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

   1 import datetime
   2 import json
   3 import os
   4 import re
   5 import socket
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_http_client,
  10     compat_parse_qs,
  11     compat_str,
  12     compat_urllib_error,
  13     compat_urllib_parse_urlparse,
  14     compat_urllib_request,
  15
  16     ExtractorError,
  17     unescapeHTML,
  18 )
  19
  20
  21 class BlipTVIE(InfoExtractor):
  22     """Information extractor for blip.tv"""
  23
  24     _VALID_URL = r'^(?:https?://)?(?:www\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
  25     _URL_EXT = r'^.*\.([a-z0-9]+)$'
  26     IE_NAME = u'blip.tv'
  27     _TEST = {
  28         u'url': u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  29         u'file': u'5779306.m4v',
  30         u'md5': u'80baf1ec5c3d2019037c1c707d676b9f',
  31         u'info_dict': {
  32             u"upload_date": u"20111205",
  33             u"description": u"md5:9bc31f227219cde65e47eeec8d2dc596",
  34             u"uploader": u"Comic Book Resources - CBR TV",
  35             u"title": u"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(u'%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(u'Invalid URL: %s' % url)
  47
  48         # See https://github.com/rg3/youtube-dl/issues/857
  49         api_mobj = re.match(r'http://a\.blip\.tv/api\.swf#(?P<video_id>[\d\w]+)', url)
  50         if api_mobj is not None:
  51             url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id')
  52         urlp = compat_urllib_parse_urlparse(url)
  53         if urlp.path.startswith('/play/'):
  54             response = self._request_webpage(url, None, False)
  55             redirecturl = response.geturl()
  56             rurlp = compat_urllib_parse_urlparse(redirecturl)
  57             file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2]
  58             url = 'http://blip.tv/a/a-' + file_id
  59             return self._real_extract(url)
  60
  61         if '?' in url:
  62             cchar = '&'
  63         else:
  64             cchar = '?'
  65         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  66         request = compat_urllib_request.Request(json_url)
  67         request.add_header('User-Agent', 'iTunes/10.6.1')
  68         self.report_extraction(mobj.group(1))
  69         info = None
  70         urlh = self._request_webpage(request, None, False,
  71             u'unable to download video info webpage')
  72
  73         try:
  74             json_code_bytes = urlh.read()
  75             json_code = json_code_bytes.decode('utf-8')
  76         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  77             raise ExtractorError(u'Unable to read video info webpage: %s' % compat_str(err))
  78
  79         try:
  80             json_data = json.loads(json_code)
  81             if 'Post' in json_data:
  82                 data = json_data['Post']
  83             else:
  84                 data = json_data
  85
  86             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  87             if 'additionalMedia' in data:
  88                 formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height']))
  89                 best_format = formats[-1]
  90                 video_url = best_format['url']
  91             else:
  92                 video_url = data['media']['url']
  93             umobj = re.match(self._URL_EXT, video_url)
  94             if umobj is None:
  95                 raise ValueError('Can not determine filename extension')
  96             ext = umobj.group(1)
  97
  98             return {
  99                 'id': compat_str(data['item_id']),
 100                 'url': video_url,
 101                 'uploader': data['display_name'],
 102                 'upload_date': upload_date,
 103                 'title': data['title'],
 104                 'ext': ext,
 105                 'format': data['media']['mimeType'],
 106                 'thumbnail': data['thumbnailUrl'],
 107                 'description': data['description'],
 108                 'player_url': data['embedUrl'],
 109                 'user_agent': 'iTunes/10.6.1',
 110             }
 111         except (ValueError, KeyError) as err:
 112             raise ExtractorError(u'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 = u'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(u'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, u'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                                           u'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 = [u'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)]