_ 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_parse_qs,
  12     compat_str,
  13     compat_urllib_error,
  14     compat_urllib_parse_urlparse,
  15     compat_urllib_request,
  16
  17     ExtractorError,
  18     unescapeHTML,
  19 )
  20
  21
  22 class BlipTVIE(InfoExtractor):
  23     """Information extractor for blip.tv"""
  24
  25     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
  26
  27     _TEST = {
  28         'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  29         'file': '5779306.mov',
  30         'md5': 'c6934ad0b6acf2bd920720ec888eb812',
  31         'info_dict': {
  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
  39     def report_direct_download(self, title):
  40         """Report information extraction."""
  41         self.to_screen('%s: Direct download detected' % title)
  42
  43     def _real_extract(self, url):
  44         mobj = re.match(self._VALID_URL, url)
  45         if mobj is None:
  46             raise ExtractorError('Invalid URL: %s' % url)
  47
  48         # See https://github.com/rg3/youtube-dl/issues/857
  49         embed_mobj = re.search(r'^(?:https?://)?(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
  50         if embed_mobj:
  51             info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
  52             info_page = self._download_webpage(info_url, embed_mobj.group(1))
  53             video_id = self._search_regex(r'data-episode-id="(\d+)', info_page,  'video_id')
  54             return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
  55
  56         if '?' in url:
  57             cchar = '&'
  58         else:
  59             cchar = '?'
  60         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  61         request = compat_urllib_request.Request(json_url)
  62         request.add_header('User-Agent', 'iTunes/10.6.1')
  63         self.report_extraction(mobj.group(1))
  64         urlh = self._request_webpage(request, None, False,
  65             'unable to download video info webpage')
  66
  67         try:
  68             json_code_bytes = urlh.read()
  69             json_code = json_code_bytes.decode('utf-8')
  70         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  71             raise ExtractorError('Unable to read video info webpage: %s' % compat_str(err))
  72
  73         try:
  74             json_data = json.loads(json_code)
  75             if 'Post' in json_data:
  76                 data = json_data['Post']
  77             else:
  78                 data = json_data
  79
  80             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  81             formats = []
  82             if 'additionalMedia' in data:
  83                 for f in sorted(data['additionalMedia'], key=lambda f: int(f['media_height'])):
  84                     if not int(f['media_width']): # filter m3u8
  85                         continue
  86                     formats.append({
  87                         'url': f['url'],
  88                         'format_id': f['role'],
  89                         'width': int(f['media_width']),
  90                         'height': int(f['media_height']),
  91                     })
  92             else:
  93                 formats.append({
  94                     'url': data['media']['url'],
  95                     'width': int(data['media']['width']),
  96                     'height': int(data['media']['height']),
  97                 })
  98
  99             self._sort_formats(formats)
 100
 101             return {
 102                 'id': compat_str(data['item_id']),
 103                 'uploader': data['display_name'],
 104                 'upload_date': upload_date,
 105                 'title': data['title'],
 106                 'thumbnail': data['thumbnailUrl'],
 107                 'description': data['description'],
 108                 'user_agent': 'iTunes/10.6.1',
 109                 'formats': formats,
 110             }
 111         except (ValueError, KeyError) as err:
 112             raise ExtractorError('Unable to parse video information: %s' % repr(err))
 113
 114
 115 class BlipTVUserIE(InfoExtractor):
 116     """Information Extractor for blip.tv users."""
 117
 118     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 119     _PAGE_SIZE = 12
 120     IE_NAME = 'blip.tv:user'
 121
 122     def _real_extract(self, url):
 123         # Extract username
 124         mobj = re.match(self._VALID_URL, url)
 125         if mobj is None:
 126             raise ExtractorError('Invalid URL: %s' % url)
 127
 128         username = mobj.group(1)
 129
 130         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 131
 132         page = self._download_webpage(url, username, 'Downloading user page')
 133         mobj = re.search(r'data-users-id="([^"]+)"', page)
 134         page_base = page_base % mobj.group(1)
 135
 136
 137         # Download video ids using BlipTV Ajax calls. Result size per
 138         # query is limited (currently to 12 videos) so we need to query
 139         # page by page until there are no video ids - it means we got
 140         # all of them.
 141
 142         video_ids = []
 143         pagenum = 1
 144
 145         while True:
 146             url = page_base + "&page=" + str(pagenum)
 147             page = self._download_webpage(url, username,
 148                                           'Downloading video ids from page %d' % pagenum)
 149
 150             # Extract video identifiers
 151             ids_in_page = []
 152
 153             for mobj in re.finditer(r'href="/([^"]+)"', page):
 154                 if mobj.group(1) not in ids_in_page:
 155                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 156
 157             video_ids.extend(ids_in_page)
 158
 159             # A little optimization - if current page is not
 160             # "full", ie. does not contain PAGE_SIZE video ids then
 161             # we can assume that this page is the last one - there
 162             # are no more ids on further pages - no need to query
 163             # again.
 164
 165             if len(ids_in_page) < self._PAGE_SIZE:
 166                 break
 167
 168             pagenum += 1
 169
 170         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 171         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 172         return [self.playlist_result(url_entries, playlist_title = username)]