]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/tvnow.py
Update changelog.
[youtubedl] / youtube_dl / extractor / tvnow.py
index e2169f2bce30a3bc42fcb422bd6978dda0175f3b..60937616f2c57e55cf093cbfe5f1d55987c75b55 100644 (file)
@@ -7,8 +7,10 @@ from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
     ExtractorError,
 from ..compat import compat_str
 from ..utils import (
     ExtractorError,
+    int_or_none,
     parse_iso8601,
     parse_duration,
     parse_iso8601,
     parse_duration,
+    try_get,
     update_url_query,
 )
 
     update_url_query,
 )
 
@@ -16,8 +18,9 @@ from ..utils import (
 class TVNowBaseIE(InfoExtractor):
     _VIDEO_FIELDS = (
         'id', 'title', 'free', 'geoblocked', 'articleLong', 'articleShort',
 class TVNowBaseIE(InfoExtractor):
     _VIDEO_FIELDS = (
         'id', 'title', 'free', 'geoblocked', 'articleLong', 'articleShort',
-        'broadcastStartDate', 'isDrm', 'duration', 'manifest.dashclear',
-        'format.defaultImage169Format', 'format.defaultImage169Logo')
+        'broadcastStartDate', 'isDrm', 'duration', 'season', 'episode',
+        'manifest.dashclear', 'manifest.hlsclear', 'manifest.smoothclear',
+        'format.title', 'format.defaultImage169Format', 'format.defaultImage169Logo')
 
     def _call_api(self, path, video_id, query):
         return self._download_json(
 
     def _call_api(self, path, video_id, query):
         return self._download_json(
@@ -28,27 +31,42 @@ class TVNowBaseIE(InfoExtractor):
         video_id = compat_str(info['id'])
         title = info['title']
 
         video_id = compat_str(info['id'])
         title = info['title']
 
-        mpd_url = info['manifest']['dashclear']
-        if not mpd_url:
+        paths = []
+        for manifest_url in (info.get('manifest') or {}).values():
+            if not manifest_url:
+                continue
+            manifest_url = update_url_query(manifest_url, {'filter': ''})
+            path = self._search_regex(r'https?://[^/]+/(.+?)\.ism/', manifest_url, 'path')
+            if path in paths:
+                continue
+            paths.append(path)
+
+            def url_repl(proto, suffix):
+                return re.sub(
+                    r'(?:hls|dash|hss)([.-])', proto + r'\1', re.sub(
+                        r'\.ism/(?:[^.]*\.(?:m3u8|mpd)|[Mm]anifest)',
+                        '.ism/' + suffix, manifest_url))
+
+            formats = self._extract_mpd_formats(
+                url_repl('dash', '.mpd'), video_id,
+                mpd_id='dash', fatal=False)
+            formats.extend(self._extract_ism_formats(
+                url_repl('hss', 'Manifest'),
+                video_id, ism_id='mss', fatal=False))
+            formats.extend(self._extract_m3u8_formats(
+                url_repl('hls', '.m3u8'), video_id, 'mp4',
+                'm3u8_native', m3u8_id='hls', fatal=False))
+            if formats:
+                break
+        else:
             if info.get('isDrm'):
                 raise ExtractorError(
                     'Video %s is DRM protected' % video_id, expected=True)
             if info.get('geoblocked'):
             if info.get('isDrm'):
                 raise ExtractorError(
                     'Video %s is DRM protected' % video_id, expected=True)
             if info.get('geoblocked'):
-                raise ExtractorError(
-                    'Video %s is not available from your location due to geo restriction' % video_id,
-                    expected=True)
+                raise self.raise_geo_restricted()
             if not info.get('free', True):
                 raise ExtractorError(
                     'Video %s is not available for free' % video_id, expected=True)
             if not info.get('free', True):
                 raise ExtractorError(
                     'Video %s is not available for free' % video_id, expected=True)
-
-        mpd_url = update_url_query(mpd_url, {'filter': ''})
-        formats = self._extract_mpd_formats(mpd_url, video_id, mpd_id='dash', fatal=False)
-        formats.extend(self._extract_ism_formats(
-            mpd_url.replace('dash.', 'hss.').replace('/.mpd', '/Manifest'),
-            video_id, ism_id='mss', fatal=False))
-        formats.extend(self._extract_m3u8_formats(
-            mpd_url.replace('dash.', 'hls.').replace('/.mpd', '/.m3u8'),
-            video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
         self._sort_formats(formats)
 
