_ Git - youtube-dl/blob - youtube_dl/extractor/bliptv.py

   1 import datetime
   2 import json
   3 import re
   4 import socket
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_http_client,
   9     compat_parse_qs,
  10     compat_str,
  11     compat_urllib_error,
  12     compat_urllib_parse_urlparse,
  13     compat_urllib_request,
  14
  15     ExtractorError,
  16     unescapeHTML,
  17 )
  18
  19
  20 class BlipTVIE(InfoExtractor):
  21     """Information extractor for blip.tv"""
  22
  23     _VALID_URL = r'^(?:https?://)?(?:www\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
  24     _URL_EXT = r'^.*\.([a-z0-9]+)$'
  25     IE_NAME = u'blip.tv'
  26     _TEST = {
  27         u'url': u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  28         u'file': u'5779306.m4v',
  29         u'md5': u'80baf1ec5c3d2019037c1c707d676b9f',
  30         u'info_dict': {
  31             u"upload_date": u"20111205",
  32             u"description": u"md5:9bc31f227219cde65e47eeec8d2dc596",
  33             u"uploader": u"Comic Book Resources - CBR TV",
  34             u"title": u"CBR EXCLUSIVE: \"Gotham City Imposters\" Bats VS Jokerz Short 3"
  35         }
  36     }
  37
  38     def report_direct_download(self, title):
  39         """Report information extraction."""
  40         self.to_screen(u'%s: Direct download detected' % title)
  41
  42     def _real_extract(self, url):
  43         mobj = re.match(self._VALID_URL, url)
  44         if mobj is None:
  45             raise ExtractorError(u'Invalid URL: %s' % url)
  46
  47         # See https://github.com/rg3/youtube-dl/issues/857
  48         api_mobj = re.match(r'http://a\.blip\.tv/api\.swf#(?P<video_id>[\d\w]+)', url)
  49         if api_mobj is not None:
  50             url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id')
  51         urlp = compat_urllib_parse_urlparse(url)
  52         if urlp.path.startswith('/play/'):
  53             response = self._request_webpage(url, None, False)
  54             redirecturl = response.geturl()
  55             rurlp = compat_urllib_parse_urlparse(redirecturl)
  56             file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2]
  57             url = 'http://blip.tv/a/a-' + file_id
  58             return self._real_extract(url)
  59
  60         if '?' in url:
  61             cchar = '&'
  62         else:
  63             cchar = '?'
  64         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  65         request = compat_urllib_request.Request(json_url)
  66         request.add_header('User-Agent', 'iTunes/10.6.1')
  67         self.report_extraction(mobj.group(1))
  68         urlh = self._request_webpage(request, None, False,
  69             u'unable to download video info webpage')
  70
  71         try:
  72             json_code_bytes = urlh.read()
  73             json_code = json_code_bytes.decode('utf-8')
  74         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  75             raise ExtractorError(u'Unable to read video info webpage: %s' % compat_str(err))
  76
  77         try:
  78             json_data = json.loads(json_code)
  79             if 'Post' in json_data:
  80                 data = json_data['Post']
  81             else:
  82                 data = json_data
  83
  84             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  85             if 'additionalMedia' in data:
  86                 formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height']))
  87                 best_format = formats[-1]
  88                 video_url = best_format['url']
  89             else:
  90                 video_url = data['media']['url']
  91             umobj = re.match(self._URL_EXT, video_url)
  92             if umobj is None:
  93                 raise ValueError('Can not determine filename extension')
  94             ext = umobj.group(1)
  95
  96             return {
  97                 'id': compat_str(data['item_id']),
  98                 'url': video_url,
  99                 'uploader': data['display_name'],
 100                 'upload_date': upload_date,
 101                 'title': data['title'],
 102                 'ext': ext,
 103                 'format': data['media']['mimeType'],
 104                 'thumbnail': data['thumbnailUrl'],
 105                 'description': data['description'],
 106                 'player_url': data['embedUrl'],
 107                 'user_agent': 'iTunes/10.6.1',
 108             }
 109         except (ValueError, KeyError) as err:
 110             raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
 111
 112
 113 class BlipTVUserIE(InfoExtractor):
 114     """Information Extractor for blip.tv users."""
 115
 116     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 117     _PAGE_SIZE = 12
 118     IE_NAME = u'blip.tv:user'
 119
 120     def _real_extract(self, url):
 121         # Extract username
 122         mobj = re.match(self._VALID_URL, url)
 123         if mobj is None:
 124             raise ExtractorError(u'Invalid URL: %s' % url)
 125
 126         username = mobj.group(1)
 127
 128         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 129
 130         page = self._download_webpage(url, username, u'Downloading user page')
 131         mobj = re.search(r'data-users-id="([^"]+)"', page)
 132         page_base = page_base % mobj.group(1)
 133
 134
 135         # Download video ids using BlipTV Ajax calls. Result size per
 136         # query is limited (currently to 12 videos) so we need to query
 137         # page by page until there are no video ids - it means we got
 138         # all of them.
 139
 140         video_ids = []
 141         pagenum = 1
 142
 143         while True:
 144             url = page_base + "&page=" + str(pagenum)
 145             page = self._download_webpage(url, username,
 146                                           u'Downloading video ids from page %d' % pagenum)
 147
 148             # Extract video identifiers
 149             ids_in_page = []
 150
 151             for mobj in re.finditer(r'href="/([^"]+)"', page):
 152                 if mobj.group(1) not in ids_in_page:
 153                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 154
 155             video_ids.extend(ids_in_page)
 156
 157             # A little optimization - if current page is not
 158             # "full", ie. does not contain PAGE_SIZE video ids then
 159             # we can assume that this page is the last one - there
 160             # are no more ids on further pages - no need to query
 161             # again.
 162
 163             if len(ids_in_page) < self._PAGE_SIZE:
 164                 break
 165
 166             pagenum += 1
 167
 168         urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
 169         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 170         return [self.playlist_result(url_entries, playlist_title = username)]