Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/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     js_to_json,
  12     mimetype2ext,
  13     try_get,
  14     unescapeHTML,
  15     parse_iso8601,
  16 )
  17
  18
  19 class DVTVIE(InfoExtractor):
  20     IE_NAME = 'dvtv'
  21     IE_DESC = 'http://video.aktualne.cz/'
  22     _VALID_URL = r'https?://video\.aktualne\.cz/(?:[^/]+/)+r~(?P<id>[0-9a-f]{32})'
  23     _TESTS = [{
  24         'url': 'http://video.aktualne.cz/dvtv/vondra-o-ceskem-stoleti-pri-pohledu-na-havla-mi-bylo-trapne/r~e5efe9ca855511e4833a0025900fea04/',
  25         'md5': '67cb83e4a955d36e1b5d31993134a0c2',
  26         'info_dict': {
  27             'id': 'dc0768de855511e49e4b0025900fea04',
  28             'ext': 'mp4',
  29             'title': 'Vondra o Českém století: Při pohledu na Havla mi bylo trapně',
  30             'duration': 1484,
  31             'upload_date': '20141217',
  32             'timestamp': 1418792400,
  33         }
  34     }, {
  35         'url': 'http://video.aktualne.cz/dvtv/dvtv-16-12-2014-utok-talibanu-boj-o-kliniku-uprchlici/r~973eb3bc854e11e498be002590604f2e/',
  36         'info_dict': {
  37             'title': r'DVTV 16. 12. 2014: útok Talibanu, boj o kliniku, uprchlíci',
  38             'id': '973eb3bc854e11e498be002590604f2e',
  39         },
  40         'playlist': [{
  41             'md5': 'da7ca6be4935532241fa9520b3ad91e4',
  42             'info_dict': {
  43                 'id': 'b0b40906854d11e4bdad0025900fea04',
  44                 'ext': 'mp4',
  45                 'title': 'Drtinová Veselovský TV 16. 12. 2014: Témata dne',
  46                 'description': 'md5:0916925dea8e30fe84222582280b47a0',
  47                 'timestamp': 1418760010,
  48                 'upload_date': '20141216',
  49             }
  50         }, {
  51             'md5': '5f7652a08b05009c1292317b449ffea2',
  52             'info_dict': {
  53                 'id': '420ad9ec854a11e4bdad0025900fea04',
  54                 'ext': 'mp4',
  55                 'title': 'Školní masakr možná změní boj s Talibanem, říká novinářka',
  56                 'description': 'md5:ff2f9f6de73c73d7cef4f756c1c1af42',
  57                 'timestamp': 1418760010,
  58                 'upload_date': '20141216',
  59             }
  60         }, {
  61             'md5': '498eb9dfa97169f409126c617e2a3d64',
  62             'info_dict': {
  63                 'id': '95d35580846a11e4b6d20025900fea04',
  64                 'ext': 'mp4',
  65                 'title': 'Boj o kliniku: Veřejný zájem, nebo právo na majetek?',
  66                 'description': 'md5:889fe610a70fee5511dc3326a089188e',
  67                 'timestamp': 1418760010,
  68                 'upload_date': '20141216',
  69             }
  70         }, {
  71             'md5': 'b8dc6b744844032dab6ba3781a7274b9',
  72             'info_dict': {
  73                 'id': '6fe14d66853511e4833a0025900fea04',
  74                 'ext': 'mp4',
  75                 'title': 'Pánek: Odmítání syrských uprchlíků je ostudou české vlády',
  76                 'description': 'md5:544f86de6d20c4815bea11bf2ac3004f',
  77                 'timestamp': 1418760010,
  78                 'upload_date': '20141216',
  79             }
  80         }],
  81     }, {
  82         'url': 'https://video.aktualne.cz/dvtv/zeman-si-jen-leci-mindraky-sobotku-nenavidi-a-babis-se-mu-te/r~960cdb3a365a11e7a83b0025900fea04/',
  83         'md5': 'f8efe9656017da948369aa099788c8ea',
  84         'info_dict': {
  85             'id': '3c496fec365911e7a6500025900fea04',
  86             'ext': 'mp4',
  87             'title': 'Zeman si jen léčí mindráky, Sobotku nenávidí a Babiš se mu teď hodí, tvrdí Kmenta',
  88             'duration': 1103,
  89             'upload_date': '20170511',
  90             'timestamp': 1494514200,
  91         },
  92         'params': {
  93             'skip_download': True,
  94         },
  95     }, {
  96         'url': 'http://video.aktualne.cz/v-cechach-poprve-zazni-zelenkova-zrestaurovana-mse/r~45b4b00483ec11e4883b002590604f2e/',
  97         'only_matching': True,
  98     }, {
