, const char *filename, int recovery) { return(xmlSAXParseFileWithData(sax,filename,recovery,NULL)); } /** * xmlRecoverDoc: * @cur: a pointer to an array of xmlChar * * DEPRECATED: Use xmlReadDoc with XML_PARSE_RECOVER. * * parse an XML in-memory document and build a tree. * In the case the document is not Well Formed, a attempt to build a * tree is tried anyway * * Returns the resulting document tree or NULL in case of failure */ xmlDocPtr xmlRecoverDoc(const xmlChar *cur) { return(xmlSAXParseDoc(NULL, cur, 1)); } /** * xmlParseFile: * @filename: the filename * * DEPRECATED: Use xmlReadFile. * * parse an XML file and build a tree. Automatic support for ZLIB/Compress * compressed document is provided by default if found at compile-time. * * Returns the resulting document tree if the file was wellformed, * NULL otherwise. */ xmlDocPtr xmlParseFile(const char *filename) { return(xmlSAXParseFile(NULL, filename, 0)); } /** * xmlRecoverFile: * @filename: the filename * * DEPRECATED: Use xmlReadFile with XML_PARSE_RECOVER. * * parse an XML file and build a tree. Automatic support for ZLIB/Compress * compressed document is provided by default if found at compile-time. * In the case the document is not Well Formed, it attempts to build * a tree anyway * * Returns the resulting document tree or NULL in case of failure */ xmlDocPtr xmlRecoverFile(const char *filename) { return(xmlSAXParseFile(NULL, filename, 1)); } /** * xmlSetupParserForBuffer: * @ctxt: an XML parser context * @buffer: a xmlChar * buffer * @filename: a file name * * DEPRECATED: Don't use. * * Setup the parser context to parse a new buffer; Clears any prior * contents from the parser context. The buffer parameter must not be * NULL, but the filename parameter can be */ void xmlSetupParserForBuffer(xmlParserCtxtPtr ctxt, const xmlChar* buffer, const char* filename) { xmlParserInputPtr input; if ((ctxt == NULL) || (buffer == NULL)) return; input = xmlNewInputStream(ctxt); if (input == NULL) { xmlErrMemory(NULL, "parsing new buffer: out of memory\n"); xmlClearParserCtxt(ctxt); return; } xmlClearParserCtxt(ctxt); if (filename != NULL) input->filename = (char *) xmlCanonicPath((const xmlChar *)filename); input->base = buffer; input->cur = buffer; input->end = &buffer[xmlStrlen(buffer)]; inputPush(ctxt, input); } /** * xmlSAXUserParseFile: * @sax: a SAX handler * @user_data: The user data returned on SAX callbacks * @filename: a file name * * DEPRECATED: Use xmlNewSAXParserCtxt and xmlCtxtReadFile. * * parse an XML file and call the given SAX handler routines. * Automatic support for ZLIB/Compress compressed document is provided * * Returns 0 in case of success or a error number otherwise */ int xmlSAXUserParseFile(xmlSAXHandlerPtr sax, void *user_data, const char *filename) { int ret = 0; xmlParserCtxtPtr ctxt; ctxt = xmlCreateFileParserCtxt(filename); if (ctxt == NULL) return -1; if (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler) xmlFree(ctxt->sax); ctxt->sax = sax; xmlDetectSAX2(ctxt); if (user_data != NULL) ctxt->userData = user_data; xmlParseDocument(ctxt); if (ctxt->wellFormed) ret = 0; else { if (ctxt->errNo != 0) ret = ctxt->errNo; else ret = -1; } if (sax != NULL) ctxt->sax = NULL; if (ctxt->myDoc != NULL) { xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } xmlFreeParserCtxt(ctxt); return ret; } #endif /* LIBXML_SAX1_ENABLED */ /************************************************************************ * * * Front ends when parsing from memory * * * ************************************************************************/ /** * xmlCreateMemoryParserCtxt: * @buffer: a pointer to a char array * @size: the size of the array * * Create a parser context for an XML in-memory document. * * Returns the new parser context or NULL */ xmlParserCtxtPtr xmlCreateMemoryParserCtxt(const char *buffer, int size) { xmlParserCtxtPtr ctxt; xmlParserInputPtr input; xmlParserInputBufferPtr buf; if (buffer == NULL) return(NULL); if (size <= 