xmlparse.c

《unix/Linux下的Curses库开发指南》综合示例程序

  freeBindingList = 0;
  inheritedBindings = 0;
  attsSize = INIT_ATTS_SIZE;
  atts = malloc(attsSize * sizeof(ATTRIBUTE));
  nSpecifiedAtts = 0;
  dataBuf = malloc(INIT_DATA_BUF_SIZE * sizeof(XML_Char));
  groupSize = 0;
  groupConnector = 0;
  hadExternalDoctype = 0;
  unknownEncodingMem = 0;
  unknownEncodingRelease = 0;
  unknownEncodingData = 0;
  unknownEncodingHandlerData = 0;
  namespaceSeparator = '!';
  ns = 0;
  poolInit(&tempPool);
  poolInit(&temp2Pool);
  protocolEncodingName = encodingName ? poolCopyString(&tempPool, encodingName) : 0;
  if (!dtdInit(&dtd) || !atts || !dataBuf
      || (encodingName && !protocolEncodingName)) {
    XML_ParserFree(parser);
    return 0;
  }
  dataBufEnd = dataBuf + INIT_DATA_BUF_SIZE;
  XmlInitEncoding(&initEncoding, &encoding, 0);
  return parser;
}

XML_Parser XML_ParserCreateNS(const XML_Char *encodingName, XML_Char nsSep)
{
  static
  const XML_Char implicitContext[] = {
    XML_T('x'), XML_T('m'), XML_T('l'), XML_T('='),
    XML_T('h'), XML_T('t'), XML_T('t'), XML_T('p'), XML_T(':'),
    XML_T('/'), XML_T('/'), XML_T('w'), XML_T('w'), XML_T('w'),
    XML_T('.'), XML_T('w'), XML_T('3'),
    XML_T('.'), XML_T('o'), XML_T('r'), XML_T('g'),
    XML_T('/'), XML_T('X'), XML_T('M'), XML_T('L'),
    XML_T('/'), XML_T('1'), XML_T('9'), XML_T('9'), XML_T('8'),
    XML_T('/'), XML_T('n'), XML_T('a'), XML_T('m'), XML_T('e'),
    XML_T('s'), XML_T('p'), XML_T('a'), XML_T('c'), XML_T('e'),
    XML_T('\0')
  };

  XML_Parser parser = XML_ParserCreate(encodingName);
  if (parser) {
    XmlInitEncodingNS(&initEncoding, &encoding, 0);
    ns = 1;
    namespaceSeparator = nsSep;
  }
  if (!setContext(parser, implicitContext)) {
    XML_ParserFree(parser);
    return 0;
  }
  return parser;
}

int XML_SetEncoding(XML_Parser parser, const XML_Char *encodingName)
{
  if (!encodingName)
    protocolEncodingName = 0;
  else {
    protocolEncodingName = poolCopyString(&tempPool, encodingName);
    if (!protocolEncodingName)
      return 0;
  }
  return 1;
}

XML_Parser XML_ExternalEntityParserCreate(XML_Parser oldParser,
					  const XML_Char *context,
					  const XML_Char *encodingName)
{
  XML_Parser parser = oldParser;
  DTD *oldDtd = &dtd;
  XML_StartElementHandler oldStartElementHandler = startElementHandler;
  XML_EndElementHandler oldEndElementHandler = endElementHandler;
  XML_CharacterDataHandler oldCharacterDataHandler = characterDataHandler;
  XML_ProcessingInstructionHandler oldProcessingInstructionHandler = processingInstructionHandler;
  XML_CommentHandler oldCommentHandler = commentHandler;
  XML_StartCdataSectionHandler oldStartCdataSectionHandler = startCdataSectionHandler;
  XML_EndCdataSectionHandler oldEndCdataSectionHandler = endCdataSectionHandler;
  XML_DefaultHandler oldDefaultHandler = defaultHandler;
  XML_StartNamespaceDeclHandler oldStartNamespaceDeclHandler = startNamespaceDeclHandler;
  XML_EndNamespaceDeclHandler oldEndNamespaceDeclHandler = endNamespaceDeclHandler;
  XML_NotStandaloneHandler oldNotStandaloneHandler = notStandaloneHandler;
  XML_ExternalEntityRefHandler oldExternalEntityRefHandler = externalEntityRefHandler;
  XML_UnknownEncodingHandler oldUnknownEncodingHandler = unknownEncodingHandler;
  void *oldUserData = userData;
  void *oldHandlerArg = handlerArg;
  int oldDefaultExpandInternalEntities = defaultExpandInternalEntities;
  void *oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg;
 
