youtube_dl/extractor/bliptv.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from .subtitles import SubtitlesInfoExtractor
   7 from ..utils import (
   8     compat_urllib_request,
   9     unescapeHTML,
  10     parse_iso8601,
  11     compat_urlparse,
  12     clean_html,
  13     compat_str,
  14 )
  15
  16
  17 class BlipTVIE(SubtitlesInfoExtractor):
  18     _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_]+)))'
  19
  20     _TESTS = [
  21         {
  22             'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  23             'md5': 'c6934ad0b6acf2bd920720ec888eb812',
  24             'info_dict': {
  25                 'id': '5779306',
  26                 'ext': 'mov',
  27                 'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
  28                 'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
  29                 'timestamp': 1323138843,
  30                 'upload_date': '20111206',
  31                 'uploader': 'cbr',
  32                 'uploader_id': '679425',
  33                 'duration': 81,
  34             }
  35         },
  36         {
  37             # https://github.com/rg3/youtube-dl/pull/2274
  38             'note': 'Video with subtitles',
  39             'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
  40             'md5': '309f9d25b820b086ca163ffac8031806',
  41             'info_dict': {
  42                 'id': '6586561',
  43                 'ext': 'mp4',
  44                 'title': 'Red vs. Blue Season 11 Episode 1',
  45                 'description': 'One-Zero-One',
  46                 'timestamp': 1371261608,
  47                 'upload_date': '20130615',
  48                 'uploader': 'redvsblue',
  49                 'uploader_id': '792887',
  50                 'duration': 279,
  51             }
  52         },
  53         {
  54             # https://bugzilla.redhat.com/show_bug.cgi?id=967465
  55             'url': 'http://a.blip.tv/api.swf#h6Uag5KbVwI',
  56             'md5': '314e87b1ebe7a48fcbfdd51b791ce5a6',
  57             'info_dict': {
  58                 'id': '6573122',
  59                 'ext': 'mov',
  60                 'upload_date': '20130520',
  61                 'description': 'Two hapless space marines argue over what to do when they realize they have an astronomically huge problem on their hands.',
  62                 'title': 'Red vs. Blue Season 11 Trailer',
  63                 'timestamp': 1369029609,
  64                 'uploader': 'redvsblue',
  65                 'uploader_id': '792887',
  66             }
  67         }
  68     ]
  69
  70     def _real_extract(self, url):
  71         mobj = re.match(self._VALID_URL, url)
  72         lookup_id = mobj.group('lookup_id')
  73
  74         # See https://github.com/rg3/youtube-dl/issues/857
  75         if lookup_id:
  76             info_page = self._download_webpage(
  77                 'http://blip.tv/play/%s.x?p=1' % lookup_id, lookup_id, 'Resolving lookup id')
  78             video_id = self._search_regex(r'data-episode-id="([0-9]+)', info_page, 'video_id')
  79         else:
  80             video_id = mobj.group('id')
  81
  82         rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
  83
  84         def blip(s):
  85             return '{http://blip.tv/dtd/blip/1.0}%s' % s
  86
  87         def media(s):
  88             return '{http://search.yahoo.com/mrss/}%s' % s
  89
  90         def itunes(s):
  91             return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s
  92
  93         item = rss.find('channel/item')
  94
  95         video_id = item.find(blip('item_id')).text
  96         title = item.find('./title').text
  97         description = clean_html(compat_str(item.find(blip('puredescription')).text))
  98         timestamp = parse_iso8601(item.find(blip('datestamp')).text)
  99         uploader = item.find(blip('user')).text
 100         uploader_id = item.find(blip('userid')).text
 101         duration = int(item.find(blip('runtime')).text)
 102         media_thumbnail = item.find(media('thumbnail'))
 103         thumbnail = media_thumbnail.get('url') if media_thumbnail is not None else item.find(itunes('image')).text
 104         categories = [category.text for category in item.findall('category')]
 105
 106         formats = []
 107         subtitles = {}
 108
 109         media_group = item.find(media('group'))
 110         for media_content in media_group.findall(media('content')):
 111             url = media_content.get('url')