0) return(NULL); ctxt = xmlNewParserCtxt(); if (ctxt == NULL) return(NULL); buf = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE); if (buf == NULL) { xmlFreeParserCtxt(ctxt); return(NULL); } input = xmlNewInputStream(ctxt); if (input == NULL) { xmlFreeParserInputBuffer(buf); xmlFreeParserCtxt(ctxt); return(NULL); } input->filename = NULL; input->buf = buf; xmlBufResetInput(input->buf->buffer, input); inputPush(ctxt, input); return(ctxt); } #ifdef LIBXML_SAX1_ENABLED /** * xmlSAXParseMemoryWithData: * @sax: the SAX handler block * @buffer: an pointer to a char array * @size: the size of the array * @recovery: work in recovery mode, i.e. tries to read no Well Formed * documents * @data: the userdata * * DEPRECATED: Use xmlNewSAXParserCtxt and xmlCtxtReadMemory. * * parse an XML in-memory block and use the given SAX function block * to handle the parsing callback. If sax is NULL, fallback to the default * DOM tree building routines. * * User data (void *) is stored within the parser context in the * context's _private member, so it is available nearly everywhere in libxml * * Returns the resulting document tree */ xmlDocPtr xmlSAXParseMemoryWithData(xmlSAXHandlerPtr sax, const char *buffer, int size, int recovery, void *data) { xmlDocPtr ret; xmlParserCtxtPtr ctxt; xmlInitParser(); ctxt = xmlCreateMemoryParserCtxt(buffer, size); if (ctxt == NULL) return(NULL); if (sax != NULL) { if (ctxt->sax != NULL) xmlFree(ctxt->sax); ctxt->sax = sax; } xmlDetectSAX2(ctxt); if (data!=NULL) { ctxt->_private=data; } ctxt->recovery = recovery; xmlParseDocument(ctxt); if ((ctxt->wellFormed) || recovery) ret = ctxt->myDoc; else { ret = NULL; xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } if (sax != NULL) ctxt->sax = NULL; xmlFreeParserCtxt(ctxt); return(ret); } /** * xmlSAXParseMemory: * @sax: the SAX handler block * @buffer: an pointer to a char array * @size: the size of the array * @recovery: work in recovery mode, i.e. tries to read not Well Formed * documents * * DEPRECATED: Use xmlNewSAXParserCtxt and xmlCtxtReadMemory. * * parse an XML in-memory block and use the given SAX function block * to handle the parsing callback. If sax is NULL, fallback to the default * DOM tree building routines. * * Returns the resulting document tree */ xmlDocPtr xmlSAXParseMemory(xmlSAXHandlerPtr sax, const char *buffer, int size, int recovery) { return xmlSAXParseMemoryWithData(sax, buffer, size, recovery, NULL); } /** * xmlParseMemory: * @buffer: an pointer to a char array * @size: the size of the array * * DEPRECATED: Use xmlReadMemory. * * parse an XML in-memory block and build a tree. * * Returns the resulting document tree */ xmlDocPtr xmlParseMemory(const char *buffer, int size) { return(xmlSAXParseMemory(NULL, buffer, size, 0)); } /** * xmlRecoverMemory: * @buffer: an pointer to a char array * @size: the size of the array * * DEPRECATED: Use xmlReadMemory with XML_PARSE_RECOVER. * * parse an XML in-memory block and build a tree. * In the case the document is not Well Formed, an attempt to * build a tree is tried anyway * * Returns the resulting document tree or NULL in case of error */ xmlDocPtr xmlRecoverMemory(const char *buffer, int size) { return(xmlSAXParseMemory(NULL, buffer, size, 1)); } /** * xmlSAXUserParseMemory: * @sax: a SAX handler * @user_data: The user data returned on SAX callbacks * @buffer: an in-memory XML document input * @size: the length of the XML document in bytes * * DEPRECATED: Use xmlNewSAXParserCtxt and xmlCtxtReadMemory. * * parse an XML in-memory buffer and call the given SAX handler routines. * * Returns 0 in case of success or a error number otherwise */ int xmlSAXUserParseMemory(xmlSAXHandlerPtr sax, void *user_data, const char *buffer, int size) { int ret = 0; xmlParserCtxtPtr ctxt; xmlInitParser(); ctxt = xmlCreateMemoryParserCtxt(buffer, size); if (ctxt == NULL) return -1; if (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler) xmlFree(ctxt->sax); ctxt->sax = sax; xmlDetectSAX2(ctxt); if (user_data != NULL) ctxt->userData = user_data; xmlParseDocument(ctxt); if (ctxt->wellFormed) ret = 0; else { if (ctxt->errNo != 0) ret = ctxt->errNo; else ret = -1; } if (sax != NULL) ctxt->sax = NULL; if (ctxt->myDoc != NULL) { xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } xmlFreeParserCtxt(ctxt); return ret; } #endif /* LIBXML_SAX1_ENABLED */ /** * xmlCreateDocParserCtxt: * @cur: a pointer to an array of xmlChar * * Creates a parser context for an XML in-memory document. * * Returns the new parser context or NULL */ xmlParserCtxtPtr xmlCreateDocParserCtxt(const xmlChar *cur) { int len; if (cur == NULL) return(NULL); len = xmlStrlen(cur); return(xmlCreateMemoryParserCtxt((const char *)cur, len)); } #ifdef LIBXML_SAX1_ENABLED /** * xmlSAXParseDoc: * @sax: the SAX handler block * @cur: a pointer to an array of xmlChar * @recovery: work in recovery mode, i.e. tries to read no Well Formed * documents * * DEPRECATED: Use xmlNewSAXParserCtxt and xmlCtxtReadDoc. * * parse an XML in-memory document and build a tree. * It use the given SAX function block to handle the parsing callback. * If sax is NULL, fallback to the default DOM tree building routines. * * Returns the resulting document tree */ xmlDocPtr xmlSAXParseDoc(xmlSAXHandlerPtr sax, const xmlChar *cur, int recovery) { xmlDocPtr ret; xmlParserCtxtPtr ctxt; xmlSAXHandlerPtr oldsax = NULL; if (cur == NULL) return(NULL); ctxt = xmlCreateDocParserCtxt(cur); if (ctxt == NULL) return(NULL); if (sax != NULL) { oldsax = ctxt->sax; ctxt->sax = sax; ctxt->userData = NULL; } xmlDetectSAX2(ctxt); xmlParseDocument(ctxt); if ((ctxt->wellFormed) || recovery) ret = ctxt->myDoc; else { ret = NULL; xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; } if (sax != NULL) ctxt->sax = oldsax; xmlFreeParserCtxt(ctxt); return(ret); } /** * xmlParseDoc: * @cur: a pointer to an array of xmlChar * * DEPRECATED: Use xmlReadDoc. * * parse an XML in-memory document and build a tree. * * Returns the resulting document tree */ xmlDocPtr xmlParseDoc(const xmlChar *cur) { return(xmlSAXParseDoc(NULL, cur, 0)); } #endif /* LIBXML_SAX1_ENABLED */ #ifdef LIBXML_LEGACY_ENABLED /************************************************************************ * * * Specific function to keep track of entities references * * and used by the XSLT debugger * * * ************************************************************************/ static xmlEntityReferenceFunc xmlEntityRefFunc = NULL; /** * xmlAddEntityReference: * @ent : A valid entity * @firstNode : A valid first node for children of entity * @lastNode : A valid last node of children entity * * Notify of a reference to an entity of type XML_EXTERNAL_GENERAL_PARSED_ENTITY */ static void xmlAddEntityReference(xmlEntityPtr ent, xmlNodePtr firstNode, xmlNodePtr lastNode) { if (xmlEntityRefFunc != NULL) { (*xmlEntityRefFunc) (ent, firstNode, lastNode); } } /** * xmlSetEntityReferenceFunc: * @func: A valid function * * Set the function to call call back when a xml reference has been made */ void xmlSetEntityReferenceFunc(xmlEntityReferenceFunc func) { xmlEntityRefFunc = func; } #endif /* LIBXML_LEGACY_ENABLED */ /************************************************************************ * * * Miscellaneous * * * ************************************************************************/ static int xmlParserInitialized = 0; /** * xmlInitParser: * * Initialization function for the XML parser. * This is not reentrant. Call once before processing in case of * use in multithreaded programs. */ void xmlInitParser(void) { /* * Note that the initialization code must not make memory allocations. */ if (xmlParserInitialized != 0) return; #ifdef LIBXML_THREAD_ENABLED __xmlGlobalInitMutexLock(); if (xmlParserInitialized == 0) { #endif #if defined(_WIN32) && (!defined(LIBXML_STATIC) || defined(LIBXML_STATIC_FOR_DLL)) if (xmlFree == free) atexit(xmlCleanupParser); #endif