xmlparse.c

一个用visual studio C++开发的XML文件解析和生成程序.

  tagNamePtr->localPart = localPart;
  tagNamePtr->uriLen = binding->uriLen;
  for (i = 0; localPart[i++];)
    ;
  n = i + binding->uriLen;
  if (n > binding->uriAlloc) {
    TAG *p;
    XML_Char *uri = malloc((n + EXPAND_SPARE) * sizeof(XML_Char));
    if (!uri)
      return XML_ERROR_NO_MEMORY;
    binding->uriAlloc = n + EXPAND_SPARE;
    memcpy(uri, binding->uri, binding->uriLen * sizeof(XML_Char));
    for (p = tagStack; p; p = p->parent)
      if (p->name.str == binding->uri)
	p->name.str = uri;
    free(binding->uri);
    binding->uri = uri;
  }
  memcpy(binding->uri + binding->uriLen, localPart, i * sizeof(XML_Char));
  tagNamePtr->str = binding->uri;
  return XML_ERROR_NONE;
}

static
int addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId, const XML_Char *uri, BINDING **bindingsPtr)
{
  BINDING *b;
  int len;
  for (len = 0; uri[len]; len++)
    ;
  if (namespaceSeparator)
    len++;
  if (freeBindingList) {
    b = freeBindingList;
    if (len > b->uriAlloc) {
      b->uri = realloc(b->uri, sizeof(XML_Char) * (len + EXPAND_SPARE));
      if (!b->uri)
	return 0;
      b->uriAlloc = len + EXPAND_SPARE;
    }
    freeBindingList = b->nextTagBinding;
  }
  else {
    b = malloc(sizeof(BINDING));
    if (!b)
      return 0;
    b->uri = malloc(sizeof(XML_Char) * (len + EXPAND_SPARE));
    if (!b->uri) {
      free(b);
      return 0;
    }
    b->uriAlloc = len + EXPAND_SPARE;
  }
  b->uriLen = len;
  memcpy(b->uri, uri, len * sizeof(XML_Char));
  if (namespaceSeparator)
    b->uri[len - 1] = namespaceSeparator;
  b->prefix = prefix;
  b->attId = attId;
  b->prevPrefixBinding = prefix->binding;
  if (*uri == XML_T('\0') && prefix == &dtd.defaultPrefix)
    prefix->binding = 0;
  else
    prefix->binding = b;
  b->nextTagBinding = *bindingsPtr;
  *bindingsPtr = b;
  if (startNamespaceDeclHandler)
    startNamespaceDeclHandler(handlerArg, prefix->name,
			      prefix->binding ? uri : 0);
  return 1;
}

/* The idea here is to avoid using stack for each CDATA section when
the whole file is parsed with one call. */

static
enum XML_Error cdataSectionProcessor(XML_Parser parser,
				     const char *start,
			    	     const char *end,
				     const char **endPtr)
{
  enum XML_Error result = doCdataSection(parser, encoding, &start, end, endPtr);
  if (start) {
    processor = contentProcessor;
    return contentProcessor(parser, start, end, endPtr);
  }
  return result;
}

/* startPtr gets set to non-null is the section is closed, and to null if
the section is not yet closed. */

static
enum XML_Error doCdataSection(XML_Parser parser,
			      const ENCODING *enc,
			      const char **startPtr,
			      const char *end,
			      const char **nextPtr)
{
  const char *s = *startPtr;
  const char **eventPP;
  const char **eventEndPP;
  if (enc == encoding) {
    eventPP = &eventPtr;
    *eventPP = s;
    eventEndPP = &eventEndPtr;
  }
  else {
    eventPP = &(openInternalEntities->internalEventPtr);
    eventEndPP = &(openInternalEntities->internalEventEndPtr);
  }
  *eventPP = s;
  *startPtr = 0;
  for (;;) {
    const char *next;
    int tok = XmlCdataSectionTok(enc, s, end, &next);
    *eventEndPP = next;
    switch (tok) {
    case XML_TOK_CDATA_SECT_CLOSE:
      if (endCdataSectionHandler)
	endCdataSectionHandler(handlerArg);
#if 0
      /* see comment under XML_TOK_CDATA_SECT_OPEN */
      else if (characterDataHandler)
	characterDataHandler(handlerArg, dataBuf, 0);
#endif
      else if (defaultHandler)
	reportDefault(parser, enc, s, next);
      *startPtr = next;
      return XML_ERROR_NONE;
    case XML_TOK_DATA_NEWLINE:
      if (characterDataHandler) {
	XML_Char c = 0xA;
	characterDataHandler(handlerArg, &c, 1);
      }
      else if (defaultHandler)
	reportDefault(parser, enc, s, next);
      break;
    case XML_TOK_DATA_CHARS:
      if (characterDataHandler) {
	if (MUST_CONVERT(enc, s)) {
	  for (;;) {
  	    ICHAR *dataPtr = (ICHAR *)dataBuf;
	    XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
	    *eventEndPP = next;
	    characterDataHandler(handlerArg, dataBuf, dataPtr - (ICHAR *)dataBuf);
	    if (s == next)
	      break;
	    *eventPP = s;
	  }
	}
	else
	  characterDataHandler(handlerArg,
		  	       (XML_Char *)s,
			       (XML_Char *)next - (XML_Char *)s);
      }
      else if (defaultHandler)
	reportDefault(parser, enc, s, next);
      break;
    case XML_TOK_INVALID:
      *eventPP = next;
      return XML_ERROR_INVALID_TOKEN;
    case XML_TOK_PARTIAL_CHAR:
      if (nextPtr) {
	*nextPtr = s;
	return XML_ERROR_NONE;
      }
      return XML_ERROR_PARTIAL_CHAR;
    case XML_TOK_PARTIAL:
    case XML_TOK_NONE:
      if (nextPtr) {
	*nextPtr = s;
	return XML_ERROR_NONE;
      }
      return XML_ERROR_UNCLOSED_CDATA_SECTION;
    default:
      abort();
    }
    *eventPP = s = next;
  }
  /* not reached */
}

