_ 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         json_data = self._download_json(request, video_id=presumptive_id)
  70
  71         if 'Post' in json_data:
  72             data = json_data['Post']
  73         else:
  74             data = json_data
  75
  76         video_id = compat_str(data['item_id'])
  77         upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  78         subtitles = {}
  79         formats = []
  80         if 'additionalMedia' in data:
  81             for f in data['additionalMedia']:
  82                 if f.get('file_type_srt') == 1:
  83                     LANGS = {
  84                         'english': 'en',
  85                     }
  86                     lang = f['role'].rpartition('-')[-1].strip().lower()
  87                     langcode = LANGS.get(lang, lang)
  88                     subtitles[langcode] = f['url']
  89                     continue
  90                 if not int(f['media_width']):  # filter m3u8
  91                     continue
  92                 formats.append({
  93                     'url': f['url'],
  94                     'format_id': f['role'],
  95                     'width': int(f['media_width']),
  96                     'height': int(f['media_height']),
  97                 })
  98         else:
  99             formats.append({
 100                 'url': data['media']['url'],
 101                 'width': int(data['media']['width']),
 102                 'height': int(data['media']['height']),
 103             })
 104         self._sort_formats(formats)
 105
 106         # subtitles
 107         video_subtitles = self.extract_subtitles(video_id, subtitles)
 108         if self._downloader.params.get('listsubtitles', False):
 109             self._list_available_subtitles(video_id, subtitles)
 110             return
 111
 112         return {
 113             'id': video_id,
 114             'uploader': data['display_name'],
 115             'upload_date': upload_date,
 116             'title': data['title'],
 117             'thumbnail': data['thumbnailUrl'],
 118             'description': data['description'],
 119             'user_agent': 'iTunes/10.6.1',
 120             'formats': formats,
 121             'subtitles': video_subtitles,
 122         }
 123
 124     def _download_subtitle_url(self, sub_lang, url):
 125         # For some weird reason, blip.tv serves a video instead of subtitles
 126         # when we request with a common UA
 127         req = compat_urllib_request.Request(url)
 128         req.add_header('Youtubedl-user-agent', 'youtube-dl')
 129         return self._download_webpage(req, None, note=False)
 130
 131
 132 class BlipTVUserIE(InfoExtractor):
 133     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 134     _PAGE_SIZE = 12
 135     IE_NAME = 'blip.tv:user'
 136
 137     def _real_extract(self, url):
 138         mobj = re.match(self._VALID_URL, url)
 139         username = mobj.group(1)
 140
 141         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 142
 143         page = self._download_webpage(url, username, 'Downloading user page')
 144         mobj = re.search(r'data-users-id="([^"]+)"', page)
 145         page_base = page_base % mobj.group(1)
 146
 147         # Download video ids using BlipTV Ajax calls. Result size per
 148         # query is limited (currently to 12 videos) so we need to query
 149         # page by page until there are no video ids - it means we got
 150         # all of them.
 151
 152         video_ids = []
 153         pagenum = 1
 154
 155         while True:
 156             url = page_base + "&page=" + str(pagenum)
 157             page = self._download_webpage(
 158                 url, username, 'Downloading video ids from page %d' % pagenum)
 159
 160             # Extract video identifiers
 161             ids_in_page = []
 162
 163             for mobj in re.finditer(r'href="/([^"]+)"', page):
 164                 if mobj.group(1) not in ids_in_page:
 165                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 166
 167             video_ids.extend(ids_in_page)
 168
 169             # A little optimization - if current page is not
 170             # "full", ie. does not contain PAGE_SIZE video ids then
 171             # we can assume that this page is the last one - there
 172             # are no more ids on further pages - no need to query
 173             # again.
 174
 175             if len(ids_in_page) < self._PAGE_SIZE:
 176                 break
 177
 178             pagenum += 1
 179
 180         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 181         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 182         return [self.playlist_result(url_entries, playlist_title=username)]