python 简易小爬虫

此脚本用于爬站点的下载链接，最终输出到txt文档中。

如果是没有防盗链设置的站点，也可以使用脚本中的下载函数尝试直接下载。

本脚本是为了短期特定目标设计的，如果使用它爬其它特征的资源链接需自行修改配置语句。

python初学者，请多多指正。

# -*- coding: utf-8 -*-
import re
import urllib
import os
import urllib2
import requests
import time
 
#download the file
def download(page, url):
	local_filename =url.split('/')[-1] + page + '.jpg'
	r = requests.get(url, stream=True)
	with open(local_filename, 'wb') as f:
		for chunk in r.iter_content(chunk_size = 1024):
			if chunk: # filter out keep-alive new chunks
				f.write(chunk)
                f.flush() 
 
	return local_filename
 
#turn the data array into urls array
def print_urls(urls):
	output_urls = []
	for link in urls:
		start_link = link.find('"')
		end_link = link.find('"', start_link+1)
		output_link = link[start_link+1: end_link]
		if output_link.find('http') == -1:
			output_link = 'http://www.XXX.com' + output_link
		if link.count('"') > 2:
			continue
		else:
			output_urls.append(output_link)
	return output_urls
 
def output_download_link_page(page):
	url = page
	s = urllib.urlopen(url).read()
	urls = []
	img_urls = 'no image on' + page
	new_stl_urls = []
 
	title = re.findall(r'<h1>.+<\/h1>', s, re.I)
	if len(title) != 0:
		title = title[0]
	else:
		title = 'no title'
 
	img_urls = print_urls(re.findall(r'href=".*?\.jpg.*?"', s, re.I))
	if len(img_urls) != 0:
		img_urls = img_urls[0]
	else:
		img_urls = 'no image' + page	
 
	stl_urls = print_urls (set(re.findall(r'href="/download/.*?"', s, re.I)))
 
	for url in stl_urls:
		#url = urllib2.urlopen(url).url
		url = requests.get(url).url
		new_stl_urls.append(url)
 
	urls.append(title)
	urls.append(img_urls)
	urls = urls + new_stl_urls
 
	return urls
 
#print output_download_link_page('http://www.XXX.com/thing/46876')
 
#output all links to download
def output_all_pages(site):
	s = urllib.urlopen(site).read()
	page = re.findall(r'href="/thing/.*?"', s, re.I)
	page = set(page)
	return print_urls(page)
 
#output all the sites to download
def generate_sites(start, end):
	sites = []
	for  num in range(start, end):
		sites.append('http://www.XXX.com/popular?query=&pg=' + str(num))
	return sites
 
#write all the results to a txt file
file_new = open ('1.txt', 'r+')
url_pakage = []
sites = generate_sites(40, 46)
count = 0
 
for site in sites:
	print site
	file_new.write( '\n' + site)
	pages = output_all_pages(site)
	for page in pages:
		urls = output_download_link_page(page)
		#
		if len(urls) >= 10:
			continue
		count = count + 1
		for url in urls:
			file_new.write(url + '\n')
	print 'done'
	time.sleep(10)
 
file_new.close()
print 'all done. all..' + str(count) + '..models'