 112             role = media_content.get(blip('role'))
 113             msg = self._download_webpage(
 114                 url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
 115                 video_id, 'Resolving URL for %s' % role)
 116             real_url = compat_urlparse.parse_qs(msg)['message'][0]
 117
 118             media_type = media_content.get('type')
 119             if media_type == 'text/srt' or url.endswith('.srt'):
 120                 LANGS = {
 121                     'english': 'en',
 122                 }
 123                 lang = role.rpartition('-')[-1].strip().lower()
 124                 langcode = LANGS.get(lang, lang)
 125                 subtitles[langcode] = url
 126             elif media_type.startswith('video/'):
 127                 formats.append({
 128                     'url': real_url,
 129                     'format_id': role,
 130                     'format_note': media_type,
 131                     'vcodec': media_content.get(blip('vcodec')),
 132                     'acodec': media_content.get(blip('acodec')),
 133                     'filesize': media_content.get('filesize'),
 134                     'width': int(media_content.get('width')),
 135                     'height': int(media_content.get('height')),
 136                 })
 137         self._sort_formats(formats)
 138
 139         # subtitles
 140         video_subtitles = self.extract_subtitles(video_id, subtitles)
 141         if self._downloader.params.get('listsubtitles', False):
 142             self._list_available_subtitles(video_id, subtitles)
 143             return
 144
 145         return {
 146             'id': video_id,
 147             'title': title,
 148             'description': description,
 149             'timestamp': timestamp,
 150             'uploader': uploader,
 151             'uploader_id': uploader_id,
 152             'duration': duration,
 153             'thumbnail': thumbnail,
 154             'categories': categories,
 155             'formats': formats,
 156             'subtitles': video_subtitles,
 157         }
 158
 159     def _download_subtitle_url(self, sub_lang, url):
 160         # For some weird reason, blip.tv serves a video instead of subtitles
 161         # when we request with a common UA
 162         req = compat_urllib_request.Request(url)
 163         req.add_header('Youtubedl-user-agent', 'youtube-dl')
 164         return self._download_webpage(req, None, note=False)
 165
 166
 167 class BlipTVUserIE(InfoExtractor):
 168     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
 169     _PAGE_SIZE = 12
 170     IE_NAME = 'blip.tv:user'
 171
 172     def _real_extract(self, url):
 173         mobj = re.match(self._VALID_URL, url)
 174         username = mobj.group(1)
 175
 176         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 177
 178         page = self._download_webpage(url, username, 'Downloading user page')
 179         mobj = re.search(r'data-users-id="([^"]+)"', page)
 180         page_base = page_base % mobj.group(1)
 181
 182         # Download video ids using BlipTV Ajax calls. Result size per
 183         # query is limited (currently to 12 videos) so we need to query
 184         # page by page until there are no video ids - it means we got
 185         # all of them.
 186
 187         video_ids = []
 188         pagenum = 1
 189
 190         while True:
 191             url = page_base + "&page=" + str(pagenum)
 192             page = self._download_webpage(
 193                 url, username, 'Downloading video ids from page %d' % pagenum)
 194
 195             # Extract video identifiers
 196             ids_in_page = []
 197
 198             for mobj in re.finditer(r'href="/([^"]+)"', page):
 199                 if mobj.group(1) not in ids_in_page:
 200                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 201
 202             video_ids.extend(ids_in_page)
 203
 204             # A little optimization - if current page is not
 205             # "full", ie. does not contain PAGE_SIZE video ids then
 206             # we can assume that this page is the last one - there
 207             # are no more ids on further pages - no need to query
 208             # again.
 209
 210             if len(ids_in_page) < self._PAGE_SIZE:
 211                 break
 212
 213             pagenum += 1
 214
 215         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 216         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 217         return [self.playlist_result(url_entries, playlist_title=username)]