python 正則表達式匹配中文

 
            # 
            !/usr/bin/python 
            
# 
            -*- coding:cp936-*- 
           

#思路，將str轉換成unicode，方可用正則表達式，前提是，要知道文件的編碼，本例中是gbk
import cPickle as mypickle
import re
import sys
if (__name__=='__main__'):
    fid1=file('above50purenames.txt','r');
    p=re.compile('(^\s+|\s+$)');
    phanzigbk=re.compile('[\\x20-\\x7f]');
    phanzi=re.compile(u'[\u4e00-\u9fa5]');#這里要加u,注意
    commlines=fid1.readlines();
    fid1.close();
    dictfamilyname={};
    dictfirstname={};
    for line in commlines:
        line=p.sub('',line);
        print type(line);
        print line;
        uline=unicode(line,'gbk');
        print type(uline);
        candidates=phanzi.findall(uline);

        print len(candidates);
        if(len(candidates)==2):
            print candidates[0];
            familynamegbk=candidates[0].encode('gbk');#把unicode型的變量變成str型的變量
            firstnamegbk=candidates[1].encode('gbk');
            if(dictfamilyname.has_key(familynamegbk)):
                dictfamilyname[familynamegbk]=dictfamilyname[familynamegbk]+1;
            else:
                dictfamilyname[familynamegbk]=1;
        
            if(dictfirstname.has_key(firstnamegbk)):
                dictfirstname[firstnamegbk]=dictfirstname[firstnamegbk]+1;
            else:
                dictfirstname[firstnamegbk]=1;

    familynameitems=dictfamilyname.items();
    print familynameitems;
    firstnameitems=dictfirstname.items();
    familynameitems.sort(key=lambda d:d[1],reverse=True);
    firstnameitems.sort(key=lambda d :d[1],reverse=True);
    fid=file('familyname.txt','w');
    for m in familynameitems:
        s=m[0]+'\t'+str(m[1]);
        fid.write(s);
        fid.write('\n');
    fid.close();
    fid=file('firstname.txt','w');
    for m in firstnameitems:
        s=m[0]+'\t'+str(m[1]);
        fid.write(s);
        fid.write('\n');
    fid.close();
    print 'finish'