ydDumper.py

#!/usr/bin/env python
#This Python file uses the following encoding: utf-8
#Author:Superfan
#Version 1.0
#Latest Update:July,31,2011
#Usage:This script will convert youdao dictonary(plus version) to Mdict source file,put it in the same directory which offline glossories exist, and run it with python VM.
#This script is for STUDY use ONLY.DO NOT USE IT FOR ANY COMMERCIAL PURPOSE!!!
# Small changes by Venj, 2011-09-05

from __future__ import division
import string,re,sys,os,time
import xml.parsers.expat

class xmlParser:
    def __init__(self,strXML):
        self.strXML=strXML
        self.xmlParsed=''
        self.name=''
        self.stack=[]
        self.entryNum=0
        self.phrsNum=0
        self.lEncounterd=False
        self.nameBegin=True
        self.phrs=[]
        self.dataBuffer=''
        self.parse()
        

    def StartElementHandler(self,name,attrs):
        #tag checker
        if name not in ['word','return-phrase','l','i','phone','trs','tr','phr','des','phrs','pos','wfs','wf','exam','f','n','syno','anto']:    
            print name
            print self.strXML
            open('tmp.xml','wb').write(self.strXML)
            raw_input()
        self.stack.append(name)
        if attrs!={}:
            print attrs
            print strXML
            raw_input()
        if name in ['trs','phrs']:self.entryNum=0
        if name=='tr':
            self.lEncounterd=False
            self.entryNum+=1
        self.xmlParsed+='<span class="%s">'%name    #write current tag
        if name=='l':
            self.lNum=+1
            if self.lEncounterd==False and 'return-phrase' not in self.stack and 'trs' in self.stack:
                self.lEncounterd=True
                self.xmlParsed+='<span class="entryNum">%d.</span>'%self.entryNum
                self.xmlParsed+=u'<span class="entryDot">■</span>'
        if name=='syno':
            self.xmlParsed+=u'<span class="tongyici">近义词</span>'
        if name=='anto':
            self.xmlParsed+=u'<span class="fanyici">反义词</span>'
        if name=='wf':
            self.xmlParsed+=u'<span class="bianxing">变形</span>'
    def EndElementHandler(self,name):
        if self.dataBuffer!='':
            if self.stack[-3:]==['phr','l','i']:
                self.phrsNum+=1
                if self.dataBuffer[-1]==';':self.dataBuffer=self.dataBuffer[0:-1]
                self.phrs.append((self.dataBuffer,self.name,self.phrsNum))
                self.xmlParsed+='<a name="phr%d">'%self.phrsNum+self.dataBuffer+'</a>'
            else:self.xmlParsed+=self.dataBuffer
            self.dataBuffer=''
        self.stack.pop()
        self.xmlParsed+=r'</span>'
    def CharacterDataHandler(self,data):
        if 'return-phrase' in self.stack:
            if re.search(r'^\d+',data)!=None and data!='' and self.nameBegin==False:
                tmp=re.search(r'(^\d+)(.*)',data)
                self.name+=' %s'%data
                self.dataBuffer+='<span class="upper">%s</span>%s'%tmp.groups()
                return
            self.nameBegin=False
            self.name+=data.replace('\n','')
        if self.stack[-1]=='phone':
            self.xmlParsed+='['+data+']'
            return
        #self.xmlParsed+=data
        self.dataBuffer+=data
    def parse(self):
        p=xml.parsers.expat.ParserCreate('UTF-8')
        p.StartElementHandler=self.StartElementHandler
        p.EndElementHandler=self.EndElementHandler
        p.CharacterDataHandler=self.CharacterDataHandler
        p.returns_unicode=True
        p.Parse(self.strXML)

def decrypt(str):
    i=0
    tmp=''
    try:
        while True:
            tmp+=chr((255-ord(str[i])))   #XD
            i+=1
    except IndexError:pass
    return tmp

def parse_dict(file):
    pointer = 0x404
    #whether or not to delete entries with the same name as a former one.(for xhy.ydic keep this value for False)
    delDuplicate=False
    f=open(file,'rb')
    buffer=''
    f.seek(pointer)
    counter=0
    phrs=[]
    names=[]
    output_file = file.split(".")[0] + '.txt'
    output=open(output_file,'wb')
    
    while True:
        if len(buffer)<5*1024*1024:
            buffer+=decrypt(f.read(10*1024*1024))
        if buffer.find(r'</word>')==-1:break
        a=buffer.find(r'</word>')
        if a==-1:break
        else:
            counter+=1
        P=xmlParser(buffer[:a+len(r'</word>')])
        buffer=buffer[(a+len(r'</word>')):]
        if delDuplicate and P.name in names:continue
        output.write(P.name.encode('UTF-8')+'\x0d\x0a')
        names.append(P.name)
        output.write('<link rel="stylesheet" type="text/css" href="sf_ecce.css"/>\x0d\x0a'+P.xmlParsed.encode('UTF-8')+'\x0d\x0a</>\x0d\x0a')
        phrs+=P.phrs
        if counter%10==0:
            print '%d\r'%counter,

    print 'Done!Total %d entries'%counter
    print 'Processing references...Total:%d'%len(phrs)
    counter=0
    for (phrName,entry,phrNum) in phrs:
        counter+=1
        if delDuplicate and phrName in names:continue
        if counter%20==0:
            print 'Percentage:%%%.2f\r'%(counter/len(phrs)),
        names.append(phrName)
        output.write(phrName.encode('utf-8'))
        output.write('\x0d\x0a<link rel="stylesheet" type="text/css" href="sf_ecce.css"/>\x0d\x0a')
        output.write(u'<span class="reference"><span class="ref_title">见:</span>'.encode('utf-8'))
        output.write('<a href="entry://%s#phr%d">%s</a></span>\x0d\x0a</>\x0d\x0a'%(entry.encode('utf-8'),phrNum,phrName.encode('utf-8')))

if len(sys.argv) < 2:
    print "Usage: %s dict_file_1 [...]" % sys.argv[0]
    exit(1)

sys.argv.pop(0)

for f in sys.argv:
    print "Processing: %s" % f
    parse_dict(f)