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