读取几个目录下及其各自的子目录中所有的txt文件,并从中筛选出domain,然后按照每个domain出现的次数进行一个排序,最后将这个排序结果写入到一个文件中。
#coding=utf8
# 统计data_naver和data_google两个目录下出现的每个domain的次数,按照从多到少进行排序
import os
import re
# 将给定目录下所有的txt文件整合到一个list中
def merge_txt(path_list):
file_list = []
for path in path_list:
for (dirpath, dirnames, filenames) in os.walk(path):
# print filenames
for filename in filenames:
filename = os.path.join(dirpath, filename)
# print filename
file_list.append(filename)
print "所有文件的总数是:",len(file_list)
return file_list
# 将文件列表中每个文件里面的url合并到一个list
def merge_to_domain_list(file_list):
all_domain_list = []
domain_pat = '"https*://(.*?)/'
# 对于每个文件,读取每行,并对每行进行处理
for file in file_list:
file_object = open(file)
# 对每行用正则匹配出url
for line in file_object:
# print line
domain_list = re.findall(domain_pat,line)
all_domain_list = all_domain_list + domain_list
file_object.close()
print all_domain_list
print "url的总数是:",len(all_domain_list)
return all_domain_list
# 对list按照出现次数排序,得出一个dict,key是url,value是count(次数)
def sort_list_by_count(domain_list):
domain_dict = {}
for domain in domain_list:
domain_count = domain_dict.get(domain)
if domain_count:
domain_count = int(domain_count)
domain_count += 1
domain_dict[domain] = domain_count
else:
domain_dict[domain] = 1
# print domain_dict
items = domain_dict.items()
backitems = [[v[1], v[0]] for v in items]
# print backitems
# backitems.sort()
backitems = sorted(backitems)
# backitems = reversed(backitems)
backitems.reverse()
print [[v[1], v[0]] for v in backitems]
list_sorted_by_count = [[v[1], v[0]] for v in backitems]
return list_sorted_by_count
# 将一个list写入文件
def write_list_to_file(sorted_list,full_path):
sorted_list_file = open(full_path,'a')
for l in sorted_list:
sorted_list_file.write(str(l) + "\n")
sorted_list_file.close()
if __name__ == '__main__':
dir = ['./data_google','./data_naver']
file_list = merge_txt(dir)
domain_list = merge_to_domain_list(file_list)
# print type(domain_list)
list_sorted_by_count = sort_list_by_count(domain_list)
full_path = './data/sorted.txt'
write_list_to_file(list_sorted_by_count, full_path)
网友评论