X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Ftvp.py;h=3954f0b93812e5bd58d950e8401c954d0ea0a7c6;hb=4c76aa06665621c7689938afd7bbdbc797b5c7ea;hp=a4997cb8965dd9c31be4f5fc4a679fb0aa147b82;hpb=e21c26daf9dfb56b9bc4f18f30e7f1c724c9189e;p=youtube-dl diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py index a4997cb89..3954f0b93 100644 --- a/youtube_dl/extractor/tvp.py +++ b/youtube_dl/extractor/tvp.py @@ -4,28 +4,46 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import ( + determine_ext, + clean_html, + get_element_by_attribute, + ExtractorError, +) class TVPIE(InfoExtractor): IE_NAME = 'tvp' IE_DESC = 'Telewizja Polska' - _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:(?!\d+/)[^/]+/)*(?P\d+)' + _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:video/(?:[^,\s]*,)*|(?:(?!\d+/)[^/]+/)*)(?P\d+)' _TESTS = [{ - 'url': 'http://vod.tvp.pl/194536/i-seria-odc-13', + 'url': 'https://vod.tvp.pl/video/czas-honoru,i-seria-odc-13,194536', 'md5': '8aa518c15e5cc32dfe8db400dc921fbb', 'info_dict': { 'id': '194536', 'ext': 'mp4', 'title': 'Czas honoru, I seria – odc. 13', + 'description': 'md5:381afa5bca72655fe94b05cfe82bf53d', }, }, { 'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176', - 'md5': 'c3b15ed1af288131115ff17a17c19dda', + 'md5': 'b0005b542e5b4de643a9690326ab1257', 'info_dict': { 'id': '17916176', 'ext': 'mp4', 'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata', + 'description': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata', + }, + }, { + # page id is not the same as video id(#7799) + 'url': 'https://wiadomosci.tvp.pl/33908820/28092017-1930', + 'md5': '84cd3c8aec4840046e5ab712416b73d0', + 'info_dict': { + 'id': '33908820', + 'ext': 'mp4', + 'title': 'Wiadomości, 28.09.2017, 19:30', + 'description': 'Wydanie główne codziennego serwisu informacyjnego.' }, }, { 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272', @@ -47,12 +65,51 @@ class TVPIE(InfoExtractor): 'only_matching': True, }] + def _real_extract(self, url): + page_id = self._match_id(url) + webpage = self._download_webpage(url, page_id) + video_id = self._search_regex([ + r']+src="[^"]*?object_id=(\d+)', + r"object_id\s*:\s*'(\d+)'", + r'data-video-id="(\d+)"'], webpage, 'video id', default=page_id) + return { + '_type': 'url_transparent', + 'url': 'tvp:' + video_id, + 'description': self._og_search_description(webpage, default=None), + 'thumbnail': self._og_search_thumbnail(webpage), + 'ie_key': 'TVPEmbed', + } + + +class TVPEmbedIE(InfoExtractor): + IE_NAME = 'tvp:embed' + IE_DESC = 'Telewizja Polska' + _VALID_URL = r'(?:tvp:|https?://[^/]+\.tvp\.(?:pl|info)/sess/tvplayer\.php\?.*?object_id=)(?P\d+)' + + _TESTS = [{ + 'url': 'http://www.tvp.pl/sess/tvplayer.php?object_id=22670268', + 'md5': '8c9cd59d16edabf39331f93bf8a766c7', + 'info_dict': { + 'id': '22670268', + 'ext': 'mp4', + 'title': 'Panorama, 07.12.2015, 15:40', + }, + }, { + 'url': 'tvp:22670268', + 'only_matching': True, + }] + def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage( 'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id) + error_massage = get_element_by_attribute('class', 'msg error', webpage) + if error_massage: + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, clean_html(error_massage)), expected=True) + title = self._search_regex( r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P.+?)\1', webpage, 'title', group='title') @@ -66,24 +123,52 @@ class TVPIE(InfoExtractor): r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None) video_url = self._search_regex( - r'0:{src:([\'"])(?P<url>.*?)\1', webpage, 'formats', group='url', default=None) - if not video_url: + r'0:{src:([\'"])(?P<url>.*?)\1', webpage, + 'formats', group='url', default=None) + if not video_url or 'material_niedostepny.mp4' in video_url: video_url = self._download_json( 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id, video_id)['video_url'] - ext = video_url.rsplit('.', 1)[-1] - if ext != 'ism/manifest': - if '/' in ext: - ext = 'mp4' + formats = [] + video_url_base = self._search_regex( + r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)', + video_url, 'video base url', default=None) + if video_url_base: + # TODO: <Group> found instead of <AdaptationSet> in MPD manifest. + # It's not mentioned in MPEG-DASH standard. Figure that out. + # formats.extend(self._extract_mpd_formats( + # video_url_base + '.ism/video.mpd', + # video_id, mpd_id='dash', fatal=False)) + formats.extend(self._extract_ism_formats( + video_url_base + '.ism/Manifest', + video_id, 'mss', fatal=False)) + formats.extend(self._extract_f4m_formats( + video_url_base + '.ism/video.f4m', + video_id, f4m_id='hds', fatal=False)) + m3u8_formats = self._extract_m3u8_formats( + video_url_base + '.ism/video.m3u8', video_id, + 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + self._sort_formats(m3u8_formats) + m3u8_formats = list(filter( + lambda f: f.get('vcodec') != 'none', m3u8_formats)) + formats.extend(m3u8_formats) + for i, m3u8_format in enumerate(m3u8_formats, 2): + http_url = '%s-%d.mp4' % (video_url_base, i) + if self._is_valid_url(http_url, video_id): + f = m3u8_format.copy() + f.update({ + 'url': http_url, + 'format_id': f['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + formats.append(f) + else: formats = [{ 'format_id': 'direct', 'url': video_url, - 'ext': ext, + 'ext': determine_ext(video_url, 'mp4'), }] - else: - m3u8_url = re.sub('([^/]*)\.ism/manifest', r'\1.ism/\1.m3u8', video_url) - formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4') self._sort_formats(formats)