htmltree.c

xml开源解析代码.版本为libxml2-2.6.29,可支持GB3212.网络消息发送XML时很有用.

    memset(outbuf, 0, (size_t) sizeof(xmlOutputBuffer));
    outbuf->buffer = buf;
    outbuf->encoder = NULL;
    outbuf->writecallback = NULL;
    outbuf->closecallback = NULL;
    outbuf->context = NULL;
    outbuf->written = 0;

    use = buf->use;
    htmlNodeDumpFormatOutput(outbuf, doc, cur, NULL, format);
    xmlFree(outbuf);
    ret = buf->use - use;
    return (ret);
}

/**
 * htmlNodeDump:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the current node
 *
 * Dump an HTML node, recursive behaviour,children are printed too,
 * and formatting returns are added.
 *
 * Returns the number of byte written or -1 in case of error
 */
int
htmlNodeDump(xmlBufferPtr buf, xmlDocPtr doc, xmlNodePtr cur) {
    xmlInitParser();

    return(htmlNodeDumpFormat(buf, doc, cur, 1));
}

/**
 * htmlNodeDumpFileFormat:
 * @out:  the FILE pointer
 * @doc:  the document
 * @cur:  the current node
 * @encoding: the document encoding
 * @format:  should formatting spaces been added
 *
 * Dump an HTML node, recursive behaviour,children are printed too.
 *
 * TODO: if encoding == NULL try to save in the doc encoding
 *
 * returns: the number of byte written or -1 in case of failure.
 */
int
htmlNodeDumpFileFormat(FILE *out, xmlDocPtr doc,
	               xmlNodePtr cur, const char *encoding, int format) {
    xmlOutputBufferPtr buf;
    xmlCharEncodingHandlerPtr handler = NULL;
    int ret;

    xmlInitParser();

    if (encoding != NULL) {
	xmlCharEncoding enc;

	enc = xmlParseCharEncoding(encoding);
	if (enc != XML_CHAR_ENCODING_UTF8) {
	    handler = xmlFindCharEncodingHandler(encoding);
	    if (handler == NULL)
		return(-1);
	}
    }

    /*
     * Fallback to HTML or ASCII when the encoding is unspecified
     */
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("HTML");
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("ascii");

    /* 
     * save the content to a temp buffer.
     */
    buf = xmlOutputBufferCreateFile(out, handler);
    if (buf == NULL) return(0);

    htmlNodeDumpFormatOutput(buf, doc, cur, encoding, format);

    ret = xmlOutputBufferClose(buf);
    return(ret);
}

/**
 * htmlNodeDumpFile:
 * @out:  the FILE pointer
 * @doc:  the document
 * @cur:  the current node
 *
 * Dump an HTML node, recursive behaviour,children are printed too,
 * and formatting returns are added.
 */
void
htmlNodeDumpFile(FILE *out, xmlDocPtr doc, xmlNodePtr cur) {
    htmlNodeDumpFileFormat(out, doc, cur, NULL, 1);
}

/**
 * htmlDocDumpMemoryFormat:
 * @cur:  the document
 * @mem:  OUT: the memory pointer
 * @size:  OUT: the memory length
 * @format:  should formatting spaces been added
 *
 * Dump an HTML document in memory and return the xmlChar * and it's size.
 * It's up to the caller to free the memory.
 */
void
htmlDocDumpMemoryFormat(xmlDocPtr cur, xmlChar**mem, int *size, int format) {
    xmlOutputBufferPtr buf;
    xmlCharEncodingHandlerPtr handler = NULL;
    const char *encoding;

    xmlInitParser();

    if ((mem == NULL) || (size == NULL))
        return;
    if (cur == NULL) {
	*mem = NULL;
	*size = 0;
	return;
    }

    encoding = (const char *) htmlGetMetaEncoding(cur);

    if (encoding != NULL) {
	xmlCharEncoding enc;

	enc = xmlParseCharEncoding(encoding);
	if (enc != cur->charset) {
	    if (cur->charset != XML_CHAR_ENCODING_UTF8) {
		/*
		 * Not supported yet
		 */
		*mem = NULL;
		*size = 0;
		return;
	    }

	    handler = xmlFindCharEncodingHandler(encoding);
	    if (handler == NULL) {
		*mem = NULL;
		*size = 0;
		return;
	    }
	} else {
	    handler = xmlFindCharEncodingHandler(encoding);
	}
    }

