Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tvp.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     clean_html,
  10     get_element_by_attribute,
  11     ExtractorError,
  12 )
  13
  14
  15 class TVPIE(InfoExtractor):
  16     IE_NAME = 'tvp'
  17     IE_DESC = 'Telewizja Polska'
  18     _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:video/(?:[^,\s]*,)*|(?:(?!\d+/)[^/]+/)*)(?P<id>\d+)'
  19
  20     _TESTS = [{
  21         'url': 'https://vod.tvp.pl/video/czas-honoru,i-seria-odc-13,194536',
  22         'md5': '8aa518c15e5cc32dfe8db400dc921fbb',
  23         'info_dict': {
  24             'id': '194536',
  25             'ext': 'mp4',
  26             'title': 'Czas honoru, I seria – odc. 13',
  27             'description': 'md5:381afa5bca72655fe94b05cfe82bf53d',
  28         },
  29     }, {
  30         'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
  31         'md5': 'b0005b542e5b4de643a9690326ab1257',
  32         'info_dict': {
  33             'id': '17916176',
  34             'ext': 'mp4',
  35             'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
  36             'description': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
  37         },
  38     }, {
  39         # page id is not the same as video id(#7799)
  40         'url': 'https://wiadomosci.tvp.pl/33908820/28092017-1930',
  41         'md5': '84cd3c8aec4840046e5ab712416b73d0',
  42         'info_dict': {
  43             'id': '33908820',
  44             'ext': 'mp4',
  45             'title': 'Wiadomości, 28.09.2017, 19:30',
  46             'description': 'Wydanie główne codziennego serwisu informacyjnego.'
  47         },
  48     }, {
  49         'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
  50         'only_matching': True,
  51     }, {
  52         'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200',
  53         'only_matching': True,
  54     }, {
  55         'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa',
  56         'only_matching': True,
  57     }, {
  58         'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach',
  59         'only_matching': True,
  60     }, {
  61         'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum',
  62         'only_matching': True,
  63     }, {
  64         'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji',
  65         'only_matching': True,
  66     }]
  67
  68     def _real_extract(self, url):
  69         page_id = self._match_id(url)
  70         webpage = self._download_webpage(url, page_id)
  71         video_id = self._search_regex([
  72             r'<iframe[^>]+src="[^"]*?object_id=(\d+)',
  73             r"object_id\s*:\s*'(\d+)'",
  74             r'data-video-id="(\d+)"'], webpage, 'video id', default=page_id)
  75         return {
  76             '_type': 'url_transparent',
  77             'url': 'tvp:' + video_id,
  78             'description': self._og_search_description(webpage, default=None),
  79             'thumbnail': self._og_search_thumbnail(webpage),
  80             'ie_key': 'TVPEmbed',
  81         }
  82
  83
  84 class TVPEmbedIE(InfoExtractor):
  85     IE_NAME = 'tvp:embed'
  86     IE_DESC = 'Telewizja Polska'
  87     _VALID_URL = r'(?:tvp:|https?://[^/]+\.tvp\.(?:pl|info)/sess/tvplayer\.php\?.*?object_id=)(?P<id>\d+)'
  88
  89     _TESTS = [{
  90         'url': 'http://www.tvp.pl/sess/tvplayer.php?object_id=22670268',
  91         'md5': '8c9cd59d16edabf39331f93bf8a766c7',
  92         'info_dict': {
  93             'id': '22670268',
  94             'ext': 'mp4',
  95             'title': 'Panorama, 07.12.2015, 15:40',
  96         },
  97     }, {
  98         'url': 'tvp:22670268',
  99         'only_matching': True,
 100     }]
 101
 102     def _real_extract(self, url):
 103         video_id = self._match_id(url)
 104
 105         webpage = self._download_webpage(
 106             'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
 107
 108         error_massage = get_element_by_attribute('class', 'msg error', webpage)
 109         if error_massage:
 110             raise ExtractorError('%s said: %s' % (
 111                 self.IE_NAME, clean_html(error_massage)), expected=True)
 112
 113         title = self._search_regex(
 114             r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1',
 115             webpage, 'title', group='title')
 116         series_title = self._search_regex(
 117             r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1',
 118             webpage, 'series', group='series', default=None)
 119         if series_title:
 120             title = '%s, %s' % (series_title, title)
 121
 122         thumbnail = self._search_regex(
 123             r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None)
 124
 125         video_url = self._search_regex(
 126             r'0:{src:([\'"])(?P<url>.*?)\1', webpage,
 127             'formats', group='url', default=None)
 128         if not video_url or 'material_niedostepny.mp4' in video_url:
 129             video_url = self._download_json(
 130                 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id,
 131                 video_id)['video_url']
 132
 133         formats = []
 134         video_url_base = self._search_regex(
 135             r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)',
 136             video_url, 'video base url', default=None)
 137         if video_url_base:
 138             # TODO: <Group> found instead of <AdaptationSet> in MPD manifest.
