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)/(?:(?!\d+/)[^/]+/)*(?P<id>\d+)'
  19
  20     _TESTS = [{
  21         'url': 'http://vod.tvp.pl/194536/i-seria-odc-13',
  22         'md5': '8aa518c15e5cc32dfe8db400dc921fbb',
  23         'info_dict': {
  24             'id': '194536',
  25             'ext': 'mp4',
  26             'title': 'Czas honoru, I seria – odc. 13',
  27         },
  28     }, {
  29         'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
  30         'md5': 'b0005b542e5b4de643a9690326ab1257',
  31         'info_dict': {
  32             'id': '17916176',
  33             'ext': 'mp4',
  34             'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
  35         },
  36     }, {
  37         'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
  38         'only_matching': True,
  39     }, {
  40         'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200',
  41         'only_matching': True,
  42     }, {
  43         'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa',
  44         'only_matching': True,
  45     }, {
  46         'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach',
  47         'only_matching': True,
  48     }, {
  49         'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum',
  50         'only_matching': True,
  51     }, {
  52         'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji',
  53         'only_matching': True,
  54     }]
  55
  56     def _real_extract(self, url):
  57         video_id = self._match_id(url)
  58
  59         webpage = self._download_webpage(
  60             'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
  61
  62         error_massage = get_element_by_attribute('class', 'msg error', webpage)
  63         if error_massage:
  64             raise ExtractorError('%s said: %s' % (
  65                 self.IE_NAME, clean_html(error_massage)), expected=True)
  66
  67         title = self._search_regex(
  68             r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1',
  69             webpage, 'title', group='title')
  70         series_title = self._search_regex(
  71             r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1',
  72             webpage, 'series', group='series', default=None)
  73         if series_title:
  74             title = '%s, %s' % (series_title, title)
  75
  76         thumbnail = self._search_regex(
  77             r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None)
  78
  79         video_url = self._search_regex(
  80             r'0:{src:([\'"])(?P<url>.*?)\1', webpage,
  81             'formats', group='url', default=None)
  82         if not video_url or 'material_niedostepny.mp4' in video_url:
  83             video_url = self._download_json(
  84                 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id,
  85                 video_id)['video_url']
  86
  87         formats = []
  88         video_url_base = self._search_regex(
  89             r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)',
  90             video_url, 'video base url', default=None)
  91         if video_url_base:
  92             # TODO: Current DASH formats are broken - $Time$ pattern in
  93             # <SegmentTemplate> not implemented yet
  94             # formats.extend(self._extract_mpd_formats(
  95             #     video_url_base + '.ism/video.mpd',
  96             #     video_id, mpd_id='dash', fatal=False))
  97             formats.extend(self._extract_f4m_formats(
  98                 video_url_base + '.ism/video.f4m',
  99                 video_id, f4m_id='hds', fatal=False))
 100             m3u8_formats = self._extract_m3u8_formats(
 101                 video_url_base + '.ism/video.m3u8', video_id,
 102                 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
 103             self._sort_formats(m3u8_formats)
 104             m3u8_formats = list(filter(
 105                 lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
 106                 m3u8_formats))
 107             formats.extend(m3u8_formats)
 108             for i, m3u8_format in enumerate(m3u8_formats, 2):
 109                 http_url = '%s-%d.mp4' % (video_url_base, i)
 110                 if self._is_valid_url(http_url, video_id):
 111                     f = m3u8_format.copy()
 112                     f.update({
 113                         'url': http_url,
 114                         'format_id': f['format_id'].replace('hls', 'http'),
 115                         'protocol': 'http',
 116                     })
 117                     formats.append(f)
 118         else:
 119             formats = [{
 120                 'format_id': 'direct',
 121                 'url': video_url,
 122                 'ext': determine_ext(video_url, 'mp4'),
 123             }]
 124
 125         self._sort_formats(formats)
 126
 127         return {
 128             'id': video_id,
 129             'title': title,
 130             'thumbnail': thumbnail,
 131             'formats': formats,
 132         }
 133
 134
 135 class TVPSeriesIE(InfoExtractor):
 136     IE_NAME = 'tvp:series'
 137     _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$'
 138
 139     _TESTS = [{
 140         'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem',
 141         'info_dict': {
 142             'title': 'Ogniem i mieczem',
 143             'id': '4278026',
 144         },
 145         'playlist_count': 4,
 146     }, {
 147         'url': 'http://vod.tvp.pl/audycje/podroze/boso-przez-swiat',
 148         'info_dict': {
 149             'title': 'Boso przez świat',
 150             'id': '9329207',
 151         },
 152         'playlist_count': 86,
 153     }]
 154
 155     def _real_extract(self, url):
 156         display_id = self._match_id(url)
 157         webpage = self._download_webpage(url, display_id, tries=5)
 158
 159         title = self._html_search_regex(
 160             r'(?s) id=[\'"]path[\'"]>(?:.*? / ){2}(.*?)</span>', webpage, 'series')
 161         playlist_id = self._search_regex(r'nodeId:\s*(\d+)', webpage, 'playlist id')
 162         playlist = self._download_webpage(
 163             'http://vod.tvp.pl/vod/seriesAjax?type=series&nodeId=%s&recommend'
 164             'edId=0&sort=&page=0&pageSize=10000' % playlist_id, display_id, tries=5,
 165             note='Downloading playlist')
 166
 167         videos_paths = re.findall(
 168             '(?s)class="shortTitle">.*?href="(/[^"]+)', playlist)
 169         entries = [
 170             self.url_result('http://vod.tvp.pl%s' % v_path, ie=TVPIE.ie_key())
 171             for v_path in videos_paths]
 172
 173         return {
 174             '_type': 'playlist',
 175             'id': playlist_id,
 176             'display_id': display_id,
 177             'title': title,
 178             'entries': entries,
 179         }