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                         formats.extend(self._extract_mpd_formats(
  63                             video_url, video_id, mpd_id='dash', fatal=False))
  64                     else:
  65                         formats.append({
  66                             'url': video_url,
  67                             'format_id': format_id,
  68                             'height': int_or_none(f.get('vertical_resolution')),
  69                             'width': int_or_none(f.get('horizontal_resolution')),
  70                             'abr': float_or_none(f.get('audio_bitrate')),
  71                             'vbr': float_or_none(f.get('video_bitrate')),
  72                         })
  73         self._sort_formats(formats)
  74
  75         meta = video.get('meta', {})
  76
  77         title = self._og_search_title(webpage, default=None) or meta['title']
  78         description = self._og_search_description(webpage, default=None) or meta.get('description')
  79         duration = meta.get('length') or meta.get('lenght')
  80         timestamp = parse_iso8601(meta.get('addDate'), ' ')
  81
  82         return {
  83             'id': video_id,
  84             'title': title,
  85             'description': description,
  86             'duration': duration,
  87             'timestamp': timestamp,
  88             'formats': formats,
  89         }
  90
  91
  92 class OnetIE(OnetBaseIE):
  93     _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)'
  94     IE_NAME = 'onet.tv'
  95
  96     _TEST = {
  97         'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
  98         'md5': 'e3ffbf47590032ac3f27249204173d50',
  99         'info_dict': {
 100             'id': 'qbpyqc',
 101             'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd',
 102             'ext': 'mp4',
 103             'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd',
 104             '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...',
 105             'upload_date': '20160705',
 106             'timestamp': 1467721580,
 107         },
 108     }
 109
 110     def _real_extract(self, url):
 111         mobj = re.match(self._VALID_URL, url)
 112         display_id, video_id = mobj.group('display_id', 'id')
 113
 114         webpage = self._download_webpage(url, display_id)
 115
 116         mvp_id = self._search_mvp_id(webpage)
 117
 118         info_dict = self._extract_from_id(mvp_id, webpage)
 119         info_dict.update({
 120             'id': video_id,
 121             'display_id': display_id,
 122         })
 123
 124         return info_dict
 125
 126
 127 class OnetChannelIE(OnetBaseIE):
 128     _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/(?P<id>[a-z]+)(?:[?#]|$)'
 129     IE_NAME = 'onet.tv:channel'
 130
 131     _TEST = {
 132         'url': 'http://onet.tv/k/openerfestival',
 133         'info_dict': {
 134             'id': 'openerfestival',
 135             'title': 'Open\'er Festival Live',
 136             'description': 'Dziękujemy, że oglądaliście transmisje. Zobaczcie nasze relacje i wywiady z artystami.',
 137         },
 138         'playlist_mincount': 46,
 139     }
 140
 141     def _real_extract(self, url):
 142         channel_id = self._match_id(url)
 143
 144         webpage = self._download_webpage(url, channel_id)
 145
 146         current_clip_info = self._parse_json(self._search_regex(
 147             r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id,
 148             transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s)))
 149         video_id = remove_start(current_clip_info['ckmId'], 'mvp:')
 150         video_name = url_basename(current_clip_info['url'])
 151
 152         if self._downloader.params.get('noplaylist'):
 153             self.to_screen(
 154                 'Downloading just video %s because of --no-playlist' % video_name)
 155             return self._extract_from_id(video_id, webpage)
 156
 157         self.to_screen(
 158             'Downloading channel %s - add --no-playlist to just download video %s' % (
 159                 channel_id, video_name))
 160         matches = re.findall(
 161             r'<a[^>]+href=[\'"](https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/[0-9a-z-]+/[0-9a-z]+)',
 162             webpage)
 163         entries = [
 164             self.url_result(video_link, OnetIE.ie_key())
 165             for video_link in matches]
 166
 167         channel_title = strip_or_none(get_element_by_class('o_channelName', webpage))
 168         channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage))
 169         return self.playlist_result(entries, channel_id, channel_title, channel_description)