_ 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?://)?(?:\w+\.)?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
  62         if '?' in url:
  63             cchar = '&'
  64         else:
  65             cchar = '?'
  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         self.report_extraction(mobj.group(1))
  70         info = None
  71         urlh = self._request_webpage(request, None, False,
  72             u'unable to download video info webpage')
  73
  74         if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download
  75             basename = url.split('/')[-1]
  76             title,ext = os.path.splitext(basename)
  77             title = title.decode('UTF-8')
  78             ext = ext.replace('.', '')
  79             self.report_direct_download(title)
  80             return {
  81                 'id': title,
  82                 'url': url,
  83                 'uploader': None,
  84                 'upload_date': None,
  85                 'title': title,
  86                 'ext': ext,
  87                 'urlhandle': urlh
  88             }
  89
  90         try:
  91             json_code_bytes = urlh.read()
  92             json_code = json_code_bytes.decode('utf-8')
  93         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  94             raise ExtractorError(u'Unable to read video info webpage: %s' % compat_str(err))
  95
  96         try:
  97             json_data = json.loads(json_code)
  98             if 'Post' in json_data:
  99                 data = json_data['Post']
 100             else:
 101                 data = json_data
 102
 103             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
 104             if 'additionalMedia' in data:
 105                 formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height']))
 106                 best_format = formats[-1]
 107                 video_url = best_format['url']
 108             else:
 109                 video_url = data['media']['url']
 110             umobj = re.match(self._URL_EXT, video_url)
 111             if umobj is None:
 112                 raise ValueError('Can not determine filename extension')
 113             ext = umobj.group(1)
 114
 115             return {
 116                 'id': compat_str(data['item_id']),
 117                 'url': video_url,
 118                 'uploader': data['display_name'],
 119                 'upload_date': upload_date,
 120                 'title': data['title'],
 121                 'ext': ext,
 122                 'format': data['media']['mimeType'],
 123                 'thumbnail': data['thumbnailUrl'],
 124                 'description': data['description'],
 125                 'player_url': data['embedUrl'],
 126                 'user_agent': 'iTunes/10.6.1',
 127             }
 128         except (ValueError, KeyError) as err:
 129             raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
 130
 131
 132 class BlipTVUserIE(InfoExtractor):
 133     """Information Extractor for blip.tv users."""
 134
 135     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 136     _PAGE_SIZE = 12
 137     IE_NAME = u'blip.tv:user'
 138
 139     def _real_extract(self, url):
 140         # Extract username
 141         mobj = re.match(self._VALID_URL, url)
 142         if mobj is None:
 143             raise ExtractorError(u'Invalid URL: %s' % url)
 144
 145         username = mobj.group(1)
 146
 147         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 148
 149         page = self._download_webpage(url, username, u'Downloading user page')
 150         mobj = re.search(r'data-users-id="([^"]+)"', page)
 151         page_base = page_base % mobj.group(1)
 152
 153
 154         # Download video ids using BlipTV Ajax calls. Result size per
 155         # query is limited (currently to 12 videos) so we need to query
 156         # page by page until there are no video ids - it means we got
 157         # all of them.
 158
 159         video_ids = []
 160         pagenum = 1
 161
 162         while True:
 163             url = page_base + "&page=" + str(pagenum)
 164             page = self._download_webpage(url, username,
 165                                           u'Downloading video ids from page %d' % pagenum)
 166
 167             # Extract video identifiers
 168             ids_in_page = []
 169
 170             for mobj in re.finditer(r'href="/([^"]+)"', page):
 171                 if mobj.group(1) not in ids_in_page:
 172                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 173
 174             video_ids.extend(ids_in_page)
 175
 176             # A little optimization - if current page is not
 177             # "full", ie. does not contain PAGE_SIZE video ids then
 178             # we can assume that this page is the last one - there
 179             # are no more ids on further pages - no need to query
 180             # again.
 181
 182             if len(ids_in_page) < self._PAGE_SIZE:
 183                 break
 184
 185             pagenum += 1
 186
 187         urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
 188         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 189         return [self.playlist_result(url_entries, playlist_title = username)]