在Python中使用元素树合并xml文件

Question

我正在尝试合并两个xml文件。这些文件包含相同的总体结构，但细节不同。

文件1.xml：

<book>
    <chapter id="113">
        <sentence id="1">
            <word id="128160">
                <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPV"/>
                <Number type="S"/>
            </word>
            <word id="128161">
                <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPF"/>
            </word>
             </sentence>
             <sentence id="2">
            <word id="128162">
                <POS Tag="P"/>
                <grammar type="PREFIX"/>
                <Tag Tag="bi+"/>
            </word>
             </sentence>
        </chapter>
</book>

文件2.xml：

<book>
    <chapter id="113">
        <sentence id="1">
            <word id="128160">
            <concept English="joke"/>
            </word>
            <word id="128161">
                <concept English="romance"/>
            </word>
             </sentence>
             <sentence id="2">
            <word id="128162">
                <concept English="happiness"/>
            </word>
             </sentence>
        </chapter>
</book>

所需输出为：

<book>
    <chapter id="113">
        <sentence id="1">
            <word id="128160">
                    <concept English="joke"/>
                    <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPV"/>
                <Number type="S"/>
            </word>
            <word id="128161">
                <concept English="romance"/>
                <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPF"/>
            </word>
             </sentence>
             <sentence id="2">
            <word id="128162">
                <concept English="happiness"/>
                <POS Tag="P"/>
                <grammar type="PREFIX"/>
                <Tag Tag="bi+"/>
            </word>
             </sentence>
        </chapter>
</book>

好吧，我试着在path中这样做，但没有得到所需的输出：

import os, os.path, sys
import glob
from xml.etree import ElementTree

output = open('merge.xml','w')
files="sample"
xml_files = glob.glob(files +"/*.xml")
xml_element_tree = None
for xml_file in xml_files:
        data = ElementTree.parse(xml_file).getroot()
        # print ElementTree.tostring(data)
        for word in data.iter('word'):
            if xml_element_tree is None:
                xml_element_tree = data 
                insertion_point = xml_element_tree.findall("book/chapter/sentence/word/*")
            else:
                insertion_point.extend(word) 
if xml_element_tree is not None:
        print>>output, ElementTree.tostring(xml_element_tree)

拜托，有什么帮助吗

Answer 1

我以前做过类似事情的一种方法是创建一个xml文档，然后附加您要查找的值。我不相信有办法把他们“合并”

xml = ET.fromstring("<book></book>")
document = ET.parse(tempFile)
childNodeList = document.findall(xpathQuery)
for node in childNodeList: 
   xml.append(node)

Answer 2

这是一个解决方案。从一个空的合并文档开始，然后在枚举文件时，将找不到的元素添加到合并文档中。你可以概括这一点，但这是第一步：

import lxml.etree
merged = lxml.etree.Element('book')
for xml_file in xml_files:
    for merge_chapter in lxml.etree.parse(xml_file):
        try:
            chapter = merged.xpath('chapter[@id=%s]' % merge_chapter.get('id'))[0]
            for merge_sentence in merge_chapter:
                try:
                    sentence = chapter.xpath('sentence[@id=%s]' % merge_sentence.get('id'))[0]
                    for merge_word in merge_sentence:
                        try:
                            word = sentence.xpath('word[@id=%s]' % merge_word.get('id'))[0]
                            for data in merge_word:
                                try:
                                    word.xpath(data.tag)[0]
                                except IndexError:
                                    # add newly discovered word data
                                    word.append(data)
                        except IndexError:
                            # add newly discovered word
                            sentence.append(merge_word)
                except IndexError:
                    # add newly discovered sentence
                    chapter.append(merge_sentence)
        except IndexError:
            # add newly discovered chapter
            merged.append(merge_chapter)

Answer 3

如果希望将File2合并到File1中，则可以循环遍历File2中的所有元素，然后将属性从File2的元素复制到File1的元素中。

在我正在做的一个项目中，我必须做类似的事情。这是我目前的解决方案，应该在Python2.7下工作。

注意，我进一步添加了在公共节点之间复制属性的需求。您将看到我将以下属性添加到：

• 鼓声
• bass='Geddy'

