_ Git - youtube-dl/blob - youtube_dl/extractor/bliptv.py

   1 from __future__ import unicode_literals
   2
   3 import datetime
   4 import json
   5 import re
   6 import socket
   7
   8 from .common import InfoExtractor
   9 from .subtitles import SubtitlesInfoExtractor
  10 from ..utils import (
  11     compat_http_client,
  12     compat_str,
  13     compat_urllib_error,
  14     compat_urllib_request,
  15
  16     ExtractorError,
  17     unescapeHTML,
  18 )
  19
  20
  21 class BlipTVIE(SubtitlesInfoExtractor):
  22     """Information extractor for blip.tv"""
  23
  24     _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(?P<presumptive_id>.+)$'
  25
  26     _TESTS = [{
  27         'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  28         'md5': 'c6934ad0b6acf2bd920720ec888eb812',
  29         'info_dict': {
  30             'id': '5779306',
  31             'ext': 'mov',
  32             'upload_date': '20111205',
  33             'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
  34             'uploader': 'Comic Book Resources - CBR TV',
  35             'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
  36         }
  37     }, {
  38         # https://github.com/rg3/youtube-dl/pull/2274
  39         'note': 'Video with subtitles',
  40         'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
  41         'md5': '309f9d25b820b086ca163ffac8031806',
  42         'info_dict': {
  43             'id': '6586561',
  44             'ext': 'mp4',
  45             'uploader': 'Red vs. Blue',
  46             'description': 'One-Zero-One',
  47             'upload_date': '20130614',
  48             'title': 'Red vs. Blue Season 11 Episode 1',
  49         }
  50     }]
  51
  52     def _real_extract(self, url):
  53         mobj = re.match(self._VALID_URL, url)
  54         presumptive_id = mobj.group('presumptive_id')
  55
  56         # See https://github.com/rg3/youtube-dl/issues/857
  57         embed_mobj = re.match(r'https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
  58         if embed_mobj:
  59             info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
  60             info_page = self._download_webpage(info_url, embed_mobj.group(1))
  61             video_id = self._search_regex(
  62                 r'data-episode-id="([0-9]+)', info_page, 'video_id')
  63             return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
  64
  65         cchar = '&' if '?' in url else '?'
  66         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  67         request = compat_urllib_request.Request(json_url)
  68         request.add_header('User-Agent', 'iTunes/10.6.1')
  69
  70         json_data = self._download_json(request, video_id=presumptive_id)
  71
  72         if 'Post' in json_data:
  73             data = json_data['Post']
  74         else:
  75             data = json_data
  76
  77         video_id = compat_str(data['item_id'])
  78         upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  79         subtitles = {}
  80         formats = []
  81         if 'additionalMedia' in data:
  82             for f in data['additionalMedia']:
  83                 if f.get('file_type_srt') == 1:
  84                     LANGS = {
  85                         'english': 'en',
  86                     }
  87                     lang = f['role'].rpartition('-')[-1].strip().lower()
  88                     langcode = LANGS.get(lang, lang)
  89                     subtitles[langcode] = f['url']
  90                     continue
  91                 if not int(f['media_width']):  # filter m3u8
  92                     continue
  93                 formats.append({
  94                     'url': f['url'],
  95                     'format_id': f['role'],
  96                     'width': int(f['media_width']),
  97                     'height': int(f['media_height']),
  98                 })
  99         else:
 100             formats.append({
 101                 'url': data['media']['url'],
 102                 'width': int(data['media']['width']),
 103                 'height': int(data['media']['height']),
 104             })
 105         self._sort_formats(formats)
 106
 107         # subtitles
 108         video_subtitles = self.extract_subtitles(video_id, subtitles)
 109         if self._downloader.params.get('listsubtitles', False):
 110             self._list_available_subtitles(video_id, subtitles)
 111             return
 112
 113         return {
 114             'id': video_id,
 115             'uploader': data['display_name'],
 116             'upload_date': upload_date,
 117             'title': data['title'],
 118             'thumbnail': data['thumbnailUrl'],
 119             'description': data['description'],
 120             'user_agent': 'iTunes/10.6.1',
 121             'formats': formats,
 122             'subtitles': video_subtitles,
 123         }
 124
 125     def _download_subtitle_url(self, sub_lang, url):
 126         # For some weird reason, blip.tv serves a video instead of subtitles
 127         # when we request with a common UA
 128         req = compat_urllib_request.Request(url)
 129         req.add_header('Youtubedl-user-agent', 'youtube-dl')
 130         return self._download_webpage(req, None, note=False)
 131
 132
 133 class BlipTVUserIE(InfoExtractor):
 134     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 135     _PAGE_SIZE = 12
 136     IE_NAME = 'blip.tv:user'
 137
 138     def _real_extract(self, url):
 139         mobj = re.match(self._VALID_URL, url)
 140         username = mobj.group(1)
 141
 142         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 143
 144         page = self._download_webpage(url, username, 'Downloading user page')
 145         mobj = re.search(r'data-users-id="([^"]+)"', page)
 146         page_base = page_base % mobj.group(1)
 147
 148         # Download video ids using BlipTV Ajax calls. Result size per
 149         # query is limited (currently to 12 videos) so we need to query
 150         # page by page until there are no video ids - it means we got
 151         # all of them.
 152
 153         video_ids = []
 154         pagenum = 1
 155
 156         while True:
 157             url = page_base + "&page=" + str(pagenum)
 158             page = self._download_webpage(
 159                 url, username, 'Downloading video ids from page %d' % pagenum)
 160
 161             # Extract video identifiers
 162             ids_in_page = []
 163
 164             for mobj in re.finditer(r'href="/([^"]+)"', page):
 165                 if mobj.group(1) not in ids_in_page:
 166                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 167
 168             video_ids.extend(ids_in_page)
 169
 170             # A little optimization - if current page is not
 171             # "full", ie. does not contain PAGE_SIZE video ids then
 172             # we can assume that this page is the last one - there
 173             # are no more ids on further pages - no need to query
 174             # again.
 175
 176             if len(ids_in_page) < self._PAGE_SIZE:
 177                 break
 178
 179             pagenum += 1
 180
 181         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 182         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 183         return [self.playlist_result(url_entries, playlist_title=username)]