Raphaël G. Git Repositories - youtubedl/blob - 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             'url': 'http://blip.tv/play/gbk766dkj4Yn',
  70             'md5': 'fe0a33f022d49399a241e84a8ea8b8e3',
  71             'info_dict': {
  72                 'id': '1749452',
  73                 'ext': 'mp4',
  74                 'upload_date': '20090208',
  75                 'description': 'Witness the first appearance of the Nostalgia Critic character, as Doug reviews the movie Transformers.',
  76                 'title': 'Nostalgia Critic: Transformers',
  77                 'timestamp': 1234068723,
  78                 'uploader': 'NostalgiaCritic',
  79                 'uploader_id': '246467',
  80             }
  81         }
  82     ]
  83
  84     def _real_extract(self, url):
  85         mobj = re.match(self._VALID_URL, url)
  86         lookup_id = mobj.group('lookup_id')
  87
  88         # See https://github.com/rg3/youtube-dl/issues/857 and
  89         # https://github.com/rg3/youtube-dl/issues/4197
  90         if lookup_id:
  91             urlh = self._request_webpage(
  92                 'http://blip.tv/play/%s' % lookup_id, lookup_id, 'Resolving lookup id')
  93             url = compat_urlparse.urlparse(urlh.geturl())
  94             qs = compat_urlparse.parse_qs(url.query)
  95             mobj = re.match(self._VALID_URL, qs['file'][0])
  96
  97         video_id = mobj.group('id')
  98
  99         rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
 100
 101         def blip(s):
 102             return '{http://blip.tv/dtd/blip/1.0}%s' % s
 103
 104         def media(s):
 105             return '{http://search.yahoo.com/mrss/}%s' % s
 106
 107         def itunes(s):
 108             return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s
 109
 110         item = rss.find('channel/item')
 111
 112         video_id = item.find(blip('item_id')).text
 113         title = item.find('./title').text
 114         description = clean_html(compat_str(item.find(blip('puredescription')).text))
 115         timestamp = parse_iso8601(item.find(blip('datestamp')).text)
 116         uploader = item.find(blip('user')).text
 117         uploader_id = item.find(blip('userid')).text
 118         duration = int(item.find(blip('runtime')).text)
 119         media_thumbnail = item.find(media('thumbnail'))
 120         thumbnail = media_thumbnail.get('url') if media_thumbnail is not None else item.find(itunes('image')).text
 121         categories = [category.text for category in item.findall('category')]
 122
 123         formats = []
 124         subtitles = {}
 125
 126         media_group = item.find(media('group'))
 127         for media_content in media_group.findall(media('content')):
 128             url = media_content.get('url')
 129             role = media_content.get(blip('role'))
 130             msg = self._download_webpage(
 131                 url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
 132                 video_id, 'Resolving URL for %s' % role)
 133             real_url = compat_urlparse.parse_qs(msg.strip())['message'][0]
 134
 135             media_type = media_content.get('type')
 136             if media_type == 'text/srt' or url.endswith('.srt'):
 137                 LANGS = {
 138                     'english': 'en',
 139                 }
 140                 lang = role.rpartition('-')[-1].strip().lower()
 141                 langcode = LANGS.get(lang, lang)
 142                 subtitles[langcode] = url
 143             elif media_type.startswith('video/'):
 144                 formats.append({
 145                     'url': real_url,
 146                     'format_id': role,
 147                     'format_note': media_type,
 148                     'vcodec': media_content.get(blip('vcodec')),
 149                     'acodec': media_content.get(blip('acodec')),
 150                     'filesize': media_content.get('filesize'),
 151                     'width': int(media_content.get('width')),
 152                     'height': int(media_content.get('height')),
 153                 })
 154         self._sort_formats(formats)
 155
 156         # subtitles
 157         video_subtitles = self.extract_subtitles(video_id, subtitles)
 158         if self._downloader.params.get('listsubtitles', False):
 159             self._list_available_subtitles(video_id, subtitles)
 160             return
 161
 162         return {
 163             'id': video_id,
 164             'title': title,
 165             'description': description,
 166             'timestamp': timestamp,
 167             'uploader': uploader,
 168             'uploader_id': uploader_id,
 169             'duration': duration,
 170             'thumbnail': thumbnail,
 171             'categories': categories,
 172             'formats': formats,
 173             'subtitles': video_subtitles,
 174         }
 175
 176     def _download_subtitle_url(self, sub_lang, url):
 177         # For some weird reason, blip.tv serves a video instead of subtitles
 178         # when we request with a common UA
 179         req = compat_urllib_request.Request(url)
 180         req.add_header('Youtubedl-user-agent', 'youtube-dl')
 181         return self._download_webpage(req, None, note=False)
 182
 183
 184 class BlipTVUserIE(InfoExtractor):
 185     _VALID_URL = r'(?:(?:https?://(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
 186     _PAGE_SIZE = 12
 187     IE_NAME = 'blip.tv:user'
 188     _TEST = {
 189         'url': 'http://blip.tv/actone',
 190         'info_dict': {
 191             'id': 'actone',
 192             'title': 'Act One: The Series',
 193         },
 194         'playlist_count': 5,
 195     }
 196
 197     def _real_extract(self, url):
 198         mobj = re.match(self._VALID_URL, url)
 199         username = mobj.group(1)
 200
 201         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 202
 203         page = self._download_webpage(url, username, 'Downloading user page')
 204         mobj = re.search(r'data-users-id="([^"]+)"', page)
 205         page_base = page_base % mobj.group(1)
 206         title = self._og_search_title(page)
 207
 208         # Download video ids using BlipTV Ajax calls. Result size per
 209         # query is limited (currently to 12 videos) so we need to query
 210         # page by page until there are no video ids - it means we got
 211         # all of them.
 212
 213         video_ids = []
 214         pagenum = 1
 215
 216         while True:
 217             url = page_base + "&page=" + str(pagenum)
 218             page = self._download_webpage(
 219                 url, username, 'Downloading video ids from page %d' % pagenum)
 220
 221             # Extract video identifiers
 222             ids_in_page = []
 223
 224             for mobj in re.finditer(r'href="/([^"]+)"', page):
 225                 if mobj.group(1) not in ids_in_page:
 226                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 227
 228             video_ids.extend(ids_in_page)
 229
 230             # A little optimization - if current page is not
 231             # "full", ie. does not contain PAGE_SIZE video ids then
 232             # we can assume that this page is the last one - there
 233             # are no more ids on further pages - no need to query
 234             # again.
 235
 236             if len(ids_in_page) < self._PAGE_SIZE:
 237                 break
 238
 239             pagenum += 1
 240
 241         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 242         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 243         return self.playlist_result(
 244             url_entries, playlist_title=title, playlist_id=username)