#ifdef XML_DTD

/* The idea here is to avoid using stack for each IGNORE section when
the whole file is parsed with one call. */

static
enum XML_Error ignoreSectionProcessor(XML_Parser parser,
				      const char *start,
				      const char *end,
				      const char **endPtr)
{
  enum XML_Error result = doIgnoreSection(parser, encoding, &start, end, endPtr);
  if (start) {
    processor = prologProcessor;
    return prologProcessor(parser, start, end, endPtr);
  }
  return result;
}

/* startPtr gets set to non-null is the section is closed, and to null if
the section is not yet closed. */

static
enum XML_Error doIgnoreSection(XML_Parser parser,
			       const ENCODING *enc,
			       const char **startPtr,
			       const char *end,
			       const char **nextPtr)
{
  const char *next;
  int tok;
  const char *s = *startPtr;
  const char **eventPP;
  const char **eventEndPP;
  if (enc == encoding) {
    eventPP = &eventPtr;
    *eventPP = s;
    eventEndPP = &eventEndPtr;
  }
  else {
    eventPP = &(openInternalEntities->internalEventPtr);
    eventEndPP = &(openInternalEntities->internalEventEndPtr);
  }
  *eventPP = s;
  *startPtr = 0;
  tok = XmlIgnoreSectionTok(enc, s, end, &next);
  *eventEndPP = next;
  switch (tok) {
  case XML_TOK_IGNORE_SECT:
    if (defaultHandler)
      reportDefault(parser, enc, s, next);
    *startPtr = next;
    return XML_ERROR_NONE;
  case XML_TOK_INVALID:
    *eventPP = next;
    return XML_ERROR_INVALID_TOKEN;
  case XML_TOK_PARTIAL_CHAR:
    if (nextPtr) {
      *nextPtr = s;
      return XML_ERROR_NONE;
    }
    return XML_ERROR_PARTIAL_CHAR;
  case XML_TOK_PARTIAL:
  case XML_TOK_NONE:
    if (nextPtr) {
      *nextPtr = s;
      return XML_ERROR_NONE;
    }
    return XML_ERROR_SYNTAX; /* XML_ERROR_UNCLOSED_IGNORE_SECTION */
  default:
    abort();
  }
  /* not reached */
}

#endif /* XML_DTD */

static enum XML_Error
initializeEncoding(XML_Parser parser)
{
  const char *s;
#ifdef XML_UNICODE
  char encodingBuf[128];
  if (!protocolEncodingName)
    s = 0;
  else {
    int i;
    for (i = 0; protocolEncodingName[i]; i++) {
      if (i == sizeof(encodingBuf) - 1
	  || (protocolEncodingName[i] & ~0x7f) != 0) {
	encodingBuf[0] = '\0';
	break;
      }
      encodingBuf[i] = (char)protocolEncodingName[i];
    }
    encodingBuf[i] = '\0';
    s = encodingBuf;
  }
#else
  s = protocolEncodingName;
#endif
  if ((ns ? XmlInitEncodingNS : XmlInitEncoding)(&initEncoding, &encoding, s))
    return XML_ERROR_NONE;
  return handleUnknownEncoding(parser, protocolEncodingName);
}

