libxml库之xmlParseMemory和xmlDocDumpFormatMemory函数
两个函数，怎么在内存中处理xml文件。有个疑问，xmlChar *xmlbuff;这样系统第一次分配了xmlbuff的大小以后。对xml做修改，可以成功吗。
有空试试。
函数原型：
xmlParseMemory xmlDocPtr xmlParseMemory (const char * buffer, int size) parse an XML in-memory block and build a tree. buffer: an pointer to a char array size: the size of the array Returns: the resulting document tree Function: xmlParserAddNodeIn
解析xml字符串 xmlDocPtr doc = xmlParseMemory(pXml, length); //根据xmldoc获得xml的根节点 xmlNodePtr cur = xmlDocGetRootElement(doc); //获得子节点:->children获得不是第一个子节点，必须用next才能获得第一个子节点 cur = cur->children; cur = cur->next; // 获得节点信息中的内容: 注意释放资源 xmlChar* key = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); xmlFree(key); //获得节点信息属性的值：属性name，注意释放资源 xmlChar* fversion = xmlGetProp(cur, "version"); xmlFree(fversion); //根节点相关函数 xmlNodePtr xmlDocGetRootElement (xmlDocPtr doc) //获取文档根节点 xmlNodePtr xmlDocSetRootElement (xmlDocPtr doc, xmlNodePtr root) //设置文档根节点 //创建子节点相关函数 xmlNodePtr xmlNewNode (xmlNsPtr ns, const xmlChar * name) //创建新节点 xmlNodePtr xmlNewChild (xmlNodePtr parent, xmlNsPtr ns, const xmlChar * name, const xmlChar * content) //创建新的子节点 xmlNodePtr xmlCopyNode (const xmlNodePtr node, int extended) //复制当前节点 //添加子节点相关函数 xmlNodePtr xmlAddChild (xmlNodePtr parent, xmlNodePtr cur) //给指定节点添加子节点 xmlNodePtr xmlAddNextSibling (xmlNodePtr cur, xmlNodePtr elem) //添加后一个兄弟节点 xmlNodePtr xmlAddPrevSibling (xmlNodePtr cur, xmlNodePtr elem) //添加前一个兄弟节点 xmlNodePtr xmlAddSibling (xmlNodePtr cur, xmlNodePtr elem) //添加兄弟节点 //属性相关函数 xmlAttrPtr xmlNewProp (xmlNodePtr node, const xmlChar * name, const xmlChar * value) //创建新节点属性 xmlChar * xmlGetProp (xmlNodePtr node, const xmlChar * name) //读取节点属性 xmlAttrPtr xmlSetProp (xmlNodePtr node, const xmlChar * name, const xmlChar * value) //设置节点属性 =xmlNodeListGetstring(doc, cur->xmlChildrenNode, 1); =xmlNodeContent(cur);
===========================================================================================
函数原型：
Function: xmlDocDumpFormatMemory void xmlDocDumpFormatMemory (xmlDocPtr cur, xmlChar ** mem, int * size, int format) Dump an XML document in memory and return the #xmlChar * and it's size. It's up to the caller to free the memory with xmlFree(). Note that @format = 1 provide node indenting only if xmlIndentTreeOutput = 1 or xmlKeepBlanksDefault(0) was called cur: the document mem: OUT: the memory pointer size: OUT: the memory length format: should formatting spaces been added
程序示例：
/**
* section: InputOutput
* synopsis: Output to char buffer
* purpose: Demonstrate the use of xmlDocDumpMemory
* to output document to a character buffer
* usage: io2
* test: io2 > io2.tmp ; diff io2.tmp io2.res ; rm -f io2.tmp
* author: John Fleck
* copy: see Copyright for the status of this software.
*/
#include <libxml/parser.h>
#if defined(LIBXML_TREE_ENABLED) && defined(LIBXML_OUTPUT_ENABLED)
int
main(void)
{
xmlNodePtr n;
xmlDocPtr doc;
xmlChar *xmlbuff;
int buffersize;
/*
* Create the document.
*/
doc = xmlNewDoc(BAD_CAST "1.0");
n = xmlNewNode(NULL, BAD_CAST "root");
xmlNodeSetContent(n, BAD_CAST "content");
xmlDocSetRootElement(doc, n);
/*
* Dump the document to a buffer and print it
* for demonstration purposes.
*/
xmlDocDumpFormatMemory(doc, &xmlbuff, &buffersize, 1);
printf("%s", (char *) xmlbuff);
/*
* Free associated memory.
*/
xmlFree(xmlbuff);
xmlFreeDoc(doc);
return (0);
}
#else
#include <stdio.h>
int
main(void)
{
fprintf(stderr,
"library not configured with tree and output support\n");
return (1);
}
#endif
来源：
http://xmlsoft.org/examples/io2.c //libxml官方网站
http://blog.163.com/miky_sun/blog/static/3369405201041942853395/