Python实现删除Android工程中的冗余字符串

android提供了一套很方便的进行资源（语言）国际化机制，为了更好地支持多语言，很多工程的翻译往往会放到类似crowdin这样的平台上。资源是全了，但是还是会有一些问题。

哪些问题

以下使用一些语言进行举例。其中values为工程默认的资源。

1.某语言的资源和某语言限定区域的资源之间。如values-fr-rca存在于values-fr相同的字符串，这种表现最为严重。
2.某语言的资源和默认的资源之间。values-fr存在与values相同的字符串，可能原因是由于values-fr存在未翻译字符串导致

为什么要去重

洁癖，容不下半点冗余。

解决思路

1.如果values-fr-rca存在于values-fr相同的字符串，去除values-fr-rca中的重复字符串，保留values-fr。这样可以保证在values-fr-rca下也可以正确读取到资源。

2.如果values-fr存在与values相同的字符串。如去除values-fr中得重复字符串，保留values的条目。

py脚本

#!/usr/bin/env python
# coding=utf-8
from os import listdir,path, system
from sys import argv
try:
    import xml.etree.celementtree as et
except importerror:
    import xml.etree.elementtree as et

def genregionlangpair(filepath):
    basiclanguage = none
    if ('values' in filepath) :
        hasregionlimit = ('r' == filepath[-3:-2])
        if (hasregionlimit):
            basiclanguage = filepath[0:-4]
            if (not path.exists(basiclanguage)) :
                return none
            belongstoenglish =  ("values-en" in basiclanguage)
            if (belongstoenglish):
                #compare with the res/values/strings.xml
                return (path.dirname(basiclanguage) + '/values/strings.xml', filepath + "/strings.xml")
            else:
                return (basiclanguage + '/strings.xml', filepath + "/strings.xml")
    return none

def genlangpair(filepath):
    def shouldgenlanpair(filepath):
        if (not 'values' in filepath ):
            return false
        if('dpi' in filepath):
            return false
        if ('dimes' in filepath):
            return false
        if ('large' in filepath):
            return false
        return true

    if(shouldgenlanpair(filepath)):
        basiclanguage = path.dirname(filepath) + '/values/strings.xml'
        targetlanguage = filepath + '/strings.xml'
        if (not path.exists(targetlanguage)):
           return none

        if (not path.samefile(basiclanguage,targetlanguage)) :
            return (basiclanguage, targetlanguage)
    return none

def gencomparelist(filepath):
    comparelists = []
    for file in listdir(filepath):
        regionpair = genregionlangpair(filepath + '/' + file)
        if (none != regionpair):
            comparelists.append(regionpair)

        languagepair = genlangpair(filepath + '/' + file)
        if (none != languagepair) :
            comparelists.append(languagepair)

    return comparelists

def getxmlentries(filepath):
    root = et.elementtree(file=filepath).getroot()
    entries = {}
    for child in root:
        attrib = child.attrib
        if (none != attrib) :
            entries[attrib.get('name')] = child.text
    print 'xmlentriescount',len(entries)
    return entries

def rewriteregionfile(sourceentries, filepath):
    if (not path.exists(filepath)):
        return
    et.register_namespace('xliff',"urn:oasis:names:tc:xliff:document:1.2")
    tree = et.elementtree(file=filepath)
    root = tree.getroot()
    print root
    totalcount = 0
    removecount = 0
    unremovecount = 0
    print len(root)
    toremovelist = []
    for child in root:
        totalcount = totalcount + 1
        attrib = child.attrib
        if (none == attrib):
            continue

        childname = attrib.get('name')

        if (sourceentries.get(childname) == child.text):
            removecount = removecount + 1
            toremovelist.append(child)
        else:
            unremovecount = unremovecount + 1
            print childname, sourceentries.get(childname), child.text
    print filepath,totalcount, removecount,unremovecount

    for aitem in toremovelist:
        root.remove(aitem)

    if (len(root) != 0 ):
        tree.write(filepath, encoding="utf-8")
    else:
        command = 'rm -rf %s'%(path.dirname(filepath))
        print command
        system(command)

def main(projectdir):
    lists = gencomparelist(projectdir + "/res/")

    for item in lists:
        print item
        src = item[0]
        dest = item[1]
        rewriteregionfile(getxmlentries(src),dest)

if __name__ == "__main__":
    if (len(argv) == 2) :
        main(argv[1])

如何使用

python removerepeatedstrings.py your_android_project_root_dir