  99         # Test live stream video (liveStarter) parsing
 100         'url': 'https://video.aktualne.cz/dvtv/zive-mistryne-sveta-eva-samkova-po-navratu-ze-sampionatu/r~182654c2288811e990fd0cc47ab5f122/',
 101         'md5': '2e552e483f2414851ca50467054f9d5d',
 102         'info_dict': {
 103             'id': '8d116360288011e98c840cc47ab5f122',
 104             'ext': 'mp4',
 105             'title': 'Živě: Mistryně světa Eva Samková po návratu ze šampionátu',
 106             'upload_date': '20190204',
 107             'timestamp': 1549289591,
 108         },
 109         'params': {
 110             # Video content is no longer available
 111             'skip_download': True,
 112         },
 113     }]
 114
 115     def _parse_video_metadata(self, js, video_id, timestamp):
 116         data = self._parse_json(js, video_id, transform_source=js_to_json)
 117         title = unescapeHTML(data['title'])
 118
 119         live_starter = try_get(data, lambda x: x['plugins']['liveStarter'], dict)
 120         if live_starter:
 121             data.update(live_starter)
 122
 123         formats = []
 124         for tracks in data.get('tracks', {}).values():
 125             for video in tracks:
 126                 video_url = video.get('src')
 127                 if not video_url:
 128                     continue
 129                 video_type = video.get('type')
 130                 ext = determine_ext(video_url, mimetype2ext(video_type))
 131                 if video_type == 'application/vnd.apple.mpegurl' or ext == 'm3u8':
 132                     formats.extend(self._extract_m3u8_formats(
 133                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 134                         m3u8_id='hls', fatal=False))
 135                 elif video_type == 'application/dash+xml' or ext == 'mpd':
 136                     formats.extend(self._extract_mpd_formats(
 137                         video_url, video_id, mpd_id='dash', fatal=False))
 138                 else:
 139                     label = video.get('label')
 140                     height = self._search_regex(
 141                         r'^(\d+)[pP]', label or '', 'height', default=None)
 142                     format_id = ['http']
 143                     for f in (ext, label):
 144                         if f:
 145                             format_id.append(f)
 146                     formats.append({
 147                         'url': video_url,
 148                         'format_id': '-'.join(format_id),
 149                         'height': int_or_none(height),
 150                     })
 151         self._sort_formats(formats)
 152
 153         return {
 154             'id': data.get('mediaid') or video_id,
 155             'title': title,
 156             'description': data.get('description'),
 157             'thumbnail': data.get('image'),
 158             'duration': int_or_none(data.get('duration')),
 159             'timestamp': int_or_none(timestamp),
 160             'formats': formats
 161         }
 162
 163     def _real_extract(self, url):
 164         video_id = self._match_id(url)
 165         webpage = self._download_webpage(url, video_id)
 166         timestamp = parse_iso8601(self._html_search_meta(
 167             'article:published_time', webpage, 'published time', default=None))
 168
 169         items = re.findall(r'(?s)playlist\.push\(({.+?})\);', webpage)
 170         if items:
 171             return self.playlist_result(
 172                 [self._parse_video_metadata(i, video_id, timestamp) for i in items],
 173                 video_id, self._html_search_meta('twitter:title', webpage))
 174
 175         item = self._search_regex(
 176             r'(?s)BBXPlayer\.setup\((.+?)\);',
 177             webpage, 'video', default=None)
 178         if item:
 179             # remove function calls (ex. htmldeentitize)
 180             # TODO this should be fixed in a general way in the js_to_json
 181             item = re.sub(r'\w+?\((.+)\)', r'\1', item)
 182             return self._parse_video_metadata(item, video_id, timestamp)
 183
 184         raise ExtractorError('Could not find neither video nor playlist')