xmlInitThreadsInternal(); xmlInitGlobalsInternal(); xmlInitMemoryInternal(); __xmlInitializeDict(); xmlInitEncodingInternal(); xmlRegisterDefaultInputCallbacks(); #ifdef LIBXML_OUTPUT_ENABLED xmlRegisterDefaultOutputCallbacks(); #endif /* LIBXML_OUTPUT_ENABLED */ #if defined(LIBXML_XPATH_ENABLED) || defined(LIBXML_SCHEMAS_ENABLED) xmlInitXPathInternal(); #endif xmlParserInitialized = 1; #ifdef LIBXML_THREAD_ENABLED } __xmlGlobalInitMutexUnlock(); #endif } /** * xmlCleanupParser: * * This function name is somewhat misleading. It does not clean up * parser state, it cleans up memory allocated by the library itself. * It is a cleanup function for the XML library. It tries to reclaim all * related global memory allocated for the library processing. * It doesn't deallocate any document related memory. One should * call xmlCleanupParser() only when the process has finished using * the library and all XML/HTML documents built with it. * See also xmlInitParser() which has the opposite function of preparing * the library for operations. * * WARNING: if your application is multithreaded or has plugin support * calling this may crash the application if another thread or * a plugin is still using libxml2. It's sometimes very hard to * guess if libxml2 is in use in the application, some libraries * or plugins may use it without notice. In case of doubt abstain * from calling this function or do it just before calling exit() * to avoid leak reports from valgrind ! */ void xmlCleanupParser(void) { if (!xmlParserInitialized) return; xmlCleanupCharEncodingHandlers(); #ifdef LIBXML_CATALOG_ENABLED xmlCatalogCleanup(); #endif xmlCleanupDictInternal(); xmlCleanupInputCallbacks(); #ifdef LIBXML_OUTPUT_ENABLED xmlCleanupOutputCallbacks(); #endif #ifdef LIBXML_SCHEMAS_ENABLED xmlSchemaCleanupTypes(); xmlRelaxNGCleanupTypes(); #endif xmlCleanupGlobalsInternal(); xmlCleanupThreadsInternal(); xmlCleanupMemoryInternal(); xmlParserInitialized = 0; } #if defined(HAVE_ATTRIBUTE_DESTRUCTOR) && !defined(LIBXML_STATIC) && \ !defined(_WIN32) static void ATTRIBUTE_DESTRUCTOR xmlDestructor(void) { /* * Calling custom deallocation functions in a destructor can cause * problems, for example with Nokogiri. */ if (xmlFree == free) xmlCleanupParser(); } #endif /************************************************************************ * * * New set (2.6.0) of simpler and more flexible APIs * * * ************************************************************************/ /** * DICT_FREE: * @str: a string * * Free a string if it is not owned by the "dict" dictionary in the * current scope */ #define DICT_FREE(str) \ if ((str) && ((!dict) || \ (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \ xmlFree((char *)(str)); /** * xmlCtxtReset: * @ctxt: an XML parser context * * Reset a parser context */ void xmlCtxtReset(xmlParserCtxtPtr ctxt) { xmlParserInputPtr input; xmlDictPtr dict; if (ctxt == NULL) return; dict = ctxt->dict; while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */ xmlFreeInputStream(input); } ctxt->inputNr = 0; ctxt->input = NULL; ctxt->spaceNr = 0; if (ctxt->spaceTab != NULL) { ctxt->spaceTab[0] = -1; ctxt->space = &ctxt->spaceTab[0]; } else { ctxt->space = NULL; } ctxt->nodeNr = 0; ctxt->node = NULL; ctxt->nameNr = 0; ctxt->name = NULL; ctxt->nsNr = 0; DICT_FREE(ctxt->version); ctxt->version = NULL; DICT_FREE(ctxt->encoding); ctxt->encoding = NULL; DICT_FREE(ctxt->directory); ctxt->directory = NULL; DICT_FREE(ctxt->extSubURI); ctxt->extSubURI = NULL; DICT_FREE(ctxt->extSubSystem); ctxt->extSubSystem = NULL; if (ctxt->myDoc != NULL) xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL; ctxt->standalone = -1; ctxt->hasExternalSubset = 0; ctxt->hasPErefs = 0; ctxt->html = 0; ctxt->external = 0; ctxt->instate = XML_PARSER_START; ctxt->token = 0; ctxt->wellFormed = 1; ctxt->nsWellFormed = 1; ctxt->disableSAX = 0; ctxt->valid = 1; #if 0 ctxt->vctxt.userData = ctxt; ctxt->vctxt.error = xmlParserValidityError; ctxt->vctxt.warning = xmlParserValidityWarning; #endif