1.正則的使用匹配

2.dic.setdefault()的使用

3、內建函式enumerate(sequence,start=0)的使用

4、內建函式sorted(),key,reversed引數設定

5、str.lower()string大小寫轉換

#coding:utf-8
import re
import os
import time
import codecs

PATH = os.path.dirname(__file__)

s = u'what a Beautiful woRld'.lower()
pattern = re.compile(u'[^a-z]+', re.U)#在非英文出進行切割
for con in pattern.split(s.lower()):#將所有英文轉化為小寫
    if len(con) <= 1:
        continue
    else:
        print con

def get_english_words():
    '''過濾出語料中夾雜在漢語中的英文單詞，並統計出現的詞頻'''
    eng_freq_dic = {}
    pattern = re.compile(u'[^a-z]+', re.U)
    cut_filename = r'E:\SVN\linguistic_model\data\combine_msg_comment.txt'
    with codecs.open(cut_filename, encoding='utf-8') as f:
        for line in f.readlines():
            for con in pattern.split(line.lower()):
                if len(con) <= 1:#過濾掉單字母
                    continue
                else:
                    count = eng_freq_dic.setdefault(con, 0) + 1 #若沒有該key，則儲存該key且設value其為0。若有則value加1
                    eng_freq_dic[con] = count#整個英文單詞及其出現的頻度
    eng_filename = os.path.join(PATH, 'english_words_original.txt')
    eng_to_write_list = sorted([(k,v) for (k, v) in eng_freq_dic.items()], key=lambda x:x[1], reverse=True)#按照詞頻的高低進行倒序排列
    codecs.open(eng_filename, mode='wb', encoding='utf-8').writelines([item[0]+'\t'+str(item[1])+'\n' for item in eng_to_write_list])#詞頻為int型別，轉化為str型別以後寫入到本地檔案中

def chose_top_n():
    '''篩選出top2000,並寫入到檔案'''
    line_list = []
    filename = os.path.join(PATH, 'english_words_original.txt')
    with codecs.open(filename, encoding='utf-8') as f:
        for index,line in enumerate(f.readlines(), start=1):#enumerate(sequence, start=0)用法，顯示可迭代序列中元素及其位置，start引數可以確定起始下標，預設情況下為0
            print index, line.strip()
            time.sleep(1)
            line_list.append(line)
            if index == 2000:
                top_filename = os.path.join(PATH, 'top_2000_english_words.txt')
                codecs.open(top_filename, mode='wb', encoding='utf-8').writelines(line_list)
                break