_ 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 ..utils import (
  10     compat_http_client,
  11     compat_str,
  12     compat_urllib_error,
  13     compat_urllib_request,
  14
  15     ExtractorError,
  16     unescapeHTML,
  17 )
  18
  19
  20 class BlipTVIE(InfoExtractor):
  21     """Information extractor for blip.tv"""
  22
  23     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
  24
  25     _TESTS = [{
  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             'upload_date': '20111205',
  32             'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
  33             'uploader': 'Comic Book Resources - CBR TV',
  34             'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
  35         }
  36     }, {
  37         # With subtitles (ignored) #2274
  38         'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
  39         'md5': '309f9d25b820b086ca163ffac8031806',
  40         'info_dict': {
  41             "id": "6586561",
  42             "ext": "mp4",
  43             "upload_date": "20130614",
  44             "uploader": "Red vs. Blue",
  45             "title": "Red vs. Blue Season 11 Episode 1",
  46             "description": "One-Zero-One"
  47         }
  48     }]
  49
  50     def report_direct_download(self, title):
  51         """Report information extraction."""
  52         self.to_screen('%s: Direct download detected' % title)
  53
  54     def _real_extract(self, url):
  55         mobj = re.match(self._VALID_URL, url)
  56         if mobj is None:
  57             raise ExtractorError('Invalid URL: %s' % url)
  58
  59         # See https://github.com/rg3/youtube-dl/issues/857
  60         embed_mobj = re.search(r'^(?:https?://)?(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
  61         if embed_mobj:
  62             info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
  63             info_page = self._download_webpage(info_url, embed_mobj.group(1))
  64             video_id = self._search_regex(r'data-episode-id="(\d+)', info_page,  'video_id')
  65             return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
  66
  67         if '?' in url:
  68             cchar = '&'
  69         else:
  70             cchar = '?'
  71         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  72         request = compat_urllib_request.Request(json_url)
  73         request.add_header('User-Agent', 'iTunes/10.6.1')
  74         self.report_extraction(mobj.group(1))
  75         urlh = self._request_webpage(request, None, False,
  76             'unable to download video info webpage')
  77
  78         try:
  79             json_code_bytes = urlh.read()
  80             json_code = json_code_bytes.decode('utf-8')
  81         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  82             raise ExtractorError('Unable to read video info webpage: %s' % compat_str(err))
  83
  84         try:
  85             json_data = json.loads(json_code)
  86             if 'Post' in json_data:
  87                 data = json_data['Post']
  88             else:
  89                 data = json_data
  90
  91             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  92             formats = []
  93             if 'additionalMedia' in data:
  94                 for f in sorted(data['additionalMedia'], key=lambda f: int(0 if f['media_height'] is None else f['media_height'])):
  95                     if not int(0 if f['media_height'] is None else f['media_height']): # filter out m3u8 and srt files
  96                         continue
  97                     formats.append({
  98                         'url': f['url'],
  99                         'format_id': f['role'],
 100                         'width': int(f['media_width']),
 101                         'height': int(f['media_height']),
 102                     })
 103             else:
 104                 formats.append({
 105                     'url': data['media']['url'],
 106                     'width': int(data['media']['width']),
 107                     'height': int(data['media']['height']),
 108                 })
 109
 110             self._sort_formats(formats)
 111
 112             return {
 113                 'id': compat_str(data['item_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             }
 122         except (ValueError, KeyError) as err:
 123             raise ExtractorError('Unable to parse video information: %s' % repr(err))
 124
 125
 126 class BlipTVUserIE(InfoExtractor):
 127     """Information Extractor for blip.tv users."""
 128
 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         # Extract username
 135         mobj = re.match(self._VALID_URL, url)
 136         if mobj is None:
 137             raise ExtractorError('Invalid URL: %s' % url)
 138
 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
 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(url, username,
 159                                           '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)]