編寫Python指令碼批量下載DesktopNexus壁紙的教程

DesktopNexus 是我最喜愛的一個壁紙下載網站，上面有許多高品質的壁紙，幾乎每天必上， 每月也必會堅持分享我這個月來收集的壁紙

但是 DesktopNexus 壁紙的下載很麻煩，而且因為壁紙會通過瀏覽器檢測你當前解析度來展示 合適你當前解析度的壁紙，再加上是國外的網站，速度上很不樂觀。

於是我寫了個指令碼，檢測輸入的頁面中壁紙頁面的連結，然後批量下載到指定檔案夾中。

指令碼使用 python 寫的，所以需要機器上安裝有 python 。
用法：

$ python desktop_nexus.py -p http://www.desktopnexus.com/tag/cat/ -s 1280x800 -o wallpapers

-p 包含 DesktopNexus 壁紙連結的頁面，比如我的壁紙分享
-s 壁紙尺寸，可選，預設為 1440x900
-o 壁紙輸出的檔案夾，可選，預設為目前的目錄下的 wallpapers， 如果不存在會自動建立

代碼：

#-*- coding: utf-8 -*-from argparse import ArgumentParserimport os, re, sysimport urllib2, cookielib, urlparseRE_WALLPAPER = r'http\:\/\/[^\/\.]+\.desktopnexus\.com\/wallpaper\/\d+\/'CHUNK_SIZE = 1024 * 3class 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) <= 70 and msg or msg[:67] + '...', progress))    sys.stdout.flush()    if progress >= 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.', eif __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='1440x900', \      help='specific the wallpaper size, default to 1440x900')  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()

