（python功能定制）复杂的xml文件对比，产生HTML展示区别

功能的设计初衷：
　　处理复杂的xml对比，屏蔽同节点先后顺序的影响

主要涉及知识点：

1、ElementTree库 ------- xml解析：

• • 导入ElementTree，import xml.etree.ElementTree as ET
  • 解析Xml文件找到根节点：
  • 直接解析XML文件并获得根节点，tree = ET.parse('country_data.xml') root = tree.getroot()
  • 解析字符串，root = ET.fromstring(country_data_as_string)
  • 遍历根节点可以获得子节点，然后就可以根据需求拿到需要的字段了，如：<APP_KEY channel = 'CSDN'> hello123456789 </APP_KEY>

• • • tag，即标签，用于标识该元素表示哪种数据，即APP_KEY
    • attrib，即属性，用Dictionary形式保存，即{'channel' = 'CSDN'}
    • text，文本字符串，可以用来存储一些数据，即hello123456789
    • tail，尾字符串，并不是必须的，例子中没有包含。

2、difflib库 ------- 提供的类和方法用来进行序列的差异化比较，它能够比对文件并生成差异结果文本或者html格式的差异化比较页面

这里使用了类difflib.HtmlDiff，用来创建一个html表格展示文件差异，他既可以进行全文本展示，也可以只展示上下文不同。

其构造函数如下：

__init__(tabsize=8, wrapcolumn=None, linejunk=None, charjunk=IS_CHARACTER_JUNK)

• • • tabsize表示制表符代表的空格个数，默认为8
    • wrapcolumn,可选参数，用来设置多少个字符时自动换行，默认None,为None时表示不自动换行（重点：可以让html显示更美观）
    • linejunk 和 charjunk，可选参数，在ndiff()中使用，

公共方法（生成一个包含表格的html文件，其内容是用来展示差异）：

make_file(fromlines, tolines [, fromdesc][, todesc][, context][, numlines])

• • • fromlines 和tolines,用于比较的内容，格式为字符串组成的列表
    • fromdesc 和 todesc，可选参数，对应的fromlines,tolines的差异化文件的标题，默认为空字符串
    • context 和 numlines，可选参数，context 为True时，只显示差异的上下文，为false，显示全文，numlines默认为5，当context为True时，控制展示上下文的行数，当context为false时,控制不同差异的高亮之间移动时“next”的开始位置（如果设置为0，当移动懂顶端时，超链接会丢失引用地址）

3、platform库 -------- 获取当前系统

4、logger库 -------- 如果使用robot framework，可以看到明显区别，可以定制日志log显示

robot framework的体验还不错，大概是因为其测试报告已经可以满足正常需要，很少有人会想去修改或者增加自己想要展示的内容，比如增加一个超链接，展示更多的内容，所以这部分花了很长时间均没有在网上找到相关资料，最后只能阅读源码。

遗憾与待优化：

　　其中有一部分内容，原先准备采用自循环的方式处理，但是过程中的数据传输逻辑容易错乱，以后会考虑把这部分优化一下。

##############################以下是代码部分，附件文件可以拖到本地执行并查看结果##################################################

 # coding=utf-8
 import re
 import xml.etree.ElementTree as ET  #解析xml的库
 import difflib   #文件对比库
 import datetime  #时间库
 import platform  #获取系统的库window、linux...
 import os
 from robot.api import logger    #不需要的话可以注释掉：robot framework框架脚本运行时会产生日志，可以利用这个库定制log
 
