fix language meta of chinese epubs in iBooks Library

zhnize.py

# -*- coding: utf-8 -*-

from biplist import readPlist,writePlist
import re
import os.path
import xml.etree.ElementTree as et


def change_lang_zh(opf):
    opftree = et.parse(opf)
    opfroot = opftree.getroot()
    # dc namespace
    title_tag = '{http://purl.org/dc/elements/1.1/}title'
    lang_tag = '{http://purl.org/dc/elements/1.1/}language'
    epub_title = opfroot[0].find(title_tag)
    epub_lang = opfroot[0].find(lang_tag)
    #Remove default namespace for avoiding ns0 markup.
    et.register_namespace("","http://www.idpf.org/2007/opf")
    if epub_title is None:
        print "Corrupted EPUB => " + opf
        return False

    #If the title of EPUB contains Chinese chracter and not contains Jap's Kana.
    if ( re.search(ur"[\u4e00-\u9fa5]+", epub_title.text) 
        and not re.search(ur"[\u3041-\u3096\u30A0-\u30FF]+", epub_title.text) ):
        # Add lang meta data if not present.
        if epub_lang is None :
            print "Language meta for %s is None" % opf
            et.SubElement(opfroot[0], lang_tag)
            epub_lang = opfroot[0].find(lang_tag)
        if ( epub_lang.text == 'en' or epub_lang.text == '' ):
            print "Changing Language to zh for %s" % opf
            epub_lang.text = 'zh'
            opftree.write(opf,encoding='UTF-8')

BOOK_LIB_PATH = os.path.join(os.path.expanduser("~"), 
    "Library/Containers/com.apple.BKAgentService/Data/Documents/iBooks/Books/")

bkplist = readPlist(BOOK_LIB_PATH + 'Books.plist')

zh_epubs = []
for item in bkplist['Books']:
    if ( not re.match(ur".*pdf$", item['BKDisplayName']) 
         and not 'BKMigrationSourcePath' in item ):
        for i in os.walk(item['path']):
            for f in i[2]:
                if re.match(r'.*\.opf$',f):
                    opf = os.path.join(i[0],f)
                    print "OPF FOUND for " + item['BKDisplayName']
                    change_lang_zh(opf)