使用Python处理XML
很久没有写博文了,也很久没有碰Python了。刚好工作需要,小小动手了下。因为是在新机器上,所以就直接装了Python 3,反正以后也是要适应,不如早点。
在使用Python处理XML的问题上,首先遇到的是编码问题。
Python并不支持gb2312,所以面对encoding="gb2312"的XML文件会出现错误。Python读取的文件本身的编码也可能导致抛出异常,这种情况下打开文件的时候就需要指定编码。此外就是XML中节点所包含的中文。
我这里呢,处理就比较简单了,只需要修改XML的encoding头部。
#!/usr/bin/env pythonimport os, sysimport redef replaceXmlEncoding(filepath, oldEncoding='gb2312', newEncoding='utf-8'):f = open(filepath, mode='r')content = f.read()content = re.sub(oldEncoding, newEncoding, content)f.close()f = open(filepath, mode='w')f.write(content)f.close()if __name__ == "__main__":replaceXmlEncoding('./ActivateAccount.xml')
接着是使用xml.etree.ElementTree来操作XML文件。 在一个类里面定义__call__函数可以使得该类可调用,比如下面代码的最后几行,在__main__函数中。这也很突出地体现了在Python的世界里,一切都是对象,包括对象本身 :)
一直觉得__main__函数用来测试真是蛮好用的。
#!/usr/bin/env pythonimport os, reimport xml.etree.ElementTree as etreeLocale_Path = "./locale.txt"class xmlExtractor(object):def __init__(self):passdef __call__(self, filepath):retDict = {}f = open(filepath, 'r')Line = len(open(filepath, 'r').readlines())retDict['Line'] = Linetree = etree.parse(f)root = tree.find("ResItem")Id = root.get("ID")retDict['Title'] = IdresItemCnt = len(list(root.findall("ResItem"))) + 1retDict['ResItemCount'] = resItemCntretDict['ChineseTip'] = 'None'for child in root:attrDict = child.attribkeyword = "Name"if(keyword in attrDict.keys() and attrDict['Name'] == "Caption"):if len(child.attrib['Value']) > 1:if child.attrib['Value'] == '~':title = child.attrib['Value']else:title = child.attrib['Value']#print(title)chs = open(Locale_Path).read()pattern = '<String id="' + title + '">[^>]+>'m = re.search(pattern, chs)if m != None:realTitle = re.sub('<[^>]+>', '', m.group(0))retDict['ChineseTip'] = realTitlef.close()return retDictif __name__ == "__main__":fo = xmlExtractor()d = fo('./ActivateAccount.xml')print(d)
最后,就是入口文件,导入上面两个文件,使用xml.dom和os.listdir来递归处理XML文件,并生成一个结果集。 一直觉得Python的UnboundLocalError错误挺有意思的,不知道是不是符号表的覆盖问题。
#!/usr/bin/env pythonfrom xmlExtractor import *from replaceXmlEncoding import *from xml.dom import minidom,Nodedoc = minidom.Document()extractor = xmlExtractor()totalLines = 0totalResItemCnt = 0totalXmlFileCnt = 0totalErrorCnt = 0errorFileList = []xmlRoot = doc.createElement("XmlResourceFile")doc.appendChild(xmlRoot)def myWalkDir(level, path):global doc, extractor, totalLines, totalResItemCnt, totalXmlFileCntglobal totalErrorCnt, errorFileListglobal xmlRootfor i in os.listdir(path):if i[-3:] == 'xml':totalXmlFileCnt += 1try:#先把xml的encoding由gb2312转换为utf-8replaceXmlEncoding(path + '\\' + i)#再提取xml文档中需要的信息info = extractor(path + '\\' + i)#在上述两行代码没有出现异常的基础上再创建节点#print(info)#print(type(i))xmlNode = doc.createElement("XmlFile")xmlRoot.appendChild(xmlNode)xmlName = doc.createElement("Filename")xmlName.setAttribute('Value', i)#xmlName.appendChild(doc.createTextNode(i))xmlNode.appendChild(xmlName)filePath = doc.createElement("Filepath")filePath.setAttribute('Value', path)#filePath.appendChild(doc.createTextNode(path))xmlNode.appendChild(filePath)titleNode = doc.createElement("Title")titleNode.setAttribute('Value', str(info['Title']))#titleNode.appendChild(doc.createTextNode(str(info['Title'])))xmlNode.appendChild(titleNode)chsNode = doc.createElement("ChineseTip")chsNode.setAttribute('Value', str(info['ChineseTip']))#chsNode.appendChild(doc.createTextNode(str(info['Chinese'])))xmlNode.appendChild(chsNode)resItemNode = doc.createElement("ResItemCount")resItemNode.setAttribute('Value', str(info['ResItemCount']))#resItemNode.appendChild(doc.createTextNode(str(info['ResItemCount'])))xmlNode.appendChild(resItemNode)lineNode = doc.createElement("LineCount")lineNode.setAttribute('Value', str(info['Line']))#lineNode.appendChild(doc.createTextNode(str(info['Line'])))xmlNode.appendChild(lineNode)descNode = doc.createElement("Description")descNode.setAttribute('Value', '')#descNode.appendChild(doc.createTextNode(''))xmlNode.appendChild(descNode)except Exception as errorDetail:totalErrorCnt += 1errorFileList.append(path + '\\' + i)print(path + '\\' + i, errorDetail)if os.path.isdir(path + '\\' + i):myWalkDir(level+1, path + '\\' + i)if __name__ == "__main__":path = os.getcwd() + '\\themes'myWalkDir(0, path)print(totalXmlFileCnt, totalErrorCnt)#print(doc.toprettyxml(indent = " "))resultXml = open("./xmlResourceList.xml", "w")resultXml.write(doc.toprettyxml(indent = " "))resultXml.close()
页:
[1]