然后我又加了一句：

• 吉他='Alex'

最后合并的文件中有三名权力三人组成员。

我还添加了<sentance id='3'/>来证明元素的顺序不再重要。

#!/usr/bin/python
from lxml import etree 
from copy import deepcopy
import lxml

xmlA='''
<book>
    <chapter id="113">

        <sentence id="1" drums='Neil'>
            <word id="128160" bass='Geddy'>
                <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPV"/>
                <Number type="S"/>
            </word>
            <word id="128161">
                <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPF"/>
            </word>
        </sentence>

        <sentence id="2">
            <word id="128162">
                <POS Tag="P"/>
                <grammar type="PREFIX"/>
                <Tag Tag="bi+"/>
            </word>
        </sentence>

    </chapter>
</book>
'''

xmlB='''
<book>
    <chapter id="113">

        <sentence id="3">
            <word id="128168">
                <concept English="sadness"/>
            </word>
        </sentence>

        <sentence id="1">
            <word id="128160">
                <concept English="joke"/>
            </word>
            <word id="128161">
                <concept English="romance"/>
            </word>
        </sentence>

        <sentence id="2" guitar='Alex'>
            <word id="128162">
                <concept English="happiness"/>
            </word>
        </sentence>


    </chapter>
</book>
'''

import re
from copy import deepcopy

##
#   @brief  Translates the relational xpath to an explicit xpath.
#   In the XML examples above, getpath will return the following for 
#   <sentance id='1'/>:
#       - xmlA = /book/chapter/sentance[1]
#       - xmlb = /book/chapter/sentance[2]
#
#   A path that is explicit in both document would be:
#       - xmlA = /book/chapter/sentance[@id='1']
#       - xmlb = /book/chapter/sentance[@id='1']
#
def convertXpath(element):
    newPath = ''
    tree    = element.getroottree()
    path    = tree.getpath(element).split('/')
    root    = tree.getroot()

    for p in path:
        if p == '':
            continue

        if re.search('\[[0-9]*\]', p):

            # Get the element at this path
            #
            node = root.xpath(newPath+'/'+p)[0]
            id=node.get('id')

            p=re.sub('\[[0-9]*\]','', p)
            newPath += '/'+p+"[@id='"+id+"']"

        else:
            newPath+='/'+p

    return newPath



def mergeXml(a,b):

    for node in a.nodes():
        path = convertXpath(node)

        # find the element in the other document
        #
        elements =  b.root.xpath(path)

        for e in elements:
            for name, value in node.items():
                if name == 'id':
                    continue
                e.set(name,value)

        if len(elements) == 0:
            # Add the node to other document
            #
            newElement = deepcopy(node)

            # Find the path to the parent
            #
            parent = node.getparent()
            path = convertXpath(parent)

            bParent = b.root.xpath(path)[0]
            bParent.append(newElement)

class XmlDoc:
    def __init__(self, xml):
        self.root = etree.fromstring(xml)
        self.tree = self.root.getroottree()

    def __str__(self):
        return etree.tostring(self.root, pretty_print=True)

    def nodes(self):
        return self.root.iter('*')



if __name__ == '__main__':
    a = XmlDoc(xmlA)
    b = XmlDoc(xmlB)

    mergeXml(a,b)
    print b

这将产生以下输出：

<book>
    <chapter id="113">

        <sentence id="3">
            <word id="128168">
                <concept English="sadness"/>
            </word>
        </sentence>

        <sentence id="1" drums="Neil">
            <word id="128160" bass="Geddy">
                <concept English="joke"/>
            <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPV"/>
                <Number type="S"/>
            </word>
            <word id="128161">
                <concept English="romance"/>
            <POS Tag="V"/>
                <grammar type="STEM"/>
                <Aspect type="IMPF"/>
            </word>
        </sentence>

        <sentence id="2" guitar="Alex">
            <word id="128162">
                <concept English="happiness"/>
            <POS Tag="P"/>
                <grammar type="PREFIX"/>
                <Tag Tag="bi+"/>
            </word>
        </sentence>


    </chapter>
</book>