    /*
     * Fallback to HTML or ASCII when the encoding is unspecified
     */
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("HTML");
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("ascii");

    buf = xmlAllocOutputBuffer(handler);
    if (buf == NULL) {
	*mem = NULL;
	*size = 0;
	return;
    }

	htmlDocContentDumpFormatOutput(buf, cur, NULL, format);

    xmlOutputBufferFlush(buf);
    if (buf->conv != NULL) {
	*size = buf->conv->use;
	*mem = xmlStrndup(buf->conv->content, *size);
    } else {
	*size = buf->buffer->use;
	*mem = xmlStrndup(buf->buffer->content, *size);
    }
    (void)xmlOutputBufferClose(buf);
}

/**
 * htmlDocDumpMemory:
 * @cur:  the document
 * @mem:  OUT: the memory pointer
 * @size:  OUT: the memory length
 *
 * Dump an HTML document in memory and return the xmlChar * and it's size.
 * It's up to the caller to free the memory.
 */
void
htmlDocDumpMemory(xmlDocPtr cur, xmlChar**mem, int *size) {
	htmlDocDumpMemoryFormat(cur, mem, size, 1);
}


/************************************************************************
 *									*
 *   		Dumping HTML tree content to an I/O output buffer	*
 *									*
 ************************************************************************/

void xmlNsListDumpOutput(xmlOutputBufferPtr buf, xmlNsPtr cur);

/**
 * htmlDtdDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @encoding:  the encoding string
 * 
 * TODO: check whether encoding is needed
 *
 * Dump the HTML document DTD, if any.
 */
static void
htmlDtdDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
	          const char *encoding ATTRIBUTE_UNUSED) {
    xmlDtdPtr cur = doc->intSubset;

    if (cur == NULL) {
	htmlSaveErr(XML_SAVE_NO_DOCTYPE, (xmlNodePtr) doc, NULL);
	return;
    }
    xmlOutputBufferWriteString(buf, "<!DOCTYPE ");
    xmlOutputBufferWriteString(buf, (const char *)cur->name);
    if (cur->ExternalID != NULL) {
	xmlOutputBufferWriteString(buf, " PUBLIC ");
	xmlBufferWriteQuotedString(buf->buffer, cur->ExternalID);
	if (cur->SystemID != NULL) {
	    xmlOutputBufferWriteString(buf, " ");
	    xmlBufferWriteQuotedString(buf->buffer, cur->SystemID);
	} 
    }  else if (cur->SystemID != NULL) {
	xmlOutputBufferWriteString(buf, " SYSTEM ");
	xmlBufferWriteQuotedString(buf->buffer, cur->SystemID);
    }
    xmlOutputBufferWriteString(buf, ">\n");
}

/**
 * htmlAttrDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the attribute pointer
 * @encoding:  the encoding string
 *
 * Dump an HTML attribute
 */
static void
htmlAttrDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc, xmlAttrPtr cur,
	           const char *encoding ATTRIBUTE_UNUSED) {
    xmlChar *value;

    /*
     * TODO: The html output method should not escape a & character
     *       occurring in an attribute value immediately followed by
     *       a { character (see Section B.7.1 of the HTML 4.0 Recommendation).
     */

    if (cur == NULL) {
	return;
    }
    xmlOutputBufferWriteString(buf, " ");
    if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
        xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
	xmlOutputBufferWriteString(buf, ":");
    }
    xmlOutputBufferWriteString(buf, (const char *)cur->name);
    if ((cur->children != NULL) && (!htmlIsBooleanAttr(cur->name))) {
	value = xmlNodeListGetString(doc, cur->children, 0);
	if (value) {
	    xmlOutputBufferWriteString(buf, "=");
	    if ((cur->ns == NULL) && (cur->parent != NULL) &&
		(cur->parent->ns == NULL) &&
		((!xmlStrcasecmp(cur->name, BAD_CAST "href")) ||
	         (!xmlStrcasecmp(cur->name, BAD_CAST "action")) ||
		 (!xmlStrcasecmp(cur->name, BAD_CAST "src")) ||
		 ((!xmlStrcasecmp(cur->name, BAD_CAST "name")) &&
		  (!xmlStrcasecmp(cur->parent->name, BAD_CAST "a"))))) {
		xmlChar *escaped;
		xmlChar *tmp = value;