         description = info.get('articleLong') or info.get('articleShort')
         self._sort_formats(formats)
 
         description = info.get('articleLong') or info.get('articleShort')
@@ -56,76 +74,123 @@ class TVNowBaseIE(InfoExtractor):
         duration = parse_duration(info.get('duration'))
 
         f = info.get('format', {})
         duration = parse_duration(info.get('duration'))
 
         f = info.get('format', {})
+
+        thumbnails = [{
+            'url': 'https://aistvnow-a.akamaihd.net/tvnow/movie/%s' % video_id,
+        }]
         thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo')
         thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo')
+        if thumbnail:
+            thumbnails.append({
+                'url': thumbnail,
+            })
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
             'description': description,
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
             'description': description,
-            'thumbnail': thumbnail,
+            'thumbnails': thumbnails,
             'timestamp': timestamp,
             'duration': duration,
             'timestamp': timestamp,
             'duration': duration,
+            'series': f.get('title'),
+            'season_number': int_or_none(info.get('season')),
+            'episode_number': int_or_none(info.get('episode')),
+            'episode': title,
             'formats': formats,
         }
 
 
 class TVNowIE(TVNowBaseIE):
             'formats': formats,
         }
 
 
 class TVNowIE(TVNowBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?tvnow\.(?:de|at|ch)/(?:rtl(?:2|plus)?|nitro|superrtl|ntv|vox)/(?P<show_id>[^/]+)/(?:(?:list/[^/]+|jahr/\d{4}/\d{1,2})/)?(?P<id>[^/]+)/(?:player|preview)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?tvnow\.(?:de|at|ch)/(?P<station>[^/]+)/
+                        (?P<show_id>[^/]+)/
+                        (?!(?:list|jahr)(?:/|$))(?P<id>[^/?\#&]+)
+                    '''
 