static enum XML_Error
processXmlDecl(XML_Parser parser, int isGeneralTextEntity,
	       const char *s, const char *next)
{
  const char *encodingName = 0;
  const ENCODING *newEncoding = 0;
  const char *version;
  int standalone = -1;
  if (!(ns
        ? XmlParseXmlDeclNS
	: XmlParseXmlDecl)(isGeneralTextEntity,
		           encoding,
		           s,
		           next,
		           &eventPtr,
		           &version,
		           &encodingName,
		           &newEncoding,
		           &standalone))
    return XML_ERROR_SYNTAX;
  if (!isGeneralTextEntity && standalone == 1) {
    dtd.standalone = 1;
#ifdef XML_DTD
    if (paramEntityParsing == XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE)
      paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER;
#endif /* XML_DTD */
  }
  if (defaultHandler)
    reportDefault(parser, encoding, s, next);
  if (!protocolEncodingName) {
    if (newEncoding) {
      if (newEncoding->minBytesPerChar != encoding->minBytesPerChar) {
	eventPtr = encodingName;
	return XML_ERROR_INCORRECT_ENCODING;
      }
      encoding = newEncoding;
    }
    else if (encodingName) {
      enum XML_Error result;
      const XML_Char *s = poolStoreString(&tempPool,
					  encoding,
					  encodingName,
					  encodingName
					  + XmlNameLength(encoding, encodingName));
      if (!s)
	return XML_ERROR_NO_MEMORY;
      result = handleUnknownEncoding(parser, s);
      poolDiscard(&tempPool);
      if (result == XML_ERROR_UNKNOWN_ENCODING)
	eventPtr = encodingName;
      return result;
    }
  }
  return XML_ERROR_NONE;
}

static enum XML_Error
handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName)
{
  if (unknownEncodingHandler) {
    XML_Encoding info;
    int i;
    for (i = 0; i < 256; i++)
      info.map[i] = -1;
    info.convert = 0;
    info.data = 0;
    info.release = 0;
    if (unknownEncodingHandler(unknownEncodingHandlerData, encodingName, &info)) {
      ENCODING *enc;
      unknownEncodingMem = malloc(XmlSizeOfUnknownEncoding());
      if (!unknownEncodingMem) {
	if (info.release)
	  info.release(info.data);
	return XML_ERROR_NO_MEMORY;
      }
      enc = (ns
	     ? XmlInitUnknownEncodingNS
	     : XmlInitUnknownEncoding)(unknownEncodingMem,
				       info.map,
				       info.convert,
				       info.data);
      if (enc) {
	unknownEncodingData = info.data;
	unknownEncodingRelease = info.release;
	encoding = enc;
	return XML_ERROR_NONE;
      }
    }
    if (info.release)
      info.release(info.data);
  }
  return XML_ERROR_UNKNOWN_ENCODING;
}

static enum XML_Error
prologInitProcessor(XML_Parser parser,
		    const char *s,
		    const char *end,
		    const char **nextPtr)
{
  enum XML_Error result = initializeEncoding(parser);
  if (result != XML_ERROR_NONE)
    return result;
  processor = prologProcessor;
  return prologProcessor(parser, s, end, nextPtr);
}

static enum XML_Error
prologProcessor(XML_Parser parser,
		const char *s,
		const char *end,
		const char **nextPtr)
{
  const char *next;
  int tok = XmlPrologTok(encoding, s, end, &next);
  return doProlog(parser, encoding, s, end, tok, next, nextPtr);
}

static enum XML_Error
doProlog(XML_Parser parser,
	 const ENCODING *enc,
	 const char *s,
	 const char *end,
	 int tok,
	 const char *next,
	 const char **nextPtr)
{
#ifdef XML_DTD
  static const XML_Char externalSubsetName[] = { '#' , '\0' };
#endif /* XML_DTD */

  const char **eventPP;
  const char **eventEndPP;
  if (enc == encoding) {
    eventPP = &eventPtr;
    eventEndPP = &eventEndPtr;
  }
  else {
    eventPP = &(openInternalEntities->internalEventPtr);
    eventEndPP = &(openInternalEntities->internalEventEndPtr);
  }
  for (;;) {
    int role;
    *eventPP = s;
    *eventEndPP = next;
    if (tok <= 0) {
      if (nextPtr != 0 && tok != XML_TOK_INVALID) {
	*nextPtr = s;
	return XML_ERROR_NONE;
      }
      switch (tok) {
      case XML_TOK_INVALID:
	*eventPP = next;
	return XML_ERROR_INVALID_TOKEN;
      case XML_TOK_PARTIAL:
	return XML_ERROR_UNCLOSED_TOKEN;
      case XML_TOK_PARTIAL_CHAR:
	return XML_ERROR_PARTIAL_CHAR;
      case XML_TOK_NONE:
#ifdef XML_DTD
	if (enc != encoding)
	  return XML_ERROR