 # listafter：将解析后的xml，转换成按序排列的list：（tag，attrib，（tag，attrib，text））
 # 此方法是被下面一个方法xmltolist()调用的，想知道具体结果，可以使用下面的方法打印解析后的结果
 def listafter(listcom1):
     listcomarr1 = []
     text1 = []
     listcomarr1.append(listcom1.tag)
     listcomarr1.append(listcom1.attrib)
     if len(listcom1) > 0:
         for listcom2 in listcom1:
             listcomarr2 = []
             text2 = []
             listcomarr2.append(listcom2.tag)
             listcomarr2.append(listcom2.attrib)
             if len(listcom2) > 0:
                 for listcom3 in listcom2:
                     listcomarr3 = []
                     text3 = []
                     listcomarr3.append(listcom3.tag)
                     listcomarr3.append(listcom3.attrib)
                     if len(listcom3) > 0:
                         for listcom4 in listcom3:
                             listcomarr4 = []
                             text4 = []
                             listcomarr4.append(listcom4.tag)
                             listcomarr4.append(listcom4.attrib)
                             if len(listcom4) > 0:
                                 for listcom5 in listcom4:
                                     listcomarr5 = []
                                     text5 = []
                                     listcomarr5.append(listcom5.tag)
                                     listcomarr5.append(listcom5.attrib)
                                     if len(listcom5) > 0:
                                         for listcom6 in listcom5:
                                             listcomarr6 = []
                                             text6 = []
                                             listcomarr6.append(listcom6.tag)
                                             listcomarr6.append(listcom6.attrib)
                                             if len(listcom6) > 0:
                                                 for listcom7 in listcom6:
                                                     listcomarr7 = []
                                                     text7 = []
                                                     listcomarr7.append(listcom7.tag)
                                                     listcomarr7.append(listcom7.attrib)
                                                     if len(listcom7) > 0:
                                                         for listcom8 in listcom7:
                                                             listcomarr8 = []
                                                             text8 = []
                                                             listcomarr8.append(listcom8.tag)
                                                             listcomarr8.append(listcom8.attrib)
                                                             if len(listcom8) > 0:
                                                                 for listcom9 in listcom8:
                                                                     listcomarr9 = []
                                                                     text9 = []
                                                                     listcomarr9.append(listcom9.tag)
                                                                     listcomarr9.append(listcom9.attrib)
                                                                     # Start：判断是否需要继续递归
                                                                     if len(listcom9) > 0:
                                                                         for listcom10 in listcom9:
                                                                             listcomarr10 = []
                                                                             text10 = []
                                                                             listcomarr10.append(listcom10.tag)
                                                                             listcomarr10.append(listcom10.attrib)
                                                                             listcomarr10.append([listcom10.text])
                                                                             text9.append(listcomarr10)
                                                                     else:
                                                                         text9.append(listcom9.text)
                                                                     # End：判断是否需要继续递归
                                                                     # list二维数组排序
                                                                     text9 = sorted(text9)
                                                                     listcomarr9.append(text9)
                                                                     text8.append(listcomarr9)
                                                             else:
                                                                 text8.append(listcom8.text)
                                                             text8 = sorted(text8)
                                                             listcomarr8.append(text8)
                                                             text7.append(listcomarr8)
                                                     else:
                                                         text7.append(listcom7.text)
                                                     text7 = sorted(text7)
                                                     listcomarr7.append(text7)
                                                     text6.append(listcomarr7)
                                             else:
                                                 text6.append(listcom6.text)
                                             text6 = sorted(text6)
                                             listcomarr6.append(text6)
                                             text5.append(listcomarr6)
                                     else:
                                         text5.append(listcom5.text)
                                     text5 = sorted(text5)
                                     listcomarr5.append(text5)
                                     text4.append(listcomarr5)
                             else:
                                 text4.append(listcom4.text)
                             text4 = sorted(text4)
                             listcomarr4.append(text4)
                             text3.append(listcomarr4)
                     else:
                         text3.append(listcom3.text)
                     text3 = sorted(text3)
                     listcomarr3.append(text3)
                     text2.append(listcomarr3)
             else:
                 text2.append(listcom2.text)
             text2 = sorted(text2)
             listcomarr2.append(text2)
             text1.append(listcomarr2)
     else:
         text1.append(listcom1.text)
     text1 = sorted(text1)
     listcomarr1.append(text1)
     return listcomarr1
 
 # 将xml内容转换成按序排列的list，返回值有3个:处理后的spmlxmllist、不需要处理的头部spmlstart、不需要处理的尾部spmlend
 # spmlstart、spmlend是为了控制不需要处理的头部和尾部，提高处理效率
 def xmltolist(spml):
     if spml.find("<spml:") != -1:
         startnum = re.search(r'<spml:[^>]*>', spml).span()[1]
         endnum = spml.rfind("</spml:")
         spmlstart = spml[:startnum].strip()
         spmlend = spml[endnum:].strip()
         spmlxml = '''<spml:modifyRequest xmlns:spml='{spml}' xmlns:subscriber="{subscriber}" xmlns:xsi="{xsi}">\n%s</spml:modifyRequest>''' % (
             spml[startnum:endnum].strip())
     elif spml.find("<PlexViewRequest") != -1:
         startnum = re.search(r'<PlexViewRequest[^>]*>', spml).span()[1]
         endnum = spml.rfind("</PlexViewRequest>")
         spmlstart = spml[:startnum].strip()
         spmlend = spml[endnum:].strip()
         spmlxml = '''<PlexViewRequest>\n%s</PlexViewRequest>''' % (spml[startnum:endnum].strip())
     else:
         spmlstart = ""
         spmlend = ""
         spmlxml = spml
     # print spmlstart
     # print endspml
     # print spmlxml
     tree = ET.fromstring(spmlxml)
     spmlxmllist = listafter(tree)
     return spmlxmllist, spmlstart, spmlend
 