     _TESTS = [{
 
     _TESTS = [{
-        # rtl
-        'url': 'https://www.tvnow.de/rtl/alarm-fuer-cobra-11/freier-fall/player?return=/rtl',
+        'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/der-neue-porsche-911-gt-3/player',
         'info_dict': {
         'info_dict': {
-            'id': '385314',
-            'display_id': 'alarm-fuer-cobra-11/freier-fall',
+            'id': '331082',
+            'display_id': 'grip-das-motormagazin/der-neue-porsche-911-gt-3',
             'ext': 'mp4',
             'ext': 'mp4',
-            'title': 'Freier Fall',
-            'description': 'md5:8c2d8f727261adf7e0dc18366124ca02',
+            'title': 'Der neue Porsche 911 GT 3',
+            'description': 'md5:6143220c661f9b0aae73b245e5d898bb',
             'thumbnail': r're:^https?://.*\.jpg$',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'timestamp': 1512677700,
-            'upload_date': '20171207',
-            'duration': 2862.0,
+            'timestamp': 1495994400,
+            'upload_date': '20170528',
+            'duration': 5283,
+            'series': 'GRIP - Das Motormagazin',
+            'season_number': 14,
+            'episode_number': 405,
+            'episode': 'Der neue Porsche 911 GT 3',
         },
     }, {
         # rtl2
         'url': 'https://www.tvnow.de/rtl2/armes-deutschland/episode-0008/player',
         },
     }, {
         # rtl2
         'url': 'https://www.tvnow.de/rtl2/armes-deutschland/episode-0008/player',
-        'only_matching': 'True',
+        'only_matching': True,
     }, {
         # rtlnitro
         'url': 'https://www.tvnow.de/nitro/alarm-fuer-cobra-11-die-autobahnpolizei/auf-eigene-faust-pilot/player',
     }, {
         # rtlnitro
         'url': 'https://www.tvnow.de/nitro/alarm-fuer-cobra-11-die-autobahnpolizei/auf-eigene-faust-pilot/player',
-        'only_matching': 'True',
+        'only_matching': True,
     }, {
         # superrtl
         'url': 'https://www.tvnow.de/superrtl/die-lustigsten-schlamassel-der-welt/u-a-ketchup-effekt/player',
     }, {
         # superrtl
         'url': 'https://www.tvnow.de/superrtl/die-lustigsten-schlamassel-der-welt/u-a-ketchup-effekt/player',
-        'only_matching': 'True',
+        'only_matching': True,
     }, {
         # ntv
         'url': 'https://www.tvnow.de/ntv/startup-news/goetter-in-weiss/player',
     }, {
         # ntv
         'url': 'https://www.tvnow.de/ntv/startup-news/goetter-in-weiss/player',
-        'only_matching': 'True',
+        'only_matching': True,
     }, {
         # vox
         'url': 'https://www.tvnow.de/vox/auto-mobil/neues-vom-automobilmarkt-2017-11-19-17-00-00/player',
     }, {
         # vox
         'url': 'https://www.tvnow.de/vox/auto-mobil/neues-vom-automobilmarkt-2017-11-19-17-00-00/player',
-        'only_matching': 'True',
+        'only_matching': True,
     }, {
         # rtlplus
         'url': 'https://www.tvnow.de/rtlplus/op-ruft-dr-bruckner/die-vernaehte-frau/player',
     }, {
         # rtlplus
         'url': 'https://www.tvnow.de/rtlplus/op-ruft-dr-bruckner/die-vernaehte-frau/player',
-        'only_matching': 'True',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/der-neue-porsche-911-gt-3',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
     }]
 
     def _real_extract(self, url):
-        display_id = '%s/%s' % re.match(self._VALID_URL, url).groups()
+        mobj = re.match(self._VALID_URL, url)
+        display_id = '%s/%s' % mobj.group(2, 3)
 
         info = self._call_api(
             'movies/' + display_id, display_id, query={
                 'fields': ','.join(self._VIDEO_FIELDS),
 
         info = self._call_api(
             'movies/' + display_id, display_id, query={
                 'fields': ','.join(self._VIDEO_FIELDS),
+                'station': mobj.group(1),
             })
 
         return self._extract_video(info, display_id)
 
 
             })
 
         return self._extract_video(info, display_id)
 
 
-class TVNowListIE(TVNowBaseIE):
-    _VALID_URL = r'(?P<base_url>https?://(?:www\.)?tvnow\.(?:de|at|ch)/(?:rtl(?:2|plus)?|nitro|superrtl|ntv|vox)/(?P<show_id>[^/]+)/)list/(?P<id>[^?/#&]+)$'
+class TVNowListBaseIE(TVNowBaseIE):
+    _SHOW_VALID_URL = r'''(?x)
+                    (?P<base_url>
+                        https?://
+                            (?:www\.)?tvnow\.(?:de|at|ch)/[^/]+/
+                            (?P<show_id>[^/]+)
+                    )
+                    '''
+
+    def _extract_list_info(self, display_id, show_id):
+        fields = list(self._SHOW_FIELDS)
+        fields.extend('formatTabs.%s' % field for field in self._SEASON_FIELDS)
+        fields.extend(
+            'formatTabs.formatTabPages.container.movies.%s' % field
+            for field in self._VIDEO_FIELDS)
+        return self._call_api(
+            'formats/seo', display_id, query={
+                'fields': ','.join(fields),
+                'name': show_id + '.php'
+            })
+
+
+class TVNowListIE(TVNowListBaseIE):
+    _VALID_URL = r'%s/(?:list|jahr)/(?P<id>[^?\#&]+)' % TVNowListBaseIE._SHOW_VALID_URL
 
