yt_dlp/extractor/dvtv.py

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