 # 将xmltolist处理形成的spmlxmllist再回头变成xml（xml中，同节点的内容已被按需排列）
 def listtoxml(spmllist1):
     kong = "  "
     spmltag1 = spmllist1[0]
     spmlattrib1 = ""
     bodyxml1 = ""
     if spmllist1[1] != {}:
         for key, value in spmllist1[1].items():
             spmlattrib1 += " %s='%s'" % (key, value)
     startxml1 = "<%s%s>" % (spmltag1, spmlattrib1)
     endxml1 = "</%s>" % (spmltag1)
     spmlxml1 = ""
     if isinstance(spmllist1[2][0], list):
         spmlxml2 = ""
         for spmllist2 in spmllist1[2]:
             spmltag2 = spmllist2[0]
             spmlattrib2 = ""
             bodyxml2 = ""
             if spmllist2[1] != {}:
                 for key, value in spmllist2[1].items():
                     spmlattrib2 += " %s='%s'" % (key, value)
             startxml2 = "<%s%s>" % (spmltag2, spmlattrib2)
             endxml2 = "</%s>" % (spmltag2)
             if isinstance(spmllist2[2][0], list):
                 spmlxml3 = ""
                 for spmllist3 in spmllist2[2]:
                     spmltag3 = spmllist3[0]
                     spmlattrib3 = ""
                     bodyxml3 = ""
                     if spmllist3[1] != {}:
                         for key, value in spmllist3[1].items():
                             spmlattrib3 += " %s='%s'" % (key, value)
                     startxml3 = "<%s%s>" % (spmltag3, spmlattrib3)
                     endxml3 = "</%s>" % (spmltag3)
                     if isinstance(spmllist3[2][0], list):
                         spmlxml4 = ""
                         for spmllist4 in spmllist3[2]:
                             spmltag4 = spmllist4[0]
                             spmlattrib4 = ""
                             bodyxml4 = ""
                             if spmllist4[1] != {}:
                                 for key, value in spmllist4[1].items():
                                     spmlattrib4 += " %s='%s'" % (key, value)
                             startxml4 = "<%s%s>" % (spmltag4, spmlattrib4)
                             endxml4 = "</%s>" % (spmltag4)
                             if isinstance(spmllist4[2][0], list):
                                 spmlxml5 = ""
                                 for spmllist5 in spmllist4[2]:
                                     spmltag5 = spmllist5[0]
                                     spmlattrib5 = ""
                                     bodyxml5 = ""
                                     if spmllist5[1] != {}:
                                         for key, value in spmllist5[1].items():
                                             spmlattrib5 += " %s='%s'" % (key, value)
                                     startxml5 = "<%s%s>" % (spmltag5, spmlattrib5)
                                     endxml5 = "</%s>" % (spmltag5)
                                     if isinstance(spmllist5[2][0], list):
                                         spmlxml6 = ""
                                         for spmllist6 in spmllist5[2]:
                                             spmltag6 = spmllist6[0]
                                             spmlattrib6 = ""
                                             bodyxml6 = ""
                                             if spmllist6[1] != {}:
                                                 for key, value in spmllist6[1].items():
                                                     spmlattrib6 += " %s='%s'" % (key, value)
                                             startxml6 = "<%s%s>" % (spmltag6, spmlattrib6)
                                             endxml6 = "</%s>" % (spmltag6)
                                             if isinstance(spmllist6[2][0], list):
                                                 spmlxml7 = ""
                                                 for spmllist7 in spmllist6[2]:
                                                     spmltag7 = spmllist7[0]
                                                     spmlattrib7 = ""
                                                     bodyxml7 = ""
                                                     if spmllist7[1] != {}:
                                                         for key, value in spmllist7[1].items():
                                                             spmlattrib7 += " %s='%s'" % (key, value)
                                                     startxml7 = "<%s%s>" % (spmltag7, spmlattrib7)
                                                     endxml7 = "</%s>" % (spmltag7)
                                                     if isinstance(spmllist7[2][0], list):
                                                         spmlxml8 = ""
                                                         for spmllist8 in spmllist7[2]:
                                                             spmltag8 = spmllist8[0]
                                                             spmlattrib8 = ""
                                                             bodyxml8 = ""
                                                             if spmllist8[1] != {}:
                                                                 for key, value in spmllist8[1].items():
                                                                     spmlattrib8 += " %s='%s'" % (key, value)
                                                             startxml8 = "<%s%s>" % (spmltag8, spmlattrib8)
                                                             endxml8 = "</%s>" % (spmltag8)
                                                             if isinstance(spmllist8[2][0], list):
                                                                 spmlxml9 = ""
                                                                 for spmllist9 in spmllist8[2]:
                                                                     spmltag9 = spmllist9[0]
                                                                     spmlattrib9 = ""
                                                                     bodyxml9 = ""
