-# -*- coding: utf-8 -*-
+# 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 = 'tvp.pl'
- _VALID_URL = r'https?://(?P<type>vod|www)\.tvp\.pl/.*/(?P<id>\d+)$'
-
- _TESTS = [
- {
- 'url': 'http://www.tvp.pl/warszawa/magazyny/campusnews/wideo/31102013/12878238',
- 'info_dict': {
- 'id': '12878238',
- 'ext': 'wmv',
- 'title': 'CAMPUSnews, 31.10.2013 - Odcinek 2',
- 'description': '',
- },
- 'skip': '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/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem/wideo/odc-2/4278035',
- 'info_dict': {
- 'id': '4278035',
- 'ext': 'wmv',
- 'title': 'Ogniem i mieczem, odc. 2',
- 'description': 'Bohun dowiaduje się o złamaniu przez kniahinię danego mu słowa i wyrusza do Rozłogów. Helenie w ostatniej chwili udaje się uciec dzięki pomocy Zagłoby.',
- },
- 'skip': 'As above',
- }, {
- 'url': 'http://vod.tvp.pl/seriale/obyczajowe/czas-honoru/sezon-1-1-13/i-seria-odc-13/194536',
- 'info_dict': {
- 'id': '194536',
- 'ext': 'mp4',
- 'title': 'Czas honoru, I seria – odc. 13',
- 'description': 'WŁADEK\nCzesław prosi Marię o dostarczenie Władkowi zarazki tyfusu. Jeśli zachoruje zostanie przewieziony do szpitala skąd łatwiej będzie go odbić. Czy matka zdecyduje się zarazić syna? Karol odwiedza Wandę przyznaje się, że ją oszukiwał, ale ostrzega też, że grozi jej aresztowanie i nalega, żeby wyjechała z Warszawy. Czy dziewczyna zdecyduje się znów oddalić od ukochanego? Rozpoczyna się akcja odbicia Władka.',
- },
- }, {
- 'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
- 'info_dict': {
- 'id': '17916176',
- 'ext': 'mp4',
- 'title': 'rozmaitosci, TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
- 'description': '',
- },
- 'params': {
- # m3u8 download
- 'skip_download': 'true',
- },
- }, {
- 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
- 'info_dict': {
- 'id': '17834272',
- 'ext': 'mp4',
- 'title': 'Na sygnale, odc. 39',
- 'description': 'Ekipa Wiktora ratuje młodą matkę, która spadła ze schodów trzymając na rękach noworodka. Okazuje się, że dziewczyna jest surogatką, a biologiczni rodzice dziecka próbują zmusić ją do oddania synka…',
- },
- 'params': {
- # m3u8 download
- 'skip_download': 'true',
- },
+class TVPIE(InfoExtractor):
+ IE_NAME = 'tvp'
+ IE_DESC = 'Telewizja Polska'
+ _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:video/(?:[^,\s]*,)*|(?:(?!\d+/)[^/]+/)*)(?P<id>\d+)'
+
+ _TESTS = [{
+ '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': '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',
+ '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):
+ page_id = self._match_id(url)
+ webpage = self._download_webpage(url, page_id)
+ video_id = self._search_regex([
+ r'<iframe[^>]+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<id>\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):
- mobj = re.match(self._VALID_URL, url)
- video_id = mobj.group('id')
+ video_id = self._match_id(url)
+
webpage = self._download_webpage(
'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
- title = self._og_search_title(webpage)
- series = self._search_regex(
- r'{name:\s*([\'"])SeriesTitle\1,\s*value:\s*\1(?P<series>.*?)\1},',
+
+ 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<title>.+?)\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 is not None and series not in title:
- title = '%s, %s' % (series, title)
- info_dict = {
- 'id': video_id,
- 'title': title,
- 'thumbnail': self._og_search_thumbnail(webpage),
- 'description': self._og_search_description(webpage, default=''),
- }
- if mobj.group('type') == 'vod' and info_dict['description'] == '':
- info_dict.update({
- 'description': self._html_search_regex(
- r'(?s)<div\s+class=[\'"]opis.*?</div>',
- self._download_webpage(url, video_id), 'description', group=0),
- })
+ 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 video_url is None:
+ 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'
- info_dict.update({
- 'ext': ext,
- 'url': 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: <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:
- m3u8_url = re.sub('([^/]*)\.ism/manifest', r'\1.ism/\1.m3u8', video_url)
- formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
- info_dict.update({
- 'formats': formats,
- })
- return info_dict
+ formats = [{
+ 'format_id': 'direct',
+ 'url': video_url,
+ 'ext': determine_ext(video_url, 'mp4'),
+ }]
+
+ self._sort_formats(formats)
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'thumbnail': thumbnail,
+ 'formats': formats,
+ }
-class TvpSeriesIE(InfoExtractor):
- IE_NAME = 'tvp.pl:Series'
+
+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 _force_download_webpage(self, url, v_id, tries=0):
- if tries >= 5:
- raise ExtractorError(
- '%s: Cannot download webpage, try again later' % v_id)
- # Sometimes happen, but in my tests second try always succeeded
- try:
- return self._download_webpage(url, v_id)
- except IncompleteRead as e:
- return self._force_download_webpage(url, v_id, tries+1)
-
+ _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._force_download_webpage(url, display_id)
+ webpage = self._download_webpage(url, display_id, tries=5)
+
title = self._html_search_regex(
- r'(?s) id=[\'"]path[\'"]>(.*?)</span>', webpage, 'series')
- title = title.split(' / ', 2)[-1]
+ r'(?s) id=[\'"]path[\'"]>(?:.*? / ){2}(.*?)</span>', webpage, 'series')
playlist_id = self._search_regex(r'nodeId:\s*(\d+)', webpage, 'playlist id')
- playlist = self._force_download_webpage(
+ playlist = self._download_webpage(
'http://vod.tvp.pl/vod/seriesAjax?type=series&nodeId=%s&recommend'
- 'edId=0&sort=&page=0&pageSize=1000000' % playlist_id, display_id)
+ '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())
+ 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,
projects / youtube-dl / blobdiff