yt_dlp/extractor/dvtv.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     int_or_none,
  11     join_nonempty,
  12     js_to_json,
  13     mimetype2ext,
  14     try_get,
  15     unescapeHTML,
  16     parse_iso8601,
  17 )
  18
  19
  20 class DVTVIE(InfoExtractor):
  21     IE_NAME = 'dvtv'
  22     IE_DESC = 'http://video.aktualne.cz/'
  23     _VALID_URL = r'https?://video\.aktualne\.cz/(?:[^/]+/)+r~(?P<id>[0-9a-f]{32})'
  24     _TESTS = [{
  25         'url': 'http://video.aktualne.cz/dvtv/vondra-o-ceskem-stoleti-pri-pohledu-na-havla-mi-bylo-trapne/r~e5efe9ca855511e4833a0025900fea04/',
  26         'md5': '67cb83e4a955d36e1b5d31993134a0c2',
  27         'info_dict': {
  28             'id': 'dc0768de855511e49e4b0025900fea04',
  29             'ext': 'mp4',
  30             'title': 'Vondra o Českém století: Při pohledu na Havla mi bylo trapně',
  31             'duration': 1484,
  32             'upload_date': '20141217',
  33             'timestamp': 1418792400,
  34         }
  35     }, {
  36         'url': 'http://video.aktualne.cz/dvtv/dvtv-16-12-2014-utok-talibanu-boj-o-kliniku-uprchlici/r~973eb3bc854e11e498be002590604f2e/',
  37         'info_dict': {
  38             'title': r'DVTV 16. 12. 2014: útok Talibanu, boj o kliniku, uprchlíci',
  39             'id': '973eb3bc854e11e498be002590604f2e',
  40         },
  41         'playlist': [{
  42             'md5': 'da7ca6be4935532241fa9520b3ad91e4',
  43             'info_dict': {
  44                 'id': 'b0b40906854d11e4bdad0025900fea04',
  45                 'ext': 'mp4',
  46                 'title': 'Drtinová Veselovský TV 16. 12. 2014: Témata dne',
  47                 'description': 'md5:0916925dea8e30fe84222582280b47a0',
  48                 'timestamp': 1418760010,
  49                 'upload_date': '20141216',
  50             }
  51         }, {
  52             'md5': '5f7652a08b05009c1292317b449ffea2',
  53             'info_dict': {
  54                 'id': '420ad9ec854a11e4bdad0025900fea04',
  55                 'ext': 'mp4',
  56                 'title': 'Školní masakr možná změní boj s Talibanem, říká novinářka',
  57                 'description': 'md5:ff2f9f6de73c73d7cef4f756c1c1af42',
  58                 'timestamp': 1418760010,
  59                 'upload_date': '20141216',
  60             }
  61         }, {
  62             'md5': '498eb9dfa97169f409126c617e2a3d64',
  63             'info_dict': {
  64                 'id': '95d35580846a11e4b6d20025900fea04',
  65                 'ext': 'mp4',
  66                 'title': 'Boj o kliniku: Veřejný zájem, nebo právo na majetek?',
  67                 'description': 'md5:889fe610a70fee5511dc3326a089188e',
  68                 'timestamp': 1418760010,
  69                 'upload_date': '20141216',
  70             }
  71         }, {
  72             'md5': 'b8dc6b744844032dab6ba3781a7274b9',
  73             'info_dict': {
  74                 'id': '6fe14d66853511e4833a0025900fea04',
  75                 'ext': 'mp4',
  76                 'title': 'Pánek: Odmítání syrských uprchlíků je ostudou české vlády',
  77                 'description': 'md5:544f86de6d20c4815bea11bf2ac3004f',
  78                 'timestamp': 1418760010,
  79                 'upload_date': '20141216',
  80             }
  81         }],
  82     }, {
  83         'url': 'https://video.aktualne.cz/dvtv/zeman-si-jen-leci-mindraky-sobotku-nenavidi-a-babis-se-mu-te/r~960cdb3a365a11e7a83b0025900fea04/',
  84         'md5': 'f8efe9656017da948369aa099788c8ea',
  85         'info_dict': {
  86             'id': '3c496fec365911e7a6500025900fea04',
  87             'ext': 'mp4',
  88             'title': 'Zeman si jen léčí mindráky, Sobotku nenávidí a Babiš se mu teď hodí, tvrdí Kmenta',
  89             'duration': 1103,
  90             'upload_date': '20170511',
  91             'timestamp': 1494514200,
  92         },
  93         'params': {
  94             'skip_download': True,
  95         },
  96     }, {
  97         'url': 'http://video.aktualne.cz/v-cechach-poprve-zazni-zelenkova-zrestaurovana-mse/r~45b4b00483ec11e4883b002590604f2e/',
