Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tvp.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     clean_html,
  10     determine_ext,
  11     ExtractorError,
  12     get_element_by_attribute,
  13     orderedSet,
  14 )
  15
  16
  17 class TVPIE(InfoExtractor):
  18     IE_NAME = 'tvp'
  19     IE_DESC = 'Telewizja Polska'
  20     _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:video/(?:[^,\s]*,)*|(?:(?!\d+/)[^/]+/)*)(?P<id>\d+)'
  21
  22     _TESTS = [{
  23         'url': 'https://vod.tvp.pl/video/czas-honoru,i-seria-odc-13,194536',
  24         'md5': 'a21eb0aa862f25414430f15fdfb9e76c',
  25         'info_dict': {
  26             'id': '194536',
  27             'ext': 'mp4',
  28             'title': 'Czas honoru, odc. 13 – Władek',
  29             'description': 'md5:437f48b93558370b031740546b696e24',
  30         },
  31     }, {
  32         'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
  33         'md5': 'b0005b542e5b4de643a9690326ab1257',
  34         'info_dict': {
  35             'id': '17916176',
  36             'ext': 'mp4',
  37             'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
  38             'description': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
  39         },
  40     }, {
  41         # page id is not the same as video id(#7799)
  42         'url': 'https://wiadomosci.tvp.pl/33908820/28092017-1930',
  43         'md5': '84cd3c8aec4840046e5ab712416b73d0',
  44         'info_dict': {
  45             'id': '33908820',
  46             'ext': 'mp4',
  47             'title': 'Wiadomości, 28.09.2017, 19:30',
  48             'description': 'Wydanie główne codziennego serwisu informacyjnego.'
  49         },
  50         'skip': 'HTTP Error 404: Not Found',
  51     }, {
  52         'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
  53         'only_matching': True,
  54     }, {
  55         'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200',
  56         'only_matching': True,
  57     }, {
  58         'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa',
  59         'only_matching': True,
  60     }, {
  61         'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach',
  62         'only_matching': True,
  63     }, {
  64         'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum',
  65         'only_matching': True,
  66     }, {
  67         'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji',
  68         'only_matching': True,
  69     }]
  70
  71     def _real_extract(self, url):
  72         page_id = self._match_id(url)
  73         webpage = self._download_webpage(url, page_id)
  74         video_id = self._search_regex([
  75             r'<iframe[^>]+src="[^"]*?object_id=(\d+)',
  76             r"object_id\s*:\s*'(\d+)'",
  77             r'data-video-id="(\d+)"'], webpage, 'video id', default=page_id)
  78         return {
  79             '_type': 'url_transparent',
  80             'url': 'tvp:' + video_id,
  81             'description': self._og_search_description(
  82                 webpage, default=None) or self._html_search_meta(
  83                 'description', webpage, default=None),
  84             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  85             'ie_key': 'TVPEmbed',
  86         }
  87
  88
  89 class TVPEmbedIE(InfoExtractor):
  90     IE_NAME = 'tvp:embed'
  91     IE_DESC = 'Telewizja Polska'
  92     _VALID_URL = r'(?:tvp:|https?://[^/]+\.tvp\.(?:pl|info)/sess/tvplayer\.php\?.*?object_id=)(?P<id>\d+)'
  93
  94     _TESTS = [{
  95         'url': 'tvp:194536',
  96         'md5': 'a21eb0aa862f25414430f15fdfb9e76c',
  97         'info_dict': {
  98             'id': '194536',
  99             'ext': 'mp4',
 100             'title': 'Czas honoru, odc. 13 – Władek',
 101         },
 102     }, {
 103         # not available
 104         'url': 'http://www.tvp.pl/sess/tvplayer.php?object_id=22670268',
 105         'md5': '8c9cd59d16edabf39331f93bf8a766c7',
 106         'info_dict': {
 107             'id': '22670268',
 108             'ext': 'mp4',
 109             'title': 'Panorama, 07.12.2015, 15:40',
 110         },
 111         'skip': 'Transmisja została zakończona lub materiał niedostępny',
 112     }, {
 113         'url': 'tvp:22670268',
 114         'only_matching': True,
 115     }]
 116
 117     def _real_extract(self, url):
 118         video_id = self._match_id(url)
 119
 120         webpage = self._download_webpage(
 121             'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
 122
 123         error = self._html_search_regex(
 124             r'(?s)<p[^>]+\bclass=["\']notAvailable__text["\'][^>]*>(.+?)</p>',
 125             webpage, 'error', default=None) or clean_html(
 126             get_element_by_attribute('class', 'msg error', webpage))
 127         if error:
 128             raise ExtractorError('%s said: %s' % (
 129                 self.IE_NAME, clean_html(error)), expected=True)
 130
 131         title = self._search_regex(
 132             r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1',
 133             webpage, 'title', group='title')
 134         series_title = self._search_regex(
 135             r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1',
 136             webpage, 'series', group='series', default=None)
 137         if series_title:
 138             title = '%s, %s' % (series_title, title)
 139
 140         thumbnail = self._search_regex(
 141             r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None)
 142
 143         video_url = self._search_regex(
 144             r'0:{src:([\'"])(?P<url>.*?)\1', webpage,
 145             'formats', group='url', default=None)
 146         if not video_url or 'material_niedostepny.mp4' in video_url:
 147             video_url = self._download_json(
 148                 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id,
 149                 video_id)['video_url']
 150
 151         formats = []
 152         video_url_base = self._search_regex(
 153             r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)',
 154             video_url, 'video base url', default=None)
 155         if video_url_base:
 156             # TODO: <Group> found instead of <AdaptationSet> in MPD manifest.
