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