rearranging the record that stardict saved to readable lines.

trans.py

#!/usr/bin/env python

#########################################################
# I wrote the little script depended on following dictionaries:
# stardict-cced-2.4.2.tar.bz2, stardict-langdao-ec-gb-2.4.2.tar.bz2,
# stardict-ncce-ec-2.4.2.tar.bz2, stardict-xiangya-medical-2.4.2.tar.bz2
#
# If you want to use other dictionary you will add or remove parts of the code.
# 
# Default is designed to remove the record of <Collins Cobuild English Dictionary>
# for Chinese users.
########################################################

import string

class cnvt():

    def __init__(self, filename):
        self.source = open(filename).read()
        self.dict   = {}
        
    def select(self):
        dict_num = len(self.source.split('<--- Collins Cobuild English Dictionary --->'))
        i = 0
        for a in range(dict_num):
         # selecting Collins by sizes of one word defination
        
            i = self.source.find('<--- Collins Cobuild English Dictionary --->', i)
            if i != -1:
                ii = self.source.find('<---', i + 4)
                tmp = self.source[i:ii]
                if ('\n\n' in tmp and tmp.split('\n\n')[-1][0].isdigit()) \
                or ('\n\n' in tmp and '=>' in tmp) or ('\n\n' not in tmp):
                    self.source = self.source.replace(tmp, '')
                i = ii
                
    def select2(self):
        dict_num = len(self.source.split('*'))
        i = 0
        for a in range(dict_num):
         # selecting quota by sizes of one word defination
        
            i = self.source.find('*', i)
            if i != -1:
                ii = self.source.find('\n', i + 1)
                tmp = self.source[i:ii + 1]
                self.source = self.source.replace(tmp, '')
                i = ii
                
    def make_dict(self):
        ldict = self.source.split('\n\n')
        if ldict[-1] == '':
            del ldict[-1]
            
        print 'There are %d words!' % len(ldict)
        
        words_list = []
        for line in ldict:
            tmp = line.split('\n')
            if tmp[-1] == '':
                del tmp[-1]
                
            head = tmp[0]           # selecting word
            del tmp[0]
            words_list.append(head)
            
            if '<---' not in line:
                print 'Error! ---> %s' % head
                
            dict_level2 = {}
                
            dict_name = ''
            for item in tmp:
                if '<---' in item:
                    dict_name = item
                    dict_level2[item] = ''
                
                else:
                    try:
                        if 'Collins Cobuild' in item:
                            dict_level2[dict_name] = dict_level2[dict_name] + ',' + item
                        else:
                            word_is_redundant = True
                            y = item.replace(' ', '')
                            y = y.replace('-', '')
                            for i in range(len(y) - 1):
                                a = ord(y[i])
                                if 48 <= a <= 122: pass
                                else:
                                     word_is_redundant = False
                                     break
                          
                            if word_is_redundant: pass
                            else:
                                dict_level2[dict_name] = dict_level2[dict_name] + ',' + item
                                
                    except KeyError:
                        print 'Error! ---> %s' % head
                
            self.dict[head] = dict_level2
            
        # comparing sizes of defination of each words, then recording definations
        dict_file = open('out', 'w')
        for word in words_list:
            defines = self.dict[word]
            lengths = {}
            for dict_name in defines.keys():
                lengths[len(defines[dict_name])] = dict_name
                
            sizes = lengths.keys()
            sizes.sort()
            
            try:
                record = defines[lengths[sizes[-1]]]   # biggest defination
            except IndexError:
                print '<@_@>!!!' + ' --->\t' + word + '  ==> ' + str(defines)
                record = ' <@_@>!!! '
                    
            dict_file.write(word + '\t\t' + '---- ' + record[1:] + '.' + '\n')
                
        dict_file.write('\n\n\n\\\\ original words:\n')
        for word in words_list:   
        # recording original words
            dict_file.write(word + '\n')
            
        dict_file.close()
                
                
if __name__ == '__main__':
    import sys, os
    argv = sys.argv
    if len(argv) == 1:
        try:
            user_home  = os.popen('printenv HOME').read()[:-1]
            do = cnvt(user_home + '/dic.txt')
        except IOError:
            print "Warning! Indicating the right address of stardict record's file)"
    else:
        do = cnvt(sys.argv[1])
    
    do.select2()
    do.make_dict()
    
    print 'Outputing file is at %s\n' % os.getcwd()
    
    raw_input()