 157             # It's not mentioned in MPEG-DASH standard. Figure that out.
 158             # formats.extend(self._extract_mpd_formats(
 159             #     video_url_base + '.ism/video.mpd',
 160             #     video_id, mpd_id='dash', fatal=False))
 161             formats.extend(self._extract_ism_formats(
 162                 video_url_base + '.ism/Manifest',
 163                 video_id, 'mss', fatal=False))
 164             formats.extend(self._extract_f4m_formats(
 165                 video_url_base + '.ism/video.f4m',
 166                 video_id, f4m_id='hds', fatal=False))
 167             m3u8_formats = self._extract_m3u8_formats(
 168                 video_url_base + '.ism/video.m3u8', video_id,
 169                 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
 170             self._sort_formats(m3u8_formats)
 171             m3u8_formats = list(filter(
 172                 lambda f: f.get('vcodec') != 'none', m3u8_formats))
 173             formats.extend(m3u8_formats)
 174             for i, m3u8_format in enumerate(m3u8_formats, 2):
 175                 http_url = '%s-%d.mp4' % (video_url_base, i)
 176                 if self._is_valid_url(http_url, video_id):
 177                     f = m3u8_format.copy()
 178                     f.update({
 179                         'url': http_url,
 180                         'format_id': f['format_id'].replace('hls', 'http'),
 181                         'protocol': 'http',
 182                     })
 183                     formats.append(f)
 184         else:
 185             formats = [{
 186                 'format_id': 'direct',
 187                 'url': video_url,
 188                 'ext': determine_ext(video_url, 'mp4'),
 189             }]
 190
 191         self._sort_formats(formats)
 192
 193         return {
 194             'id': video_id,
 195             'title': title,
 196             'thumbnail': thumbnail,
 197             'formats': formats,
 198         }
 199
 200
 201 class TVPWebsiteIE(InfoExtractor):
 202     IE_NAME = 'tvp:series'
 203     _VALID_URL = r'https?://vod\.tvp\.pl/website/(?P<display_id>[^,]+),(?P<id>\d+)'
 204
 205     _TESTS = [{
 206         # series
 207         'url': 'https://vod.tvp.pl/website/lzy-cennet,38678312/video',
 208         'info_dict': {
 209             'id': '38678312',
 210         },
 211         'playlist_count': 115,
 212     }, {
 213         # film
 214         'url': 'https://vod.tvp.pl/website/gloria,35139666',
 215         'info_dict': {
 216             'id': '36637049',
 217             'ext': 'mp4',
 218             'title': 'Gloria, Gloria',
 219         },
 220         'params': {
 221             'skip_download': True,
 222         },
 223         'add_ie': ['TVPEmbed'],
 224     }, {
 225         'url': 'https://vod.tvp.pl/website/lzy-cennet,38678312',
 226         'only_matching': True,
 227     }]
 228
 229     def _entries(self, display_id, playlist_id):
 230         url = 'https://vod.tvp.pl/website/%s,%s/video' % (display_id, playlist_id)
 231         for page_num in itertools.count(1):
 232             page = self._download_webpage(
 233                 url, display_id, 'Downloading page %d' % page_num,
 234                 query={'page': page_num})
 235
 236             video_ids = orderedSet(re.findall(
 237                 r'<a[^>]+\bhref=["\']/video/%s,[^,]+,(\d+)' % display_id,
 238                 page))
 239
 240             if not video_ids:
 241                 break
 242
 243             for video_id in video_ids:
 244                 yield self.url_result(
 245                     'tvp:%s' % video_id, ie=TVPEmbedIE.ie_key(),
 246                     video_id=video_id)
 247
 248     def _real_extract(self, url):
 249         mobj = re.match(self._VALID_URL, url)
 250         display_id, playlist_id = mobj.group('display_id', 'id')
 251         return self.playlist_result(
 252             self._entries(display_id, playlist_id), playlist_id)