ctxt->record_info = 0; ctxt->checkIndex = 0; ctxt->endCheckState = 0; ctxt->inSubset = 0; ctxt->errNo = XML_ERR_OK; ctxt->depth = 0; ctxt->charset = XML_CHAR_ENCODING_UTF8; ctxt->catalogs = NULL; ctxt->sizeentities = 0; ctxt->sizeentcopy = 0; xmlInitNodeInfoSeq(&ctxt->node_seq); if (ctxt->attsDefault != NULL) { xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator); ctxt->attsDefault = NULL; } if (ctxt->attsSpecial != NULL) { xmlHashFree(ctxt->attsSpecial, NULL); ctxt->attsSpecial = NULL; } #ifdef LIBXML_CATALOG_ENABLED if (ctxt->catalogs != NULL) xmlCatalogFreeLocal(ctxt->catalogs); #endif ctxt->nbErrors = 0; ctxt->nbWarnings = 0; if (ctxt->lastError.code != XML_ERR_OK) xmlResetError(&ctxt->lastError); } /** * xmlCtxtResetPush: * @ctxt: an XML parser context * @chunk: a pointer to an array of chars * @size: number of chars in the array * @filename: an optional file name or URI * @encoding: the document encoding, or NULL * * Reset a push parser context * * Returns 0 in case of success and 1 in case of error */ int xmlCtxtResetPush(xmlParserCtxtPtr ctxt, const char *chunk, int size, const char *filename, const char *encoding) { xmlParserInputPtr inputStream; xmlParserInputBufferPtr buf; xmlCharEncoding enc = XML_CHAR_ENCODING_NONE; if (ctxt == NULL) return(1); if ((encoding == NULL) && (chunk != NULL) && (size >= 4)) enc = xmlDetectCharEncoding((const xmlChar *) chunk, size); buf = xmlAllocParserInputBuffer(enc); if (buf == NULL) return(1); if (ctxt == NULL) { xmlFreeParserInputBuffer(buf); return(1); } xmlCtxtReset(ctxt); if (filename == NULL) { ctxt->directory = NULL; } else { ctxt->directory = xmlParserGetDirectory(filename); } inputStream = xmlNewInputStream(ctxt); if (inputStream == NULL) { xmlFreeParserInputBuffer(buf); return(1); } if (filename == NULL) inputStream->filename = NULL; else inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) filename); inputStream->buf = buf; xmlBufResetInput(buf->buffer, inputStream); inputPush(ctxt, inputStream); if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) && (ctxt->input->buf != NULL)) { size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input); size_t cur = ctxt->input->cur - ctxt->input->base; int res; res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk); xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur); if (res < 0) { xmlFatalErr(ctxt, ctxt->input->buf->error, NULL); xmlHaltParser(ctxt); return(1); } #ifdef DEBUG_PUSH xmlGenericError(xmlGenericErrorContext, "PP: pushed %d\n", size); #endif } if (encoding != NULL) { xmlCharEncodingHandlerPtr hdlr; if (ctxt->encoding != NULL) xmlFree((xmlChar *) ctxt->encoding); ctxt->encoding = xmlStrdup((const xmlChar *) encoding); hdlr = xmlFindCharEncodingHandler(encoding); if (hdlr != NULL) { xmlSwitchToEncoding(ctxt, hdlr); } else { xmlFatalErrMsgStr(ctxt, XML_ERR_UNSUPPORTED_ENCODING, "Unsupported encoding %s\n", BAD_CAST encoding); } } else if (enc != XML_CHAR_ENCODING_NONE) { xmlSwitchEncoding(ctxt, enc); } return(0); } /** * xmlCtxtUseOptionsInternal: * @ctxt: an XML parser context * @options: a combination of xmlParserOption * @encoding: the user provided encoding to use * * Applies the options to the parser context * * Returns 0 in case of success, the set of unknown or unimplemented options * in case of error. */ static int xmlCtxtUseOptionsInternal(xmlParserCtxtPtr ctxt, int options, const char *encoding) { if (ctxt == NULL) return(-1); if (encoding != NULL) { if (ctxt->encoding != NULL) xmlFree((xmlChar *) ctxt->encoding); ctxt->encoding = xmlStrdup((const xmlChar *) encoding); } if (options & XML_PARSE_RECOVER) { ctxt->recovery = 1; options -= XML_PARSE_RECOVER; ctxt->options |= XML_PARSE_RECOVER; } else ctxt->recovery = 0; if (options & XML_PARSE_DTDLOAD) { ctxt->loadsubset = XML_DETECT_IDS; options -= XML_PARSE_DTDLOAD; ctxt->options |= XML_PARSE_DTDLOAD; } else ctxt->loadsubset = 0; if (options & XML_PARSE_DTDATTR) { ctxt->loadsubset |= XML_COMPLETE_ATTRS; options -= XML_PARSE_DTDATTR; ctxt->options |= XML_PARSE_DTDATTR; } if (options & XML_PARSE_NOENT) { ctxt->replaceEntities = 1; /* ctxt->loadsubset |= XML_DETECT_IDS; */ options -= XML_PARSE_NOENT; ctxt->options |= XML_PARSE_NOENT; } else ctxt->replaceEntities = 0; if (options & XML_PARSE_PEDANTIC) { ctxt->pedantic = 1; options -= XML_PARSE_PEDANTIC; ctxt->options |= XML_PARSE_PEDANTIC; } else ctxt->pedantic = 0; if (options & XML_PARSE_NOBLANKS) { ctxt->keepBlanks = 0; ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace; options -= XML_PARSE_NOBLANKS; ctxt->options |= XML_PARSE_NOBLANKS; } else ctxt->keepBlanks = 1; if (options & XML_PARSE_DTDVALID) { ctxt->validate = 1; if (options & XML_PARSE_NOWARNING) ctxt->vctxt.warning = NULL; if (options & XML_PARSE_NOERROR) ctxt->vctxt.error = NULL; options -= XML_PARSE_DTDVALID; ctxt->options |= XML_PARSE_DTDVALID; } else ctxt->validate = 0; if (options & XML_PARSE_NOWARNING) { ctxt->sax->warning = NULL; options -= XML_PARSE_NOWARNING; } if (options & XML_PARSE_NOERROR) { ctxt->sax->error = NULL; ctxt->sax->fatalError = NULL; options -= XML_PARSE_NOERROR; } #ifdef LIBXML_SAX1_ENABLED if (options & XML_PARSE_SAX1) { ctxt->sax->startElementNs = NULL; ctxt->sax->endElementNs = NULL; ctxt->sax->initialized = 1; options -= XML_PARSE_SAX1; ctxt->options |= XML_PARSE_SAX1; } #endif /* LIBXML_SAX1_ENABLED */ if (options & XML_PARSE_NODICT) { ctxt->dictNames = 0; options -= XML_PARSE_NODICT; ctxt->options |= XML_PARSE_NODICT; } else { ctxt->dictNames = 1; } if (options & XML_PARSE_NOCDATA) { ctxt->sax->cdataBlock = NULL; options -= XML_PARSE_NOCDATA; ctxt->options |= XML_PARSE_NOCDATA; } if (options & XML_PARSE_NSCLEAN) { ctxt->options |= XML_PARSE_NSCLEAN; options -= XML_PARSE_NSCLEAN; } if (options & XML_PARSE_NONET) { ctxt->options |= XML_PARSE_NONET; options -= XML_PARSE_NONET; } if (options & XML_PARSE_COMPACT) { ctxt->options |= XML_PARSE_COMPACT; options -= XML_PARSE_COMPACT; } if (options & XML_PARSE_OLD10) { ctxt->options |= XML_PARSE_OLD10; options -= XML_PARSE_OLD10; } if (options & XML_PARSE_NOBASEFIX) { ctxt->options |= XML_PARSE_NOBASEFIX; options -= XML_PARSE_NOBASEFIX; } if (options & XML_PARSE_HUGE) { ctxt->options |= XML_PARSE_HUGE; options -= XML_PARSE_HUGE; if (ctxt->dict != NULL) xmlDictSetLimit(ctxt->dict, 0); } if (options & XML_PARSE_OLDSAX) { ctxt->options |= XML_PARSE_OLDSAX; options -= XML_PARSE_OLDSAX; } if (options & XML_PARSE_IGNORE_ENC) { ctxt->options |= XML_PARSE_IGNORE_ENC; options -= XML_PARSE_IGNORE_ENC; } if (options & XML_PARSE_BIG_LINES) { ctxt->options |= XML_PARSE_BIG_LINES; options -= XML_PARSE_BIG_LINES; } ctxt->linenumbers = 1; return (options); } /** * xmlCtxtUseOptions: * @ctxt: an XML parser context * @options: a combination of xmlParserOption * * Applies the options to the parser context * * Returns 0 in case of success, the set of unknown or unimplemented options * in case of error. */ int xmlCtxtUseOptions(xmlParserCtxtPtr ctxt, int options) { return(xmlCtxtUseOptionsInternal(ctxt, options, NULL)); } /** * xmlDoRead: * @ctxt: an XML parser context * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * @reuse: keep the context for reuse * * Common front-end for the xmlRead functions * * Returns the resulting document tree or NULL */ static xmlDocPtr xmlDoRead(xmlParserCtxtPtr ctxt, const char *URL, const char *encoding, int options, int reuse) { xmlDocPtr ret; xmlCtxtUseOptionsInternal(ctxt, options, encoding); if (encoding != NULL) { xmlCharEncodingHandlerPtr hdlr; /* * TODO: We should consider to set XML_PARSE_IGNORE_ENC if the * caller provided an encoding. Otherwise, we might switch to * the encoding from the XML declaration which is likely to * break things. Also see xmlSwitchInputEncoding. */ hdlr = xmlFindCharEncodingHandler(encoding); if (hdlr != NULL) xmlSwitchToEncoding(ctxt, hdlr); } if ((URL != NULL) && (ctxt->input != NULL) && (ctxt->input->filename == NULL)) ctxt->input->filename = (char *) xmlStrdup((const xmlChar *) URL); xmlParseDocument(ctxt); if ((ctxt->wellFormed) || ctxt->recovery) ret = ctxt->myDoc; else { ret = NULL; if (ctxt->myDoc != NULL) { xmlFreeDoc(ctxt->myDoc); } } ctxt->myDoc = NULL; if (!reuse) { xmlFreeParserCtxt(ctxt); } return (ret); } /** * xmlReadDoc: * @cur: a pointer to a zero terminated string * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML in-memory document and build a tree. * * Returns the resulting document tree */ xmlDocPtr xmlReadDoc(const xmlChar * cur, const char *URL, const char *encoding, int options) { xmlParserCtxtPtr ctxt; if (cur == NULL) return (NULL); xmlInitParser(); ctxt = xmlCreateDocParserCtxt(cur); if (ctxt == NULL) return (NULL); return (xmlDoRead(ctxt, URL, encoding, options, 0)); } /** * xmlReadFile: * @filename: a file or URL * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML file from the filesystem or the network. * * Returns the resulting document tree */ xmlDocPtr xmlReadFile(const char *filename, const char *encoding, int options) { xmlParserCtxtPtr ctxt; xmlInitParser(); ctxt = xmlCreateURLParserCtxt(filename, options); if (ctxt == NULL) return (NULL); return (xmlDoRead(ctxt, NULL, encoding, options, 0)); } /** * xmlReadMemory: * @buffer: a pointer to a char array * @size: the size of the array * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML in-memory document and build a tree. * * Returns the resulting document tree */ xmlDocPtr xmlReadMemory(const char *buffer, int size, const char *URL, const char *encoding, int options) { xmlParserCtxtPtr ctxt; xmlInitParser(); ctxt = xmlCreateMemoryParserCtxt(buffer, size); if (ctxt == NULL) return (NULL); return (xmlDoRead(ctxt, URL, encoding, options, 0)); } /** * xmlReadFd: * @fd: an open file descriptor * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML from a file descriptor and build a tree. * NOTE that the file descriptor will not be closed when the * reader is closed or reset. * * Returns the resulting document tree */ xmlDocPtr xmlReadFd(int fd, const char *URL, const char *encoding, int options) { xmlParserCtxtPtr ctxt; xmlParserInputBufferPtr input; xmlParserInputPtr stream; if (fd < 0) return (NULL); xmlInitParser(); input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE); if (input == NULL) return (NULL); input->closecallback = NULL; ctxt = xmlNewParserCtxt(); if (ctxt == NULL) { xmlFreeParserInputBuffer(input); return (NULL); } stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); if (stream == NULL) { xmlFreeParserInputBuffer(input); xmlFreeParserCtxt(ctxt); return (NULL); } inputPush(ctxt, stream); return (xmlDoRead(ctxt, URL, encoding, options, 0)); } /** * xmlReadIO: * @ioread: an I/O read function * @ioclose: an I/O close function * @ioctx: an I/O handler * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML document from I/O functions and source and build a tree. * * Returns the resulting document tree */ xmlDocPtr xmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose, void *ioctx, const char *URL, const char *encoding, int options) { xmlParserCtxtPtr ctxt; xmlParserInputBufferPtr input; xmlParserInputPtr stream; if (ioread == NULL) return (NULL); xmlInitParser(); input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx, XML_CHAR_ENCODING_NONE); if (input == NULL) { if (ioclose != NULL) ioclose(ioctx); return (NULL); } ctxt = xmlNewParserCtxt(); if (ctxt == NULL) { xmlFreeParserInputBuffer(input); return (NULL); } stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); if (stream == NULL) { xmlFreeParserInputBuffer(input); xmlFreeParserCtxt(ctxt); return (NULL); } inputPush(ctxt, stream); return (xmlDoRead(ctxt, URL, encoding, options, 0)); } /** * xmlCtxtReadDoc: * @ctxt: an XML parser context * @cur: a pointer to a zero terminated string * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML in-memory document and build a tree. * This reuses the existing @ctxt parser context * * Returns the resulting document