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
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_request,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     clean_html,
  15     int_or_none,
  16     parse_iso8601,
  17     unescapeHTML,
  18 )
  19
  20
  21 class BlipTVIE(SubtitlesInfoExtractor):
  22     _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_]+)))'
  23
  24     _TESTS = [
  25         {
  26             'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  27             'md5': 'c6934ad0b6acf2bd920720ec888eb812',
  28             'info_dict': {
  29                 'id': '5779306',
  30                 'ext': 'mov',
  31                 'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
  32                 'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
  33                 'timestamp': 1323138843,
  34                 'upload_date': '20111206',
  35                 'uploader': 'cbr',
  36                 'uploader_id': '679425',
  37                 'duration': 81,
  38             }
  39         },
  40         {
  41             # https://github.com/rg3/youtube-dl/pull/2274
  42             'note': 'Video with subtitles',
  43             'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
  44             'md5': '309f9d25b820b086ca163ffac8031806',
  45             'info_dict': {
  46                 'id': '6586561',
  47                 'ext': 'mp4',
  48                 'title': 'Red vs. Blue Season 11 Episode 1',
  49                 'description': 'One-Zero-One',
  50                 'timestamp': 1371261608,
  51                 'upload_date': '20130615',
  52                 'uploader': 'redvsblue',
  53                 'uploader_id': '792887',
  54                 'duration': 279,
  55             }
  56         },
  57         {
  58             # https://bugzilla.redhat.com/show_bug.cgi?id=967465
  59             'url': 'http://a.blip.tv/api.swf#h6Uag5KbVwI',
  60             'md5': '314e87b1ebe7a48fcbfdd51b791ce5a6',
  61             'info_dict': {
  62                 'id': '6573122',
  63                 'ext': 'mov',
  64                 'upload_date': '20130520',
  65                 'description': 'Two hapless space marines argue over what to do when they realize they have an astronomically huge problem on their hands.',
  66                 'title': 'Red vs. Blue Season 11 Trailer',
  67                 'timestamp': 1369029609,
  68                 'uploader': 'redvsblue',
  69                 'uploader_id': '792887',
  70             }
  71         },
  72         {
  73             'url': 'http://blip.tv/play/gbk766dkj4Yn',
  74             'md5': 'fe0a33f022d49399a241e84a8ea8b8e3',
  75             'info_dict': {
  76                 'id': '1749452',
  77                 'ext': 'mp4',
  78                 'upload_date': '20090208',
  79                 'description': 'Witness the first appearance of the Nostalgia Critic character, as Doug reviews the movie Transformers.',
  80                 'title': 'Nostalgia Critic: Transformers',
  81                 'timestamp': 1234068723,
  82                 'uploader': 'NostalgiaCritic',
  83                 'uploader_id': '246467',
  84             }
  85         },
  86         {
  87             # https://github.com/rg3/youtube-dl/pull/4404
  88             'note': 'Audio only',
  89             'url': 'http://blip.tv/hilarios-productions/weekly-manga-recap-kingdom-7119982',
  90             'md5': '76c0a56f24e769ceaab21fbb6416a351',
  91             'info_dict': {
  92                 'id': '7103299',
  93                 'ext': 'flv',
  94                 'title': 'Weekly Manga Recap: Kingdom',
  95                 'description': 'And then Shin breaks the enemy line, and he&apos;s all like HWAH! And then he slices a guy and it&apos;s all like FWASHING! And... it&apos;s really hard to describe the best parts of this series without breaking down into sound effects, okay?',
  96                 'timestamp': 1417660321,
  97                 'upload_date': '20141204',
  98                 'uploader': 'The Rollo T',
  99                 'uploader_id': '407429',
 100                 'duration': 7251,
 101                 'vcodec': 'none',
 102             }
 103         },
 104     ]
 105
 106     def _real_extract(self, url):
 107         mobj = re.match(self._VALID_URL, url)
 108         lookup_id = mobj.group('lookup_id')
 109
 110         # See https://github.com/rg3/youtube-dl/issues/857 and
 111         # https://github.com/rg3/youtube-dl/issues/4197
 112         if lookup_id:
 113             urlh = self._request_webpage(
 114                 'http://blip.tv/play/%s' % lookup_id, lookup_id, 'Resolving lookup id')
 115             url = compat_urlparse.urlparse(urlh.geturl())
 116             qs = compat_urlparse.parse_qs(url.query)
 117             mobj = re.match(self._VALID_URL, qs['file'][0])
 118
 119         video_id = mobj.group('id')
 120
 121         rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
 122
 123         def blip(s):
 124             return '{http://blip.tv/dtd/blip/1.0}%s' % s
 125
 126         def media(s):
 127             return '{http://search.yahoo.com/mrss/}%s' % s
 128
 129         def itunes(s):
 130             return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s
 131
 132         item = rss.find('channel/item')
 133
 134         video_id = item.find(blip('item_id')).text
 135         title = item.find('./title').text
 136         description = clean_html(compat_str(item.find(blip('puredescription')).text))
 137         timestamp = parse_iso8601(item.find(blip('datestamp')).text)