                                                                     if spmllist9[1] != {}:
                                                                         for key, value in spmllist9[1].items():
                                                                             spmlattrib9 += " %s='%s'" % (key, value)
                                                                     startxml9 = "<%s%s>" % (spmltag9, spmlattrib9)
                                                                     endxml9 = "</%s>" % (spmltag9)
                                                                     if isinstance(spmllist9[2][0], list):
                                                                         spmlxml10 = ""
                                                                         for spmllist10 in spmllist9[2]:
                                                                             spmltag10 = spmllist10[0]
                                                                             spmlattrib10 = ""
                                                                             bodyxml10 = ""
                                                                             if spmllist10[1] != {}:
                                                                                 for key, value in spmllist10[1].items():
                                                                                     spmlattrib10 += " %s='%s'" % (
                                                                                         key, value)
                                                                             startxml10 = "<%s%s>" % (
                                                                                 spmltag10, spmlattrib10)
                                                                             endxml10 = "</%s>" % (spmltag10)
                                                                             bodyxml10 = spmllist10[2][0]
                                                                             spmlxml10 += "\n%s%s%s%s" % (
                                                                                 kong * 9, startxml10, bodyxml10,
                                                                                 endxml10)
                                                                         spmlxml9 += "\n%s%s%s\n%s%s" % (
                                                                             kong * 8, startxml9, spmlxml10, kong * 8,
                                                                             endxml9)
                                                                     else:
                                                                         bodyxml9 = spmllist9[2][0]
                                                                         spmlxml9 += "\n%s%s%s%s" % (
                                                                             kong * 8, startxml9, bodyxml9, endxml9)
                                                                 spmlxml8 += "\n%s%s%s\n%s%s" % (
                                                                     kong * 7, startxml8, spmlxml9, kong * 7, endxml8)
                                                             else:
                                                                 bodyxml8 = spmllist8[2][0]
                                                                 spmlxml8 += "\n%s%s%s%s" % (
                                                                     kong * 7, startxml8, bodyxml8, endxml8)
                                                         spmlxml7 += "\n%s%s%s\n%s%s" % (
                                                             kong * 6, startxml7, spmlxml8, kong * 6, endxml7)
                                                     else:
                                                         bodyxml7 = spmllist7[2][0]
                                                         spmlxml7 += "\n%s%s%s%s" % (
                                                             kong * 6, startxml7, bodyxml7, endxml7)
                                                 spmlxml6 += "\n%s%s%s\n%s%s" % (
                                                     kong * 5, startxml6, spmlxml7, kong * 5, endxml6)
                                             else:
                                                 bodyxml6 = spmllist6[2][0]
                                                 spmlxml6 += "\n%s%s%s%s" % (kong * 5, startxml6, bodyxml6, endxml6)
                                         spmlxml5 += "\n%s%s%s\n%s%s" % (
                                             kong * 4, startxml5, spmlxml6, kong * 4, endxml5)
                                     else:
                                         bodyxml5 = spmllist5[2][0]
                                         spmlxml5 += "\n%s%s%s%s" % (kong * 4, startxml5, bodyxml5, endxml5)
                                 spmlxml4 += "\n%s%s%s\n%s%s" % (kong * 3, startxml4, spmlxml5, kong * 3, endxml4)
                             else:
                                 bodyxml4 = spmllist4[2][0]
                                 spmlxml4 += "\n%s%s%s%s" % (kong * 3, startxml4, bodyxml4, endxml4)
                         spmlxml3 += "\n%s%s%s\n%s%s" % (kong * 2, startxml3, spmlxml4, kong * 2, endxml3)
                     else:
                         bodyxml3 = spmllist3[2][0]
                         spmlxml3 += "\n%s%s%s%s" % (kong * 2, startxml3, bodyxml3, endxml3)
                 spmlxml2 += "\n%s%s%s\n%s%s" % (kong * 1, startxml2, spmlxml3, kong * 1, endxml2)
             else:
                 bodyxml2 = spmllist2[2][0]
                 spmlxml2 += "\n%s%s%s%s" % (kong * 1, startxml2, bodyxml2, endxml2)
         spmlxml1 += "\n%s%s\n%s" % (startxml1, spmlxml2, endxml1)
     else:
         bodyxml1 = spmllist1[2][0]
         spmlxml1 += "\n%s%s%s" % (startxml1, bodyxml1, endxml1)
     return spmlxml1
 