 139             # It's not mentioned in MPEG-DASH standard. Figure that out.
 140             # formats.extend(self._extract_mpd_formats(
 141             #     video_url_base + '.ism/video.mpd',
 142             #     video_id, mpd_id='dash', fatal=False))
 143             formats.extend(self._extract_ism_formats(
 144                 video_url_base + '.ism/Manifest',
 145                 video_id, 'mss', fatal=False))
 146             formats.extend(self._extract_f4m_formats(
 147                 video_url_base + '.ism/video.f4m',
 148                 video_id, f4m_id='hds', fatal=False))
 149             m3u8_formats = self._extract_m3u8_formats(
 150                 video_url_base + '.ism/video.m3u8', video_id,
 151                 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
 152             self._sort_formats(m3u8_formats)
 153             m3u8_formats = list(filter(
 154                 lambda f: f.get('vcodec') != 'none', m3u8_formats))
 155             formats.extend(m3u8_formats)
 156             for i, m3u8_format in enumerate(m3u8_formats, 2):
 157                 http_url = '%s-%d.mp4' % (video_url_base, i)
 158                 if self._is_valid_url(http_url, video_id):
 159                     f = m3u8_format.copy()
 160                     f.update({
 161                         'url': http_url,
 162                         'format_id': f['format_id'].replace('hls', 'http'),
 163                         'protocol': 'http',
 164                     })
 165                     formats.append(f)
 166         else:
 167             formats = [{
 168                 'format_id': 'direct',
 169                 'url': video_url,
 170                 'ext': determine_ext(video_url, 'mp4'),
 171             }]
 172
 173         self._sort_formats(formats)
 174
 175         return {
 176             'id': video_id,
 177             'title': title,
 178             'thumbnail': thumbnail,
 179             'formats': formats,
 180         }
 181
 182
 183 class TVPSeriesIE(InfoExtractor):
 184     IE_NAME = 'tvp:series'
 185     _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$'
 186
 187     _TESTS = [{
 188         'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem',
 189         'info_dict': {
 190             'title': 'Ogniem i mieczem',
 191             'id': '4278026',
 192         },
 193         'playlist_count': 4,
 194     }, {
 195         'url': 'http://vod.tvp.pl/audycje/podroze/boso-przez-swiat',
 196         'info_dict': {
 197             'title': 'Boso przez świat',
 198             'id': '9329207',
 199         },
 200         'playlist_count': 86,
 201     }]
 202
 203     def _real_extract(self, url):
 204         display_id = self._match_id(url)
 205         webpage = self._download_webpage(url, display_id, tries=5)
 206
 207         title = self._html_search_regex(
 208             r'(?s) id=[\'"]path[\'"]>(?:.*? / ){2}(.*?)</span>', webpage, 'series')
 209         playlist_id = self._search_regex(r'nodeId:\s*(\d+)', webpage, 'playlist id')
 210         playlist = self._download_webpage(
 211             'http://vod.tvp.pl/vod/seriesAjax?type=series&nodeId=%s&recommend'
 212             'edId=0&sort=&page=0&pageSize=10000' % playlist_id, display_id, tries=5,
 213             note='Downloading playlist')
 214
 215         videos_paths = re.findall(
 216             '(?s)class="shortTitle">.*?href="(/[^"]+)', playlist)
 217         entries = [
 218             self.url_result('http://vod.tvp.pl%s' % v_path, ie=TVPIE.ie_key())
 219             for v_path in videos_paths]
 220
 221         return {
 222             '_type': 'playlist',
 223             'id': playlist_id,
 224             'display_id': display_id,
 225             'title': title,
 226             'entries': entries,
 227         }