Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bliptv.py

   1 from __future__ import unicode_literals
   2
   3 import datetime
   4 import re
   5
   6 from .common import InfoExtractor
   7 from .subtitles import SubtitlesInfoExtractor
   8 from ..utils import (
   9     compat_str,
  10     compat_urllib_request,
  11
  12     unescapeHTML,
  13 )
  14
  15
  16 class BlipTVIE(SubtitlesInfoExtractor):
  17     """Information extractor for blip.tv"""
  18
  19     _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(?P<presumptive_id>.+)$'
  20
  21     _TESTS = [{
  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             'upload_date': '20111205',
  28             'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
  29             'uploader': 'Comic Book Resources - CBR TV',
  30             'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
  31         }
  32     }, {
  33         # https://github.com/rg3/youtube-dl/pull/2274
  34         'note': 'Video with subtitles',
  35         'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
  36         'md5': '309f9d25b820b086ca163ffac8031806',
  37         'info_dict': {
  38             'id': '6586561',
  39             'ext': 'mp4',
  40             'uploader': 'Red vs. Blue',
  41             'description': 'One-Zero-One',
  42             'upload_date': '20130614',
  43             'title': 'Red vs. Blue Season 11 Episode 1',
  44         }
  45     }]
  46
  47     def _real_extract(self, url):
  48         mobj = re.match(self._VALID_URL, url)
  49         presumptive_id = mobj.group('presumptive_id')
  50
  51         # See https://github.com/rg3/youtube-dl/issues/857
  52         embed_mobj = re.match(r'https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
  53         if embed_mobj:
  54             info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
  55             info_page = self._download_webpage(info_url, embed_mobj.group(1))
  56             video_id = self._search_regex(
  57                 r'data-episode-id="([0-9]+)', info_page, 'video_id')
  58             return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
  59
  60         cchar = '&' if '?' in url else '?'
  61         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  62         request = compat_urllib_request.Request(json_url)
  63         request.add_header('User-Agent', 'iTunes/10.6.1')
  64
  65         json_data = self._download_json(request, video_id=presumptive_id)
  66
  67         if 'Post' in json_data:
  68             data = json_data['Post']
  69         else:
  70             data = json_data
  71
  72         video_id = compat_str(data['item_id'])
  73         upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  74         subtitles = {}
  75         formats = []
  76         if 'additionalMedia' in data:
  77             for f in data['additionalMedia']:
  78                 if f.get('file_type_srt') == 1:
  79                     LANGS = {
  80                         'english': 'en',
  81                     }
  82                     lang = f['role'].rpartition('-')[-1].strip().lower()
  83                     langcode = LANGS.get(lang, lang)
  84                     subtitles[langcode] = f['url']
  85                     continue
  86                 if not int(f['media_width']):  # filter m3u8
  87                     continue
  88                 formats.append({
  89                     'url': f['url'],
  90                     'format_id': f['role'],
  91                     'width': int(f['media_width']),
  92                     'height': int(f['media_height']),
  93                 })
  94         else:
  95             formats.append({
  96                 'url': data['media']['url'],
  97                 'width': int(data['media']['width']),
  98                 'height': int(data['media']['height']),
  99             })
 100         self._sort_formats(formats)
 101
 102         # subtitles
 103         video_subtitles = self.extract_subtitles(video_id, subtitles)
 104         if self._downloader.params.get('listsubtitles', False):
 105             self._list_available_subtitles(video_id, subtitles)
 106             return
 107
 108         return {
 109             'id': video_id,
 110             'uploader': data['display_name'],
 111             'upload_date': upload_date,
 112             'title': data['title'],
 113             'thumbnail': data['thumbnailUrl'],
 114             'description': data['description'],
 115             'user_agent': 'iTunes/10.6.1',
 116             'formats': formats,
 117             'subtitles': video_subtitles,
 118         }
 119
 120     def _download_subtitle_url(self, sub_lang, url):
 121         # For some weird reason, blip.tv serves a video instead of subtitles
 122         # when we request with a common UA
 123         req = compat_urllib_request.Request(url)
 124         req.add_header('Youtubedl-user-agent', 'youtube-dl')
 125         return self._download_webpage(req, None, note=False)
 126
 127
 128 class BlipTVUserIE(InfoExtractor):
 129     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 130     _PAGE_SIZE = 12
 131     IE_NAME = 'blip.tv:user'
 132
 133     def _real_extract(self, url):
 134         mobj = re.match(self._VALID_URL, url)
 135         username = mobj.group(1)
 136
 137         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 138
 139         page = self._download_webpage(url, username, 'Downloading user page')
 140         mobj = re.search(r'data-users-id="([^"]+)"', page)
 141         page_base = page_base % mobj.group(1)
 142
 143         # Download video ids using BlipTV Ajax calls. Result size per
 144         # query is limited (currently to 12 videos) so we need to query
 145         # page by page until there are no video ids - it means we got
 146         # all of them.
 147
 148         video_ids = []
 149         pagenum = 1
 150
 151         while True:
 152             url = page_base + "&page=" + str(pagenum)
 153             page = self._download_webpage(
 154                 url, username, 'Downloading video ids from page %d' % pagenum)
 155
 156             # Extract video identifiers
 157             ids_in_page = []
 158
 159             for mobj in re.finditer(r'href="/([^"]+)"', page):
 160                 if mobj.group(1) not in ids_in_page:
 161                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 162
 163             video_ids.extend(ids_in_page)
 164
 165             # A little optimization - if current page is not
 166             # "full", ie. does not contain PAGE_SIZE video ids then
 167             # we can assume that this page is the last one - there
 168             # are no more ids on further pages - no need to query
 169             # again.
 170
 171             if len(ids_in_page) < self._PAGE_SIZE:
 172                 break
 173
 174             pagenum += 1
 175
 176         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 177         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 178         return [self.playlist_result(url_entries, playlist_title=username)]