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

   1 import datetime
   2 import json
   3 import re
   4 import socket
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_http_client,
   9     compat_parse_qs,
  10     compat_str,
  11     compat_urllib_error,
  12     compat_urllib_parse_urlparse,
  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#))(.+)$'
  24     _URL_EXT = r'^.*\.([a-z0-9]+)$'
  25     IE_NAME = u'blip.tv'
  26     _TEST = {
  27         u'url': u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  28         u'file': u'5779306.m4v',
  29         u'md5': u'80baf1ec5c3d2019037c1c707d676b9f',
  30         u'info_dict': {
  31             u"upload_date": u"20111205",
  32             u"description": u"md5:9bc31f227219cde65e47eeec8d2dc596",
  33             u"uploader": u"Comic Book Resources - CBR TV",
  34             u"title": u"CBR EXCLUSIVE: \"Gotham City Imposters\" Bats VS Jokerz Short 3"
  35         }
  36     }
  37
  38     def report_direct_download(self, title):
  39         """Report information extraction."""
  40         self.to_screen(u'%s: Direct download detected' % title)
  41
  42     def _real_extract(self, url):
  43         mobj = re.match(self._VALID_URL, url)
  44         if mobj is None:
  45             raise ExtractorError(u'Invalid URL: %s' % url)
  46
  47         # See https://github.com/rg3/youtube-dl/issues/857
  48         embed_mobj = re.search(r'^(?:https?://)?(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
  49         if embed_mobj:
  50             info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
  51             info_page = self._download_webpage(info_url, embed_mobj.group(1))
  52             video_id = self._search_regex(r'data-episode-id="(\d+)', info_page, u'video_id')
  53             return self.url_result('http://blip.tv/a/a-'+video_id, 'BlipTV')
  54
  55         if '?' in url:
  56             cchar = '&'
  57         else:
  58             cchar = '?'
  59         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  60         request = compat_urllib_request.Request(json_url)
  61         request.add_header('User-Agent', 'iTunes/10.6.1')
  62         self.report_extraction(mobj.group(1))
  63         urlh = self._request_webpage(request, None, False,
  64             u'unable to download video info webpage')
  65
  66         try:
  67             json_code_bytes = urlh.read()
  68             json_code = json_code_bytes.decode('utf-8')
  69         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  70             raise ExtractorError(u'Unable to read video info webpage: %s' % compat_str(err))
  71
  72         try:
  73             json_data = json.loads(json_code)
  74             if 'Post' in json_data:
  75                 data = json_data['Post']
  76             else:
  77                 data = json_data
  78
  79             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  80             if 'additionalMedia' in data:
  81                 formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height']))
  82                 best_format = formats[-1]
  83                 video_url = best_format['url']
  84             else:
  85                 video_url = data['media']['url']
  86             umobj = re.match(self._URL_EXT, video_url)
  87             if umobj is None:
  88                 raise ValueError('Can not determine filename extension')
  89             ext = umobj.group(1)
  90
  91             return {
  92                 'id': compat_str(data['item_id']),
  93                 'url': video_url,
  94                 'uploader': data['display_name'],
  95                 'upload_date': upload_date,
  96                 'title': data['title'],
  97                 'ext': ext,
  98                 'format': data['media']['mimeType'],
  99                 'thumbnail': data['thumbnailUrl'],
 100                 'description': data['description'],
 101                 'player_url': data['embedUrl'],
 102                 'user_agent': 'iTunes/10.6.1',
 103             }
 104         except (ValueError, KeyError) as err:
 105             raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
 106
 107
 108 class BlipTVUserIE(InfoExtractor):
 109     """Information Extractor for blip.tv users."""
 110
 111     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 112     _PAGE_SIZE = 12
 113     IE_NAME = u'blip.tv:user'
 114
 115     def _real_extract(self, url):
 116         # Extract username
 117         mobj = re.match(self._VALID_URL, url)
 118         if mobj is None:
 119             raise ExtractorError(u'Invalid URL: %s' % url)
 120
 121         username = mobj.group(1)
 122
 123         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 124
 125         page = self._download_webpage(url, username, u'Downloading user page')
 126         mobj = re.search(r'data-users-id="([^"]+)"', page)
 127         page_base = page_base % mobj.group(1)
 128
 129
 130         # Download video ids using BlipTV Ajax calls. Result size per
 131         # query is limited (currently to 12 videos) so we need to query
 132         # page by page until there are no video ids - it means we got
 133         # all of them.
 134
 135         video_ids = []
 136         pagenum = 1
 137
 138         while True:
 139             url = page_base + "&page=" + str(pagenum)
 140             page = self._download_webpage(url, username,
 141                                           u'Downloading video ids from page %d' % pagenum)
 142
 143             # Extract video identifiers
 144             ids_in_page = []
 145
 146             for mobj in re.finditer(r'href="/([^"]+)"', page):
 147                 if mobj.group(1) not in ids_in_page:
 148                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 149
 150             video_ids.extend(ids_in_page)
 151
 152             # A little optimization - if current page is not
 153             # "full", ie. does not contain PAGE_SIZE video ids then
 154             # we can assume that this page is the last one - there
 155             # are no more ids on further pages - no need to query
 156             # again.
 157
 158             if len(ids_in_page) < self._PAGE_SIZE:
 159                 break
 160
 161             pagenum += 1
 162
 163         urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
 164         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 165         return [self.playlist_result(url_entries, playlist_title = username)]