 138         uploader = item.find(blip('user')).text
 139         uploader_id = item.find(blip('userid')).text
 140         duration = int(item.find(blip('runtime')).text)
 141         media_thumbnail = item.find(media('thumbnail'))
 142         thumbnail = media_thumbnail.get('url') if media_thumbnail is not None else item.find(itunes('image')).text
 143         categories = [category.text for category in item.findall('category')]
 144
 145         formats = []
 146         subtitles = {}
 147
 148         media_group = item.find(media('group'))
 149         for media_content in media_group.findall(media('content')):
 150             url = media_content.get('url')
 151             role = media_content.get(blip('role'))
 152             msg = self._download_webpage(
 153                 url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
 154                 video_id, 'Resolving URL for %s' % role)
 155             real_url = compat_urlparse.parse_qs(msg.strip())['message'][0]
 156
 157             media_type = media_content.get('type')
 158             if media_type == 'text/srt' or url.endswith('.srt'):
 159                 LANGS = {
 160                     'english': 'en',
 161                 }
 162                 lang = role.rpartition('-')[-1].strip().lower()
 163                 langcode = LANGS.get(lang, lang)
 164                 subtitles[langcode] = url
 165             elif media_type.startswith('video/'):
 166                 formats.append({
 167                     'url': real_url,
 168                     'format_id': role,
 169                     'format_note': media_type,
 170                     'vcodec': media_content.get(blip('vcodec')) or 'none',
 171                     'acodec': media_content.get(blip('acodec')),
 172                     'filesize': media_content.get('filesize'),
 173                     'width': int_or_none(media_content.get('width')),
 174                     'height': int_or_none(media_content.get('height')),
 175                 })
 176         self._sort_formats(formats)
 177
 178         # subtitles
 179         video_subtitles = self.extract_subtitles(video_id, subtitles)
 180         if self._downloader.params.get('listsubtitles', False):
 181             self._list_available_subtitles(video_id, subtitles)
 182             return
 183
 184         return {
 185             'id': video_id,
 186             'title': title,
 187             'description': description,
 188             'timestamp': timestamp,
 189             'uploader': uploader,
 190             'uploader_id': uploader_id,
 191             'duration': duration,
 192             'thumbnail': thumbnail,
 193             'categories': categories,
 194             'formats': formats,
 195             'subtitles': video_subtitles,
 196         }
 197
 198     def _download_subtitle_url(self, sub_lang, url):
 199         # For some weird reason, blip.tv serves a video instead of subtitles
 200         # when we request with a common UA
 201         req = compat_urllib_request.Request(url)
 202         req.add_header('User-Agent', 'youtube-dl')
 203         return self._download_webpage(req, None, note=False)
 204
 205
 206 class BlipTVUserIE(InfoExtractor):
 207     _VALID_URL = r'(?:(?:https?://(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
 208     _PAGE_SIZE = 12
 209     IE_NAME = 'blip.tv:user'
 210     _TEST = {
 211         'url': 'http://blip.tv/actone',
 212         'info_dict': {
 213             'id': 'actone',
 214             'title': 'Act One: The Series',
 215         },
 216         'playlist_count': 5,
 217     }
 218
 219     def _real_extract(self, url):
 220         mobj = re.match(self._VALID_URL, url)
 221         username = mobj.group(1)
 222
 223         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 224
 225         page = self._download_webpage(url, username, 'Downloading user page')
 226         mobj = re.search(r'data-users-id="([^"]+)"', page)
 227         page_base = page_base % mobj.group(1)
 228         title = self._og_search_title(page)
 229
 230         # Download video ids using BlipTV Ajax calls. Result size per
 231         # query is limited (currently to 12 videos) so we need to query
 232         # page by page until there are no video ids - it means we got
 233         # all of them.
 234
 235         video_ids = []
 236         pagenum = 1
 237
 238         while True:
 239             url = page_base + "&page=" + str(pagenum)
 240             page = self._download_webpage(
 241                 url, username, 'Downloading video ids from page %d' % pagenum)
 242
 243             # Extract video identifiers
 244             ids_in_page = []
 245
 246             for mobj in re.finditer(r'href="/([^"]+)"', page):
 247                 if mobj.group(1) not in ids_in_page:
 248                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 249
 250             video_ids.extend(ids_in_page)
 251
 252             # A little optimization - if current page is not
 253             # "full", ie. does not contain PAGE_SIZE video ids then
 254             # we can assume that this page is the last one - there
 255             # are no more ids on further pages - no need to query
 256             # again.
 257
 258             if len(ids_in_page) < self._PAGE_SIZE:
 259                 break
 260
 261             pagenum += 1
 262
 263         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 264         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 265         return self.playlist_result(
 266             url_entries, playlist_title=title, playlist_id=username)