  parser = (ns
            ? XML_ParserCreateNS(encodingName, namespaceSeparator)
	    : XML_ParserCreate(encodingName));
  if (!parser)
    return 0;
  startElementHandler = oldStartElementHandler;
  endElementHandler = oldEndElementHandler;
  characterDataHandler = oldCharacterDataHandler;
  processingInstructionHandler = oldProcessingInstructionHandler;
  commentHandler = oldCommentHandler;
  startCdataSectionHandler = oldStartCdataSectionHandler;
  endCdataSectionHandler = oldEndCdataSectionHandler;
  defaultHandler = oldDefaultHandler;
  startNamespaceDeclHandler = oldStartNamespaceDeclHandler;
  endNamespaceDeclHandler = oldEndNamespaceDeclHandler;
  notStandaloneHandler = oldNotStandaloneHandler;
  externalEntityRefHandler = oldExternalEntityRefHandler;
  unknownEncodingHandler = oldUnknownEncodingHandler;
  userData = oldUserData;
  if (oldUserData == oldHandlerArg)
    handlerArg = userData;
  else
    handlerArg = parser;
  if (oldExternalEntityRefHandlerArg != oldParser)
    externalEntityRefHandlerArg = oldExternalEntityRefHandlerArg;
  defaultExpandInternalEntities = oldDefaultExpandInternalEntities;
  if (!dtdCopy(&dtd, oldDtd) || !setContext(parser, context)) {
    XML_ParserFree(parser);
    return 0;
  }
  processor = externalEntityInitProcessor;
  return parser;
}

static
void destroyBindings(BINDING *bindings)
{
  for (;;) {
    BINDING *b = bindings;
    if (!b)
      break;
    bindings = b->nextTagBinding;
    free(b->uri);
    free(b);
  }
}

void XML_ParserFree(XML_Parser parser)
{
  for (;;) {
    TAG *p;
    if (tagStack == 0) {
      if (freeTagList == 0)
	break;
      tagStack = freeTagList;
      freeTagList = 0;
    }
    p = tagStack;
    tagStack = tagStack->parent;
    free(p->buf);
    destroyBindings(p->bindings);
    free(p);
  }
  destroyBindings(freeBindingList);
  destroyBindings(inheritedBindings);
  poolDestroy(&tempPool);
  poolDestroy(&temp2Pool);
  dtdDestroy(&dtd);
  free((void *)atts);
  free(groupConnector);
  free(buffer);
  free(dataBuf);
  free(unknownEncodingMem);
  if (unknownEncodingRelease)
    unknownEncodingRelease(unknownEncodingData);
  free(parser);
}

void XML_UseParserAsHandlerArg(XML_Parser parser)
{
  handlerArg = parser;
}

void XML_SetUserData(XML_Parser parser, void *p)
{
  if (handlerArg == userData)
    handlerArg = userData = p;
  else
    userData = p;
}

int XML_SetBase(XML_Parser parser, const XML_Char *p)
{
  if (p) {
    p = poolCopyString(&dtd.pool, p);
    if (!p)
      return 0;
    dtd.base = p;
  }
  else
    dtd.base = 0;
  return 1;
}

const XML_Char *XML_GetBase(XML_Parser parser)
{
  return dtd.base;
}

int XML_GetSpecifiedAttributeCount(XML_Parser parser)
{
  return nSpecifiedAtts;
}

void XML_SetElementHandler(XML_Parser parser,
			   XML_StartElementHandler start,
			   XML_EndElementHandler end)
{
  startElementHandler = start;
  endElementHandler = end;
}

void XML_SetCharacterDataHandler(XML_Parser parser,
				 XML_CharacterDataHandler handler)
{
  characterDataHandler = handler;
}

void XML_SetProcessingInstructionHandler(XML_Parser parser,
					 XML_ProcessingInstructionHandler handler)
{
  processingInstructionHandler = handler;
}

void XML_SetCommentHandler(XML_Parser parser,
			   XML_CommentHandler handler)
{
  commentHandler = handler;
}

void XML_SetCdataSectionHandler(XML_Parser parser,
				XML_StartCdataSectionHandler start,
			        XML_EndCdataSectionHandler end)
{
  startCdataSectionHandler = start;
  endCdataSectionHandler = end;
}

void XML_SetDefaultHandler(XML_Parser parser,
			   XML_DefaultHandler handler)
{
  defaultHandler = handler;
  defaultExpandInternalEntities = 0;
}

void XML_SetDefaultHandlerExpand(XML_Parser parser,
				 XML_DefaultHandler handler)
{
  defaultHandler = handler;
  defaultExpandInternalEntities = 1;
}

void XML_SetUnparsedEntityDeclHandler(XML_Parser parser,
				      XML_UnparsedEntityDeclHandler handler)
{
  unparsedEntityDeclHandler = handler;
}

void XML_SetNotationDeclHandler(XML_Parser parser,
				XML_NotationDeclHandler handler)
{
  notationDeclHandler = handler;
}

void XML_SetNamespaceDeclHandler(XML_Parser parser,
				 XML_StartNamespaceDeclHandler start,
				 XML_EndNamespaceDeclHandler end)
{
  startNamespaceDeclHandler = start;
  endNamespaceDeclHandler = end;
}

void XML_SetNotStandaloneHandler(XML_Parser parser,
				 XML_NotStandaloneHandler handler)
{
  notStandaloneHandler = handler;
}

void XML_SetExternalEntityRefHandler(XML_Parser parser,
				     XML_ExternalEntityRefHandler handler)
{
  externalEntityRefHandler = handler;
}