 # 将startspml, xmlspml, endspml组合起来，其中有一部分内容需要根据实际情况处理
 def regroupspml(startspml, xmlspml, endspml):
     xmlspml = str(xmlspml).replace("{{", "").replace("}}", ":").strip().splitlines()
     if endspml != "":
         startspml = str(startspml.strip()).replace("\"", "\'")
         startspml = re.sub(" +>", ">", startspml)
         startspml = startspml.splitlines()
         endspml = str(endspml.strip()).splitlines()
         spmlxmlcom = startspml + xmlspml[1:-1] + endspml
     else:
         spmlxmlcom = xmlspml
     return spmlxmlcom
 
 # 对按序排列的xml进行内容比对，生成html文件，可以很直接的看出内容区别
 def diffspml(spmlxml1, spmlxml2):
     spmlxmllist1, spmlstart1, spmlend1 = xmltolist(spmlxml1)
     spmlxmllist2, spmlstart2, spmlend2 = xmltolist(spmlxml2)
     spmlxmlcom1 = listtoxml(spmlxmllist1)
     spmlxmlcom2 = listtoxml(spmlxmllist2)
     spmlxmlcom1 = regroupspml(spmlstart1, spmlxmlcom1, spmlend1)
     spmlxmlcom2 = regroupspml(spmlstart2, spmlxmlcom2, spmlend2)
     # print spmlstart1
     # print spmlend1
     if spmlxmlcom1 == spmlxmlcom2:
         return 0
     else:
         global diffspmNum
         global outputhtml_dir
         try:
             diffspmNum += 1
         except:
             diffspmNum = 1
             system = platform.system()
             if ('Windows' in system):
                 outputhtml_dir = "c:/RobotLog"
             else:
                 outputhtml_dir = "/tmp/RobotLog"
             outputhtml_dir = "%s/%s" % (outputhtml_dir, datetime.datetime.now().strftime('%Y%m%d_%H%M%S'))
             os.makedirs(outputhtml_dir)
         Loghtmldir = "%s/%s.html" % (outputhtml_dir, diffspmNum)
         # logger.write("<a href=\"%s\">%s</a>" % (Loghtmldir, Loghtmldir), "HTML")
         hd = difflib.HtmlDiff(8,65)
         with open(Loghtmldir, 'w') as fo:
             fo.write(hd.make_file(spmlxmlcom1, spmlxmlcom2))
             fo.close()
         return Loghtmldir

#############################################以上是代码部分#################################################################

 spmlxml1='''
 <PlexViewRequest SessionId="${sessionid}" ProvisioningGroup="volte" Command="ed-ngfs-subscriber-v2"><SubParty><PrimaryPUID>+86${ISDN}@${domain}</PrimaryPUID><PartyId>+86${ISDN}</PartyId></SubParty><SeqRinging><RingingList>null^null^true^true^10^false`+86${msisdn1}^STANDARD^true^true^10^false</RingingList><DefaultAnswerTimeout>10</DefaultAnswerTimeout><Send181Mode>TAS_181_NONE</Send181Mode><Activated>true</Activated><PublicUID>+86${ISDN}@${domain}</PublicUID><Assigned>true</Assigned></SeqRinging></PlexViewRequest>
 '''
 spmlxml2='''
 <PlexViewRequest SessionId="${sessionid}" ProvisioningGroup="volte" Command="ed-ngfs-subscriber-v2">
    <SubParty>
       <PrimaryPUID>+86${ISDN}@${domain}</PrimaryPUID>
       <PartyId>+86${ISDN}</PartyId>
    </SubParty>
    <SeqRinging>
       <RingingList>null^null^true^true^10^false`+86${msisdn1}^STANDARD^true^true^10^false</RingingList>
       <DefaultAnswerTimeout>10</DefaultAnswerTimeout>
       <Send181Mode>TAS_180_NONE</Send181Mode>
       <Activated>true</Activated>
       <PublicUID>+86${ISDN}@${domain}</PublicUID>
    </SeqRinging>
 </PlexViewRequest>
 '''
 print diffspml(spmlxml1, spmlxml2)

#####################################以上列出来了本公司使用的xml格式（还可以更复杂），方法中有部分内容是根据本身需要，特别处理的####################################