python中lxml的应用

首先下载lxml， http://www.lfd.uci.edu/~gohlke/pythonlibs/ ， 然后添加引用

from lxml import _elementpath as DONTUSE

from lxml import etree

具体示例：

1.添加命名空间

#set namespace

nsmap = { "xsi" : "http://www.w3.org/2001/XMLSchema-instance" }

g_statisticsRoot = etree.Element( "DcmStatistics" , nsmap = nsmap)

2.添加xml schema引用

#add xsd reference

g_statisticsRoot.set( "{http://www.w3.org/2001/XMLSchema-instance}noNamespaceSchemaLocation" , "DcmStatistics.xsd" )

3.添加注释

#add comment， 利用addprevious添加到根节点的前面

comment = etree.Comment( "create by jiangong.li" )

g_statisticsRoot.addprevious(comment)

4.尝试多种编码来解析xml

def decodingXml(xmlFile):

    tree = None

    encoding = "utf-8"

    while ( True ):

        try :

            parser = etree.XMLParser(remove_blank_text= True , encoding=encoding, remove_comments = False )

            tree = etree.parse(xmlFile, parser)

        except Exception as e:

            if (encoding != "gb18030" ):

                encoding = "gb18030"

                continue

            else :

                print( "\nPAR XML ERROR, decoding error." )

                break

        break

    return tree

5.遍历xml下的所有子节点，不止直属第一级子节点. iter()

     for element in root.iter():

          element . tail = None

6.遍历xml下的第一级子节点. iterchildren()

     for e in srcParentNode.iterchildren():

        if e is srcParentNode:

            continue

        name = ""

        #statistics node

        if   e.tag == "element" :

            name = "Element"

        elif e.tag == "sequence" :

            name = "Sequence"

        elif e.tag == "item" :

            name = "Item"

        else :

            print( "\nUnsupported element type: %s\n" %(e.tag))

            name = e.tag

            # Only parse element/sequence/item

             continue      

7.添加子节点到尾部. append()

def getXmlElement(nodeName, parentNode):

    if parentNode  == None:

        raise Exception( "parent node is None" )

    nodes = parentNode.xpath( './' +nodeName)

    if len(nodes) == 0:

        node = etree.Element(nodeName)

        parentNode.append(node)

        return node

    else :

        return nodes[0]

8.格式化成str输出

etree.tostring(g_statisticsRoot, encoding= "UTF-8" , xml_declaration= True , pretty_print= True , with_comments= True )

9.保存成xml文件

        statisticsResult =  open(g_xmlName, "bw+")

        statisticsResult.write(etree.tostring(g_statisticsRoot, encoding= "UTF-8" , xml_declaration= True , pretty_print= True , with_comments= True ))

        statisticsResult.flush()

        statisticsResult.close()