tree */ xmlDocPtr xmlCtxtReadDoc(xmlParserCtxtPtr ctxt, const xmlChar * cur, const char *URL, const char *encoding, int options) { if (cur == NULL) return (NULL); return (xmlCtxtReadMemory(ctxt, (const char *) cur, xmlStrlen(cur), URL, encoding, options)); } /** * xmlCtxtReadFile: * @ctxt: an XML parser context * @filename: a file or URL * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML file from the filesystem or the network. * This reuses the existing @ctxt parser context * * Returns the resulting document tree */ xmlDocPtr xmlCtxtReadFile(xmlParserCtxtPtr ctxt, const char *filename, const char *encoding, int options) { xmlParserInputPtr stream; if (filename == NULL) return (NULL); if (ctxt == NULL) return (NULL); xmlInitParser(); xmlCtxtReset(ctxt); stream = xmlLoadExternalEntity(filename, NULL, ctxt); if (stream == NULL) { return (NULL); } inputPush(ctxt, stream); return (xmlDoRead(ctxt, NULL, encoding, options, 1)); } /** * xmlCtxtReadMemory: * @ctxt: an XML parser context * @buffer: a pointer to a char array * @size: the size of the array * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML in-memory document and build a tree. * This reuses the existing @ctxt parser context * * Returns the resulting document tree */ xmlDocPtr xmlCtxtReadMemory(xmlParserCtxtPtr ctxt, const char *buffer, int size, const char *URL, const char *encoding, int options) { xmlParserInputBufferPtr input; xmlParserInputPtr stream; if (ctxt == NULL) return (NULL); if (buffer == NULL) return (NULL); xmlInitParser(); xmlCtxtReset(ctxt); input = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE); if (input == NULL) { return(NULL); } stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); if (stream == NULL) { xmlFreeParserInputBuffer(input); return(NULL); } inputPush(ctxt, stream); return (xmlDoRead(ctxt, URL, encoding, options, 1)); } /** * xmlCtxtReadFd: * @ctxt: an XML parser context * @fd: an open file descriptor * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML from a file descriptor and build a tree. * This reuses the existing @ctxt parser context * NOTE that the file descriptor will not be closed when the * reader is closed or reset. * * Returns the resulting document tree */ xmlDocPtr xmlCtxtReadFd(xmlParserCtxtPtr ctxt, int fd, const char *URL, const char *encoding, int options) { xmlParserInputBufferPtr input; xmlParserInputPtr stream; if (fd < 0) return (NULL); if (ctxt == NULL) return (NULL); xmlInitParser(); xmlCtxtReset(ctxt); input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE); if (input == NULL) return (NULL); input->closecallback = NULL; stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); if (stream == NULL) { xmlFreeParserInputBuffer(input); return (NULL); } inputPush(ctxt, stream); return (xmlDoRead(ctxt, URL, encoding, options, 1)); } /** * xmlCtxtReadIO: * @ctxt: an XML parser context * @ioread: an I/O read function * @ioclose: an I/O close function * @ioctx: an I/O handler * @URL: the base URL to use for the document * @encoding: the document encoding, or NULL * @options: a combination of xmlParserOption * * parse an XML document from I/O functions and source and build a tree. * This reuses the existing @ctxt parser context * * Returns the resulting document tree */ xmlDocPtr xmlCtxtReadIO(xmlParserCtxtPtr ctxt, xmlInputReadCallback ioread, xmlInputCloseCallback ioclose, void *ioctx, const char *URL, const char *encoding, int options) { xmlParserInputBufferPtr input; xmlParserInputPtr stream; if (ioread == NULL) return (NULL); if (ctxt == NULL) return (NULL); xmlInitParser(); xmlCtxtReset(ctxt); input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx, XML_CHAR_ENCODING_NONE); if (input == NULL) { if (ioclose != NULL) ioclose(ioctx); return (NULL); } stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); if (stream == NULL) { xmlFreeParserInputBuffer(input); return (NULL); } inputPush(ctxt, stream); return (xmlDoRead(ctxt, URL, encoding, options, 1)); }