  98         'only_matching': True,
  99     }, {
 100         # Test live stream video (liveStarter) parsing
 101         'url': 'https://video.aktualne.cz/dvtv/zive-mistryne-sveta-eva-samkova-po-navratu-ze-sampionatu/r~182654c2288811e990fd0cc47ab5f122/',
 102         'md5': '2e552e483f2414851ca50467054f9d5d',
 103         'info_dict': {
 104             'id': '8d116360288011e98c840cc47ab5f122',
 105             'ext': 'mp4',
 106             'title': 'Živě: Mistryně světa Eva Samková po návratu ze šampionátu',
 107             'upload_date': '20190204',
 108             'timestamp': 1549289591,
 109         },
 110         'params': {
 111             # Video content is no longer available
 112             'skip_download': True,
 113         },
 114     }]
 115
 116     def _parse_video_metadata(self, js, video_id, timestamp):
 117         data = self._parse_json(js, video_id, transform_source=js_to_json)
 118         title = unescapeHTML(data['title'])
 119
 120         live_starter = try_get(data, lambda x: x['plugins']['liveStarter'], dict)
 121         if live_starter:
 122             data.update(live_starter)
 123
 124         formats = []
 125         for tracks in data.get('tracks', {}).values():
 126             for video in tracks:
 127                 video_url = video.get('src')
 128                 if not video_url:
 129                     continue
 130                 video_type = video.get('type')
 131                 ext = determine_ext(video_url, mimetype2ext(video_type))
 132                 if video_type == 'application/vnd.apple.mpegurl' or ext == 'm3u8':
 133                     formats.extend(self._extract_m3u8_formats(
 134                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 135                         m3u8_id='hls', fatal=False))
 136                 elif video_type == 'application/dash+xml' or ext == 'mpd':
 137                     formats.extend(self._extract_mpd_formats(
 138                         video_url, video_id, mpd_id='dash', fatal=False))
 139                 else:
 140                     label = video.get('label')
 141                     height = self._search_regex(
 142                         r'^(\d+)[pP]', label or '', 'height', default=None)
 143                     formats.append({
 144                         'url': video_url,
 145                         'format_id': join_nonempty('http', ext, label),
 146                         'height': int_or_none(height),
 147                     })
 148         self._sort_formats(formats)
 149
 150         return {
 151             'id': data.get('mediaid') or video_id,
 152             'title': title,
 153             'description': data.get('description'),
 154             'thumbnail': data.get('image'),
 155             'duration': int_or_none(data.get('duration')),
 156             'timestamp': int_or_none(timestamp),
 157             'formats': formats
 158         }
 159
 160     def _real_extract(self, url):
 161         video_id = self._match_id(url)
 162         webpage = self._download_webpage(url, video_id)
 163         timestamp = parse_iso8601(self._html_search_meta(
 164             'article:published_time', webpage, 'published time', default=None))
 165
 166         items = re.findall(r'(?s)playlist\.push\(({.+?})\);', webpage)
 167         if items:
 168             return self.playlist_result(
 169                 [self._parse_video_metadata(i, video_id, timestamp) for i in items],
 170                 video_id, self._html_search_meta('twitter:title', webpage))
 171
 172         item = self._search_regex(
 173             r'(?s)BBXPlayer\.setup\((.+?)\);',
 174             webpage, 'video', default=None)
 175         if item:
 176             # remove function calls (ex. htmldeentitize)
 177             # TODO this should be fixed in a general way in the js_to_json
 178             item = re.sub(r'\w+?\((.+)\)', r'\1', item)
 179             return self._parse_video_metadata(item, video_id, timestamp)
 180
 181         raise ExtractorError('Could not find neither video nor playlist')