youtube_dl/extractor/photobucket.py

   1 import datetime
   2 import json
   3 import re
   4
   5 from .common import InfoExtractor
   6
   7 from ..utils import (
   8     ExtractorError,
   9 )
  10
  11 class PhotobucketIE(InfoExtractor):
  12     """Information extractor for photobucket.com."""
  13
  14     # TODO: the original _VALID_URL was:
  15     # r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)'
  16     # Check if it's necessary to keep the old extracion process
  17     _VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P<id>.*)\.(?P<ext>(flv)|(mp4))'
  18     IE_NAME = u'photobucket'
  19     _TEST = {
  20         u'url': u'http://media.photobucket.com/user/rachaneronas/media/TiredofLinkBuildingTryBacklinkMyDomaincom_zpsc0c3b9fa.mp4.html?filters[term]=search&filters[primary]=videos&filters[secondary]=images&sort=1&o=0',
  21         u'file': u'zpsc0c3b9fa.mp4',
  22         u'md5': u'7dabfb92b0a31f6c16cebc0f8e60ff99',
  23         u'info_dict': {
  24             u"upload_date": u"20130504",
  25             u"uploader": u"rachaneronas",
  26             u"title": u"Tired of Link Building? Try BacklinkMyDomain.com!"
  27         }
  28     }
  29
  30     def _real_extract(self, url):
  31         # Extract id from URL
  32         mobj = re.match(self._VALID_URL, url)
  33         if mobj is None:
  34             raise ExtractorError(u'Invalid URL: %s' % url)
  35
  36         video_id = mobj.group('id')
  37
  38         video_extension = mobj.group('ext')
  39
  40         # Retrieve video webpage to extract further information
  41         webpage = self._download_webpage(url, video_id)
  42
  43         # Extract URL, uploader, and title from webpage
  44         self.report_extraction(video_id)
  45         # We try first by looking the javascript code:
  46         mobj = re.search(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (?P<json>.*?)\);', webpage)
  47         if mobj is not None:
  48             info = json.loads(mobj.group('json'))
  49             return [{
  50                 'id':       video_id,
  51                 'url':      info[u'downloadUrl'],
  52                 'uploader': info[u'username'],
  53                 'upload_date':  datetime.date.fromtimestamp(info[u'creationDate']).strftime('%Y%m%d'),
  54                 'title':    info[u'title'],
  55                 'ext':      video_extension,
  56                 'thumbnail': info[u'thumbUrl'],
  57             }]
  58
  59         # We try looking in other parts of the webpage
  60         video_url = self._search_regex(r'<link rel="video_src" href=".*\?file=([^"]+)" />',
  61             webpage, u'video URL')
  62
  63         mobj = re.search(r'<title>(.*) video by (.*) - Photobucket</title>', webpage)
  64         if mobj is None:
  65             raise ExtractorError(u'Unable to extract title')
  66         video_title = mobj.group(1).decode('utf-8')
  67         video_uploader = mobj.group(2).decode('utf-8')
  68
  69         return [{
  70             'id':       video_id.decode('utf-8'),
  71             'url':      video_url.decode('utf-8'),
  72             'uploader': video_uploader,
  73             'upload_date':  None,
  74             'title':    video_title,
  75             'ext':      video_extension.decode('utf-8'),
  76         }]