X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/b238854ce845f3796daac74edab2e8a373e8ba1a..d1d8237037fb87a57bd1d073c3b9e642d4a75016:/youtube_dl/extractor/tvp.py?ds=inline diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py index bfed9dd..5070082 100644 --- a/youtube_dl/extractor/tvp.py +++ b/youtube_dl/extractor/tvp.py @@ -1,42 +1,179 @@ -import json +# coding: utf-8 +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 = u'tvp.pl' - _VALID_URL = r'https?://www\.tvp\.pl/.*?wideo/(?P\d+)/(?P\d+)' +class TVPIE(InfoExtractor): + IE_NAME = 'tvp' + IE_DESC = 'Telewizja Polska' + _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:(?!\d+/)[^/]+/)*(?P\d+)' - _TEST = { - u'url': u'http://www.tvp.pl/warszawa/magazyny/campusnews/wideo/31102013/12878238', - u'md5': u'148408967a6a468953c0a75cbdaf0d7a', - u'file': u'12878238.wmv', - u'info_dict': { - u'title': u'31.10.2013 - Odcinek 2', - u'description': u'31.10.2013 - Odcinek 2', + _TESTS = [{ + 'url': 'http://vod.tvp.pl/194536/i-seria-odc-13', + 'md5': '8aa518c15e5cc32dfe8db400dc921fbb', + 'info_dict': { + 'id': '194536', + 'ext': 'mp4', + 'title': 'Czas honoru, I seria – odc. 13', + }, + }, { + 'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176', + 'md5': 'b0005b542e5b4de643a9690326ab1257', + 'info_dict': { + 'id': '17916176', + 'ext': 'mp4', + 'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata', }, - u'skip': u'Download has to use same server IP as extraction. Therefore, a good (load-balancing) DNS resolver will make the download fail.' - } + }, { + 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272', + 'only_matching': True, + }, { + 'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200', + 'only_matching': True, + }, { + 'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa', + 'only_matching': True, + }, { + 'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach', + 'only_matching': True, + }, { + 'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum', + 'only_matching': True, + }, { + 'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) - json_url = 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id - json_params = self._download_webpage( - json_url, video_id, u"Downloading video metadata") - - params = json.loads(json_params) - self.report_extraction(video_id) - video_url = params['video_url'] - - title = self._og_search_title(webpage, fatal=True) + 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') + series_title = self._search_regex( + r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1', + webpage, 'series', group='series', default=None) + if series_title: + title = '%s, %s' % (series_title, title) + + thumbnail = self._search_regex( + 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 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'] + + 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: Current DASH formats are broken - $Time$ pattern in + # <SegmentTemplate> not implemented yet + # formats.extend(self._extract_mpd_formats( + # video_url_base + '.ism/video.mpd', + # video_id, mpd_id='dash', 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' and f.get('resolution') != 'multiple', + 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': determine_ext(video_url, 'mp4'), + }] + + self._sort_formats(formats) + return { 'id': video_id, 'title': title, - 'ext': 'wmv', - 'url': video_url, - 'description': self._og_search_description(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), + 'thumbnail': thumbnail, + 'formats': formats, + } + + +class TVPSeriesIE(InfoExtractor): + IE_NAME = 'tvp:series' + _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$' + + _TESTS = [{ + 'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem', + 'info_dict': { + 'title': 'Ogniem i mieczem', + 'id': '4278026', + }, + 'playlist_count': 4, + }, { + 'url': 'http://vod.tvp.pl/audycje/podroze/boso-przez-swiat', + 'info_dict': { + 'title': 'Boso przez świat', + 'id': '9329207', + }, + 'playlist_count': 86, + }] + + 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] + + return { + '_type': 'playlist', + 'id': playlist_id, + 'display_id': display_id, + 'title': title, + 'entries': entries, }