编写python脚本批量下载desktopnexus壁纸的教程

desktopnexus 是我最喜爱的一个壁纸下载网站,上面有许多高质量的壁纸,几乎每天必上, 每月也必会坚持分享我这个月来收集的壁纸

但是 desktopnexus 壁纸的下载很麻烦,而且因为壁纸会通过浏览器检测你当前分辨率来展示 合适你当前分辨率的壁纸,再加上是国外的网站,速度上很不乐观。

于是我写了个脚本,检测输入的页面中壁纸页面的链接,然后批量下载到指定文件夹中。

脚本使用 python 写的,所以需要机器上安装有 python 。
用法:

$ python desktop_nexus.py -p http://www.desktopnexus.com/tag/cat/ -s 1280×800 -o wallpapers

-p 包含 desktopnexus 壁纸链接的页面,比如我的壁纸分享
-s 壁纸尺寸,可选,缺省为 1440×900
-o 壁纸输出的文件夹,可选,缺省为当前目录下的 wallpapers, 如果不存在会自动创建

代码:

#-*- coding: utf-8 -*-
from argparse import argumentparser
import os, re, sys
import urllib2, cookielib, urlparse
re_wallpaper = r’http\:\/\/[^\/\.]+\.desktopnexus\.com\/wallpaper\/\d+\/’
chunk_size = 1024 * 3
class desktopnexus:
def __init__(self, page=none, size=none, output_dir=none):
self.page = page
self.size = size
self.output_dir = output_dir
def start(self):
print ‘making output directory:’, self.output_dir
if not os.path.exists(self.output_dir):
os.makedirs(self.output_dir)
# setup cookie
cookie = cookielib.cookiejar()
processer = urllib2.httpcookieprocessor(cookie)
opener = urllib2.build_opener(processer)
urllib2.install_opener(opener)
self._read_page()
def _get_pic_info(self, url):
pic_id = url.split(‘/’)[-2]
html = urllib2.urlopen(url).read()
pattern = r’.*?)\”‘ % pic_id
match = re.search(pattern, html, flags=re.i|re.m|re.s)
if match:
return {‘id’: pic_id,
‘token’: match.group(‘token’),
‘size’: self.size}
else:
raise exception(‘cound not find wallpaper’)
def _get_pic_file(self, pic_info):
redirect_url = ‘http://www.desktopnexus.com/dl/inline/%(id)s/%(size)s/%(token)s’ % pic_info
request = urllib2.urlopen(redirect_url)
return request.geturl()
def _download_pic(self, url):
pic_info = self._get_pic_info(url)
pic_file = self._get_pic_file(pic_info)
filename = os.path.split(urlparse.urlparse(pic_file).path)[-1]
filename = os.path.join(self.output_dir, filename)
with open(filename, ‘wb’) as output:
resp = urllib2.urlopen(pic_file)
total_size = int(resp.info().get(‘content-length’))
saved_size = 0.0
while saved_size != total_size:
chunk = resp.read(chunk_size)
saved_size += len(chunk)
output.write(chunk)
self._print_progress(‘saving file: %s’ % filename, \
saved_size / total_size * 100)
def _print_progress(self, msg, progress):
sys.stdout.write(‘%-71s%3d%%\r’ \
% (len(msg) = 100:
sys.stdout.write(‘\n’)
def _read_page(self):
try:
print ‘fetching content:’, self.page
html = urllib2.urlopen(self.page).read()
links = set(re.findall(re_wallpaper, html, re.m|re.i))
count = len(links)
print ‘downloading wallpapers:’
for i, link in enumerate(links):
print ‘[%d/%d]: %s’ % (i + 1, count, link)
try:
self._download_pic(link)
except exception as e:
print ‘error downloading wallpaper.’, e.message
except exception as e:
print ‘error fetching content.’, e
if __name__ == ‘__main__’:
# setup argparser
parser = argumentparser(‘python desktop_nexus.py’)
parser.add_argument(‘-p’, ‘–page’, dest=’page’, required=true, \
help=’specific a page that includes wallpaper list’)
parser.add_argument(‘-s’, ‘–size’, dest=’size’, default=’1440×900′, \
help=’specific the wallpaper size, default to 1440×900′)
parser.add_argument(‘-o’, ‘–output’, dest=’output_dir’, default=’wallpapers’, \
help=’specific the output directory, default to “wallpapers”‘)
args = parser.parse_args()
dn = desktopnexus(**args.__dict__)
dn.start()

Posted in 未分类