youtube_dl/extractor/onet.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     float_or_none,
  11     get_element_by_class,
  12     int_or_none,
  13     js_to_json,
  14     parse_iso8601,
  15     remove_start,
  16     strip_or_none,
  17     url_basename,
  18 )
  19
  20
  21 class OnetBaseIE(InfoExtractor):
  22     def _search_mvp_id(self, webpage):
  23         return self._search_regex(
  24             r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
  25
  26     def _extract_from_id(self, video_id, webpage):
  27         response = self._download_json(
  28             'http://qi.ckm.onetapi.pl/', video_id,
  29             query={
  30                 'body[id]': video_id,
  31                 'body[jsonrpc]': '2.0',
  32                 'body[method]': 'get_asset_detail',
  33                 'body[params][ID_Publikacji]': video_id,
  34                 'body[params][Service]': 'www.onet.pl',
  35                 'content-type': 'application/jsonp',
  36                 'x-onet-app': 'player.front.onetapi.pl',
  37             })
  38
  39         error = response.get('error')
  40         if error:
  41             raise ExtractorError(
  42                 '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
  43
  44         video = response['result'].get('0')
  45
  46         formats = []
  47         for _, formats_dict in video['formats'].items():
  48             if not isinstance(formats_dict, dict):
  49                 continue
  50             for format_id, format_list in formats_dict.items():
  51                 if not isinstance(format_list, list):
  52                     continue
  53                 for f in format_list:
  54                     video_url = f.get('url')
  55                     if not video_url:
  56                         continue
  57                     ext = determine_ext(video_url)
  58                     if format_id == 'ism':
  59                         # TODO: Support Microsoft Smooth Streaming
  60                         continue
  61                     elif ext == 'mpd':
  62                         # TODO: Current DASH formats are broken - $Time$ pattern in
  63                         # <SegmentTemplate> not implemented yet
  64                         # formats.extend(self._extract_mpd_formats(
  65                         #    video_url, video_id, mpd_id='dash', fatal=False))
  66                         continue
  67                     else:
  68                         formats.append({
  69                             'url': video_url,
  70                             'format_id': format_id,
  71                             'height': int_or_none(f.get('vertical_resolution')),
  72                             'width': int_or_none(f.get('horizontal_resolution')),
  73                             'abr': float_or_none(f.get('audio_bitrate')),
  74                             'vbr': float_or_none(f.get('video_bitrate')),
  75                         })
  76         self._sort_formats(formats)
  77
  78         meta = video.get('meta', {})
  79
  80         title = self._og_search_title(webpage, default=None) or meta['title']
  81         description = self._og_search_description(webpage, default=None) or meta.get('description')
  82         duration = meta.get('length') or meta.get('lenght')
  83         timestamp = parse_iso8601(meta.get('addDate'), ' ')
  84
  85         return {
  86             'id': video_id,
  87             'title': title,
  88             'description': description,
  89             'duration': duration,
  90             'timestamp': timestamp,
  91             'formats': formats,
  92         }
  93
  94
  95 class OnetIE(OnetBaseIE):
  96     _VALID_URL = 'https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)'
  97     IE_NAME = 'onet.tv'
  98
  99     _TEST = {
 100         'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
 101         'info_dict': {
 102             'id': 'qbpyqc',
 103             'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd',
 104             'ext': 'mp4',
 105             'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd',
 106             'description': 'Trzy samochody, których nigdy nie użyto, prywatne spa, hotel dekorowany czarnym suknem czy nielegalne używki. Organizatorzy koncertów i festiwali muszą stawać przed nie lada wyzwaniem zapraszając gwia...',
 107             'upload_date': '20160705',
 108             'timestamp': 1467721580,
 109         },
 110     }
 111
 112     def _real_extract(self, url):
 113         mobj = re.match(self._VALID_URL, url)
 114         display_id, video_id = mobj.group('display_id', 'id')
 115
 116         webpage = self._download_webpage(url, display_id)
 117
 118         mvp_id = self._search_mvp_id(webpage)
 119
 120         info_dict = self._extract_from_id(mvp_id, webpage)
 121         info_dict.update({
 122             'id': video_id,
 123             'display_id': display_id,
 124         })
 125
 126         return info_dict
 127
 128
 129 class OnetChannelIE(OnetBaseIE):
 130     _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/(?P<id>[a-z]+)(?:[?#]|$)'
 131     IE_NAME = 'onet.tv:channel'
 132
 133     _TEST = {
 134         'url': 'http://onet.tv/k/openerfestival',
 135         'info_dict': {
 136             'id': 'openerfestival',
 137             'title': 'Open\'er Festival Live',
 138             'description': 'Dziękujemy, że oglądaliście transmisje. Zobaczcie nasze relacje i wywiady z artystami.',
 139         },
 140         'playlist_mincount': 46,
 141     }
 142
 143     def _real_extract(self, url):
 144         channel_id = self._match_id(url)
 145
 146         webpage = self._download_webpage(url, channel_id)
 147
 148         current_clip_info = self._parse_json(self._search_regex(
 149             r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id,
 150             transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s)))
 151         video_id = remove_start(current_clip_info['ckmId'], 'mvp:')
 152         video_name = url_basename(current_clip_info['url'])
 153
 154         if self._downloader.params.get('noplaylist'):
 155             self.to_screen(
 156                 'Downloading just video %s because of --no-playlist' % video_name)
 157             return self._extract_from_id(video_id, webpage)
 158
 159         self.to_screen(
 160             'Downloading channel %s - add --no-playlist to just download video %s' % (
 161                 channel_id, video_name))
 162         matches = re.findall(
 163             r'<a[^>]+href=[\'"](https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/[0-9a-z-]+/[0-9a-z]+)',
 164             webpage)
 165         entries = [
 166             self.url_result(video_link, OnetIE.ie_key())
 167             for video_link in matches]
 168
 169         channel_title = strip_or_none(get_element_by_class('o_channelName', webpage))
 170         channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage))
 171         return self.playlist_result(entries, channel_id, channel_title, channel_description)