Python multi threaded grab tianya post content sample
- 2020-04-02 13:34:21
- OfStack
Using re, urllib, threading multi-threading to grab the content of the tianya post, set the url as the first page of the tianya post to be grabbed, and set the file_name as the file name after downloading
#coding:utf-8
import urllib
import re
import threading
import os, time
class Down_Tianya(threading.Thread):
""" Multithreaded download """
def __init__(self, url, num, dt):
threading.Thread.__init__(self)
self.url = url
self.num = num
self.txt_dict = dt
def run(self):
print 'downling from %s' % self.url
self.down_text()
def down_text(self):
""" According to incoming url Grab the contents of each page, press the number of pages as the key to save the dictionary """
html_content =urllib.urlopen(self.url).read()
text_pattern = re.compile('<span> Time: (.*?)</span>.*?<!-- <div class="host-ico"> The original poster </div> -->.*?<div class="bbs-content.*?>s*(.*?)</div>', re.DOTALL)
text = text_pattern.findall(html_content)
text_join = ['rnrnrnrn'.join(item) for item in text]
self.txt_dict[self.num] = text_join
def page(url):
""" Fetch the total number of pages based on the first page address """
html_page = urllib.urlopen(url).read()
page_pattern = re.compile(r'<a href="S*?">(d*)</a>s*<a href="S*?" class="S*?"> On the next page </a>')
page_result = page_pattern.search(html_page)
if page_result:
page_num = int(page_result.group(1))
return page_num
def write_text(dict, fn):
""" To write a dictionary content key (number of pages) to text, with each key value per page list The list of """
tx_file = open(fn, 'w+')
pn = len(dict)
for i in range(1, pn+1):
tx_list = dict[i]
for tx in tx_list:
tx = tx.replace('<br>', 'rn').replace('<br />', 'rn').replace(' ', '')
tx_file.write(tx.strip()+'rn'*4)
tx_file.close()
def main():
url = 'http://bbs.tianya.cn/post-16-996521-1.shtml'
file_name ='abc.txt'
my_page = page(url)
my_dict = {}
print 'page num is : %s' % my_page
threads = []
""" Structure by number of pages urls Multi-threaded download """
for num in range(1, my_page+1):
myurl = '%s%s.shtml' % (url[:-7], num)
downlist = Down_Tianya(myurl, num, my_dict)
downlist.start()
threads.append(downlist)
""" Check that the download is complete before writing """
for t in threads:
t.join()
write_text(my_dict, file_name)
print 'All download finished. Save file at directory: %s' % os.getcwd()
if __name__ == '__main__':
main()
Down_tianya. Py
#coding:utf-8
import urllib
import re
import threading
import os
class Down_Tianya(threading.Thread):
""" Multithreaded download """
def __init__(self, url, num, dt):
threading.Thread.__init__(self)
self.url = url
self.num = num
self.txt_dict = dt
def run(self):
print 'downling from %s' % self.url
self.down_text()
def down_text(self):
""" According to incoming url Grab the contents of each page, press the number of pages as the key to save the dictionary """
html_content =urllib.urlopen(self.url).read()
text_pattern = re.compile('<div class="atl-item".*?<span> Time: (.*?)</span>.*?<!-- <div class="host-ico"> The original poster </div> -->.*?<div class="bbs-content.*?>s*(.*?)</div>', re.DOTALL)
text = text_pattern.findall(html_content)
text_join = ['rnrnrnrn'.join(item) for item in text]
self.txt_dict[self.num] = text_join
def page(url):
""" Fetch the total number of pages based on the first page address """
html_page = urllib.urlopen(url).read()
page_pattern = re.compile(r'<a href="S*?">(d*)</a>s*<a href="S*?" class="S*?"> On the next page </a>')
page_result = page_pattern.search(html_page)
if page_result:
page_num = int(page_result.group(1))
return page_num
def write_text(dict, fn):
""" To write a dictionary content key (number of pages) to text, with each key value per page list The list of """
tx_file = open(fn, 'w+')
pn = len(dict)
for i in range(1, pn+1):
tx_list = dict[i]
for tx in tx_list:
tx = tx.replace('<br>', 'rn').replace('<br />', 'rn').replace(' ', '')
tx_file.write(tx.strip()+'rn'*4)
tx_file.close()
def main():
url = 'http://bbs.tianya.cn/post-16-996521-1.shtml'
file_name ='abc.txt'
my_page = page(url)
my_dict = {}
print 'page num is : %s' % my_page
threads = []
""" Structure by number of pages urls Multi-threaded download """
for num in range(1, my_page+1):
myurl = '%s%s.shtml' % (url[:-7], num)
downlist = Down_Tianya(myurl, num, my_dict)
downlist.start()
threads.append(downlist)
""" Check that the download is complete before writing """
for t in threads:
t.join()
write_text(my_dict, file_name)
print 'All download finished. Save file at directory: %s' % os.getcwd()
if __name__ == '__main__':
main()