debian/README.source: Change 'markup'.

[youtubedl] / youtube_dl / extractor / tvp.py

diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py
index cc26f417a0c208efd946aa91c72086471d69ccaa..accff75b5b9d4f959836a56e004b8464696ff7f7 100644 (file)
--- a/youtube_dl/extractor/tvp.py
+++ b/youtube_dl/extractor/tvp.py
@@ -1,142 +1,252 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

+import itertools

 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

+from ..utils import (
+    clean_html,
+    determine_ext,
+    ExtractorError,
+    get_element_by_attribute,
+    orderedSet,
+)

 
 
 
 

-class TvpIE(InfoExtractor):
-    IE_NAME = 'tvp.pl'
-    _VALID_URL = r'https?://(?:vod|www)\.tvp\.pl/.*/(?P<id>\d+)$'
+class TVPIE(InfoExtractor):
+    IE_NAME = 'tvp'
+    IE_DESC = 'Telewizja Polska'
+    _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:video/(?:[^,\s]*,)*|(?:(?!\d+/)[^/]+/)*)(?P<id>\d+)'

 
     _TESTS = [{
 
     _TESTS = [{

-        '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.',
-        },
-    }, {
-        'url': 'http://vod.tvp.pl/seriale/obyczajowe/czas-honoru/sezon-1-1-13/i-seria-odc-13/194536',
+        'url': 'https://vod.tvp.pl/video/czas-honoru,i-seria-odc-13,194536',
+        'md5': 'a21eb0aa862f25414430f15fdfb9e76c',

         'info_dict': {
             'id': '194536',
             'ext': 'mp4',
         '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.',
+            'title': 'Czas honoru, odc. 13 – Władek',
+            'description': 'md5:437f48b93558370b031740546b696e24',

         },
     }, {
         'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
         },
     }, {
         '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',
         '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',

         },
         },

-        'params': {
-            # m3u8 download
-            'skip_download': 'true',
+    }, {
+        # 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',
     }, {
         '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) 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<id>\d+)'
+
+    _TESTS = [{
+        'url': 'tvp:194536',
+        'md5': 'a21eb0aa862f25414430f15fdfb9e76c',

         'info_dict': {
         'info_dict': {

-            'id': '17834272',
+            'id': '194536',

             'ext': 'mp4',
             '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…',
+            'title': 'Czas honoru, odc. 13 – Władek',

         },
         },

-        'params': {
-            # m3u8 download
-            'skip_download': 'true',
+    }, {
+        # 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)
     }]
 
     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)
 
         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 = self._html_search_regex(
+            r'(?s)<p[^>]+\bclass=["\']notAvailable__text["\'][^>]*>(.+?)</p>',
+            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)), 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)
             webpage, 'series', group='series', default=None)

-        if series is not None and series not in title:
-            title = '%s, %s' % (series, title)
-        description = self._og_search_description(webpage, 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(
 
         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']
 
             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,
             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)
 
         return {
             'id': video_id,
             'title': title,
 
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': title,

-            'thumbnail': self._og_search_thumbnail(webpage),
-            'description': description,
+            'thumbnail': thumbnail,

             'formats': formats,
         }
 
 
             'formats': formats,
         }
 
 

-class TvpSeriesIE(InfoExtractor):
-    IE_NAME = 'tvp.pl:Series'
-    _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$'
+class TVPWebsiteIE(InfoExtractor):
+    IE_NAME = 'tvp:series'
+    _VALID_URL = r'https?://vod\.tvp\.pl/website/(?P<display_id>[^,]+),(?P<id>\d+)'

 
     _TESTS = [{
 
     _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': {
         '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': {
         'info_dict': {

-            'title': 'Boso przez świat',
-            'id': '9329207',
+            'id': '36637049',
+            'ext': 'mp4',
+            'title': 'Gloria, Gloria',

         },
         },

-        'playlist_count': 86,
+        'params': {
+            'skip_download': True,
+        },
+        '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)