void XML_SetExternalEntityRefHandlerArg(XML_Parser parser, void *arg)
{
  if (arg)
    externalEntityRefHandlerArg = arg;
  else
    externalEntityRefHandlerArg = parser;
}

void XML_SetUnknownEncodingHandler(XML_Parser parser,
				   XML_UnknownEncodingHandler handler,
				   void *data)
{
  unknownEncodingHandler = handler;
  unknownEncodingHandlerData = data;
}

int XML_Parse(XML_Parser parser, const char *s, int len, int isFinal)
{
  if (len == 0) {
    if (!isFinal)
      return 1;
    positionPtr = bufferPtr;
    errorCode = processor(parser, bufferPtr, parseEndPtr = bufferEnd, 0);
    if (errorCode == XML_ERROR_NONE)
      return 1;
    eventEndPtr = eventPtr;
    return 0;
  }
  else if (bufferPtr == bufferEnd) {
    const char *end;
    int nLeftOver;
    parseEndByteIndex += len;
    positionPtr = s;
    if (isFinal) {
      errorCode = processor(parser, s, parseEndPtr = s + len, 0);
      if (errorCode == XML_ERROR_NONE)
	return 1;
      eventEndPtr = eventPtr;
      return 0;
    }
    errorCode = processor(parser, s, parseEndPtr = s + len, &end);
    if (errorCode != XML_ERROR_NONE) {
      eventEndPtr = eventPtr;
      return 0;
    }
    XmlUpdatePosition(encoding, positionPtr, end, &position);
    nLeftOver = s + len - end;
    if (nLeftOver) {
      if (buffer == 0 || nLeftOver > bufferLim - buffer) {
	/* FIXME avoid integer overflow */
	buffer = buffer == 0 ? malloc(len * 2) : realloc(buffer, len * 2);
	if (!buffer) {
	  errorCode = XML_ERROR_NO_MEMORY;
	  eventPtr = eventEndPtr = 0;
	  return 0;
	}
	bufferLim = buffer + len * 2;
      }
      memcpy(buffer, end, nLeftOver);
      bufferPtr = buffer;
      bufferEnd = buffer + nLeftOver;
    }
    return 1;
  }
  else {
    memcpy(XML_GetBuffer(parser, len), s, len);
    return XML_ParseBuffer(parser, len, isFinal);
  }
}

int XML_ParseBuffer(XML_Parser parser, int len, int isFinal)
{
  const char *start = bufferPtr;
  positionPtr = start;
  bufferEnd += len;
  parseEndByteIndex += len;
  errorCode = processor(parser, start, parseEndPtr = bufferEnd,
			isFinal ? (const char **)0 : &bufferPtr);
  if (errorCode == XML_ERROR_NONE) {
    if (!isFinal)
      XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
    return 1;
  }
  else {
    eventEndPtr = eventPtr;
    return 0;
  }
}

void *XML_GetBuffer(XML_Parser parser, int len)
{
  if (len > bufferLim - bufferEnd) {
    /* FIXME avoid integer overflow */
    int neededSize = len + (bufferEnd - bufferPtr);
    if (neededSize  <= bufferLim - buffer) {
      memmove(buffer, bufferPtr, bufferEnd - bufferPtr);
      bufferEnd = buffer + (bufferEnd - bufferPtr);
      bufferPtr = buffer;
    }
    else {
      char *newBuf;
      int bufferSize = bufferLim - bufferPtr;
      if (bufferSize == 0)
	bufferSize = INIT_BUFFER_SIZE;
      do {
	bufferSize *= 2;
      } while (bufferSize < neededSize);
      newBuf = malloc(bufferSize);
      if (newBuf == 0) {
	errorCode = XML_ERROR_NO_MEMORY;
	return 0;
      }
      bufferLim = newBuf + bufferSize;
      if (bufferPtr) {
	memcpy(newBuf, bufferPtr, bufferEnd - bufferPtr);
	free(buffer);
      }
      bufferEnd = newBuf + (bufferEnd - bufferPtr);
      bufferPtr = buffer = newBuf;
    }
  }
  return bufferEnd;
}

enum XML_Error XML_GetErrorCode(XML_Parser parser)
{
  return errorCode;
}

long XML_GetCurrentByteIndex(XML_Parser parser)
{
  if (eventPtr)
    return parseEndByteIndex - (parseEndPtr - eventPtr);
  return -1;
}

int XML_GetCurrentByteCount(XML_Parser parser)
{
  if (eventEndPtr && eventPtr)
    return eventEndPtr - eventPtr;
  return 0;
}

int XML_GetCurrentLineNumber(XML_Parser parser)
{
  if (eventPtr) {
    XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
    positionPtr = eventPtr;
  }
  return position.lineNumber + 1;
}

int XML_GetCurrentColumnNumber(XML_Parser parser)
{
  if (eventPtr) {
    XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
    positionPtr = eventPtr;
  }
  return position.columnNumber;
}

void XML_DefaultCurrent(XML_Parser parser)
{
  if (defaultHandler) {
    if (openInternalEntities)
      reportDefault(parser,