X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/fe979149c83b5a935f7d28baf75848a9137316fd..7734719b79e09597eb492309ba53519877399b72:/youtube_dl/extractor/tvp.py diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py index 5070082..accff75 100644 --- a/youtube_dl/extractor/tvp.py +++ b/youtube_dl/extractor/tvp.py @@ -1,29 +1,32 @@ # coding: utf-8 from __future__ import unicode_literals +import itertools import re from .common import InfoExtractor from ..utils import ( - determine_ext, clean_html, - get_element_by_attribute, + determine_ext, ExtractorError, + get_element_by_attribute, + orderedSet, ) 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', - 'md5': '8aa518c15e5cc32dfe8db400dc921fbb', + 'url': 'https://vod.tvp.pl/video/czas-honoru,i-seria-odc-13,194536', + 'md5': 'a21eb0aa862f25414430f15fdfb9e76c', 'info_dict': { 'id': '194536', 'ext': 'mp4', - 'title': 'Czas honoru, I seria – odc. 13', + 'title': 'Czas honoru, odc. 13 – Władek', + 'description': 'md5:437f48b93558370b031740546b696e24', }, }, { 'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176', @@ -32,7 +35,19 @@ class TVPIE(InfoExtractor): '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.' }, + 'skip': 'HTTP Error 404: Not Found', }, { 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272', 'only_matching': True, @@ -53,16 +68,65 @@ 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) or self._html_search_meta( + 'description', webpage, default=None), + 'thumbnail': self._og_search_thumbnail(webpage, default=None), + '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': 'tvp:194536', + 'md5': 'a21eb0aa862f25414430f15fdfb9e76c', + 'info_dict': { + 'id': '194536', + 'ext': 'mp4', + 'title': 'Czas honoru, odc. 13 – Władek', + }, + }, { + # not available + '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', + }, + 'skip': 'Transmisja została zakończona lub materiał niedostępny', + }, { + '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: + error = self._html_search_regex( + r'(?s)]+\bclass=["\']notAvailable__text["\'][^>]*>(.+?)

', + webpage, 'error', default=None) or clean_html( + get_element_by_attribute('class', 'msg error', webpage)) + if error: raise ExtractorError('%s said: %s' % ( - self.IE_NAME, clean_html(error_massage)), expected=True) + self.IE_NAME, clean_html(error)), expected=True) title = self._search_regex( r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P.+?)\1', @@ -89,11 +153,14 @@ class TVPIE(InfoExtractor): r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)', video_url, 'video base url', default=None) if video_url_base: - # TODO: Current DASH formats are broken - $Time$ pattern in - # <SegmentTemplate> not implemented yet + # 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)) @@ -102,8 +169,7 @@ class TVPIE(InfoExtractor): 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) self._sort_formats(m3u8_formats) m3u8_formats = list(filter( - lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple', - m3u8_formats)) + 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) @@ -132,48 +198,55 @@ class TVPIE(InfoExtractor): } -class TVPSeriesIE(InfoExtractor): +class TVPWebsiteIE(InfoExtractor): IE_NAME = 'tvp:series' - _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$' + _VALID_URL = r'https?://vod\.tvp\.pl/website/(?P<display_id>[^,]+),(?P<id>\d+)' _TESTS = [{ - 'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem', + # series + 'url': 'https://vod.tvp.pl/website/lzy-cennet,38678312/video', 'info_dict': { - 'title': 'Ogniem i mieczem', - 'id': '4278026', + 'id': '38678312', }, - 'playlist_count': 4, + 'playlist_count': 115, }, { - 'url': 'http://vod.tvp.pl/audycje/podroze/boso-przez-swiat', + # film + 'url': 'https://vod.tvp.pl/website/gloria,35139666', 'info_dict': { - 'title': 'Boso przez świat', - 'id': '9329207', + 'id': '36637049', + 'ext': 'mp4', + 'title': 'Gloria, Gloria', + }, + 'params': { + 'skip_download': True, }, - 'playlist_count': 86, + 'add_ie': ['TVPEmbed'], + }, { + 'url': 'https://vod.tvp.pl/website/lzy-cennet,38678312', + 'only_matching': True, }] - def _real_extract(self, url): - display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id, tries=5) - - title = self._html_search_regex( - r'(?s) id=[\'"]path[\'"]>(?:.*? / ){2}(.*?)</span>', webpage, 'series') - playlist_id = self._search_regex(r'nodeId:\s*(\d+)', webpage, 'playlist id') - playlist = self._download_webpage( - 'http://vod.tvp.pl/vod/seriesAjax?type=series&nodeId=%s&recommend' - 'edId=0&sort=&page=0&pageSize=10000' % playlist_id, display_id, tries=5, - note='Downloading playlist') - - videos_paths = re.findall( - '(?s)class="shortTitle">.*?href="(/[^"]+)', playlist) - entries = [ - self.url_result('http://vod.tvp.pl%s' % v_path, ie=TVPIE.ie_key()) - for v_path in videos_paths] + def _entries(self, display_id, playlist_id): + url = 'https://vod.tvp.pl/website/%s,%s/video' % (display_id, playlist_id) + for page_num in itertools.count(1): + page = self._download_webpage( + url, display_id, 'Downloading page %d' % page_num, + query={'page': page_num}) - return { - '_type': 'playlist', - 'id': playlist_id, - 'display_id': display_id, - 'title': title, - 'entries': entries, - } + video_ids = orderedSet(re.findall( + r'<a[^>]+\bhref=["\']/video/%s,[^,]+,(\d+)' % display_id, + page)) + + if not video_ids: + break + + for video_id in video_ids: + yield self.url_result( + 'tvp:%s' % video_id, ie=TVPEmbedIE.ie_key(), + video_id=video_id) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + display_id, playlist_id = mobj.group('display_id', 'id') + return self.playlist_result( + self._entries(display_id, playlist_id), playlist_id)