		while (IS_BLANK_CH(*tmp)) tmp++;

		escaped = xmlURIEscapeStr(tmp, BAD_CAST"@/:=?;#%&,+");
		if (escaped != NULL) {
		    xmlBufferWriteQuotedString(buf->buffer, escaped);
		    xmlFree(escaped);
		} else {
		    xmlBufferWriteQuotedString(buf->buffer, value);
		}
	    } else {
		xmlBufferWriteQuotedString(buf->buffer, value);
	    }
	    xmlFree(value);
	} else  {
	    xmlOutputBufferWriteString(buf, "=\"\"");
	}
    }
}

/**
 * htmlAttrListDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the first attribute pointer
 * @encoding:  the encoding string
 *
 * Dump a list of HTML attributes
 */
static void
htmlAttrListDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc, xmlAttrPtr cur, const char *encoding) {
    if (cur == NULL) {
	return;
    }
    while (cur != NULL) {
        htmlAttrDumpOutput(buf, doc, cur, encoding);
	cur = cur->next;
    }
}



/**
 * htmlNodeListDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the first node
 * @encoding:  the encoding string
 * @format:  should formatting spaces been added
 *
 * Dump an HTML node list, recursive behaviour,children are printed too.
 */
static void
htmlNodeListDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
	               xmlNodePtr cur, const char *encoding, int format) {
    if (cur == NULL) {
	return;
    }
    while (cur != NULL) {
        htmlNodeDumpFormatOutput(buf, doc, cur, encoding, format);
	cur = cur->next;
    }
}

/**
 * htmlNodeDumpFormatOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the current node
 * @encoding:  the encoding string
 * @format:  should formatting spaces been added
 *
 * Dump an HTML node, recursive behaviour,children are printed too.
 */
void
htmlNodeDumpFormatOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
	                 xmlNodePtr cur, const char *encoding, int format) {
    const htmlElemDesc * info;

    xmlInitParser();

    if ((cur == NULL) || (buf == NULL)) {
	return;
    }
    /*
     * Special cases.
     */
    if (cur->type == XML_DTD_NODE)
	return;
    if ((cur->type == XML_HTML_DOCUMENT_NODE) ||
        (cur->type == XML_DOCUMENT_NODE)){
	htmlDocContentDumpOutput(buf, (xmlDocPtr) cur, encoding);
	return;
    }
    if (cur->type == XML_ATTRIBUTE_NODE) {
        htmlAttrDumpOutput(buf, doc, (xmlAttrPtr) cur, encoding);
	return;
    }
    if (cur->type == HTML_TEXT_NODE) {
	if (cur->content != NULL) {
	    if (((cur->name == (const xmlChar *)xmlStringText) ||
		 (cur->name != (const xmlChar *)xmlStringTextNoenc)) &&
		((cur->parent == NULL) ||
		 ((xmlStrcasecmp(cur->parent->name, BAD_CAST "script")) &&
		  (xmlStrcasecmp(cur->parent->name, BAD_CAST "style"))))) {
		xmlChar *buffer;

		buffer = xmlEncodeEntitiesReentrant(doc, cur->content);
		if (buffer != NULL) {
		    xmlOutputBufferWriteString(buf, (const char *)buffer);
		    xmlFree(buffer);
		}
	    } else {
		xmlOutputBufferWriteString(buf, (const char *)cur->content);
	    }
	}
	return;
    }
    if (cur->type == HTML_COMMENT_NODE) {
	if (cur->content != NULL) {
	    xmlOutputBufferWriteString(buf, "<!--");
	    xmlOutputBufferWriteString(buf, (const char *)cur->content);
	    xmlOutputBufferWriteString(buf, "-->");
	}