     _SHOW_FIELDS = ('title', )
     _SEASON_FIELDS = ('id', 'headline', 'seoheadline', )
 
     _SHOW_FIELDS = ('title', )
     _SEASON_FIELDS = ('id', 'headline', 'seoheadline', )
@@ -138,38 +203,94 @@ class TVNowListIE(TVNowBaseIE):
             'title': '30 Minuten Deutschland - Aktuell',
         },
         'playlist_mincount': 1,
             'title': '30 Minuten Deutschland - Aktuell',
         },
         'playlist_mincount': 1,
+    }, {
+        'url': 'https://www.tvnow.de/vox/ab-ins-beet/list/staffel-14',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/jahr/2018/3',
+        'only_matching': True,
     }]
 
     }]
 
+    @classmethod
+    def suitable(cls, url):
+        return (False if TVNowIE.suitable(url)
+                else super(TVNowListIE, cls).suitable(url))
+
     def _real_extract(self, url):
         base_url, show_id, season_id = re.match(self._VALID_URL, url).groups()
 
     def _real_extract(self, url):
         base_url, show_id, season_id = re.match(self._VALID_URL, url).groups()
 
-        fields = []
-        fields.extend(self._SHOW_FIELDS)
-        fields.extend('formatTabs.%s' % field for field in self._SEASON_FIELDS)
-        fields.extend(
-            'formatTabs.formatTabPages.container.movies.%s' % field
-            for field in self._VIDEO_FIELDS)
-
-        list_info = self._call_api(
-            'formats/seo', season_id, query={
-                'fields': ','.join(fields),
-                'name': show_id + '.php'
-            })
+        list_info = self._extract_list_info(season_id, show_id)
 
         season = next(
             season for season in list_info['formatTabs']['items']
             if season.get('seoheadline') == season_id)
 
 
         season = next(
             season for season in list_info['formatTabs']['items']
             if season.get('seoheadline') == season_id)
 
-        title = '%s - %s' % (list_info['title'], season['headline'])
+        title = list_info.get('title')
+        headline = season.get('headline')
+        if title and headline:
+            title = '%s - %s' % (title, headline)
+        else:
+            title = headline or title
 
         entries = []
         for container in season['formatTabPages']['items']:
 
         entries = []
         for container in season['formatTabPages']['items']:
-            for info in ((container.get('container') or {}).get('movies') or {}).get('items') or []:
+            items = try_get(
+                container, lambda x: x['container']['movies']['items'],
+                list) or []
+            for info in items:
                 seo_url = info.get('seoUrl')
                 if not seo_url:
                     continue
                 seo_url = info.get('seoUrl')
                 if not seo_url:
                     continue
+                video_id = info.get('id')
                 entries.append(self.url_result(
                 entries.append(self.url_result(
-                    base_url + seo_url + '/player', 'TVNow', info.get('id')))
+                    '%s/%s/player' % (base_url, seo_url), TVNowIE.ie_key(),
+                    compat_str(video_id) if video_id else None))
 
         return self.playlist_result(
             entries, compat_str(season.get('id') or season_id), title)
 
         return self.playlist_result(
             entries, compat_str(season.get('id') or season_id), title)
+
+
+class TVNowShowIE(TVNowListBaseIE):
+    _VALID_URL = TVNowListBaseIE._SHOW_VALID_URL
+
+    _SHOW_FIELDS = ('id', 'title', )
+    _SEASON_FIELDS = ('id', 'headline', 'seoheadline', )
+    _VIDEO_FIELDS = ()
+
+    _TESTS = [{
+        'url': 'https://www.tvnow.at/vox/ab-ins-beet',
+        'info_dict': {
+            'id': 'ab-ins-beet',
+            'title': 'Ab ins Beet!',
+        },
+        'playlist_mincount': 7,
+    }, {
+        'url': 'https://www.tvnow.at/vox/ab-ins-beet/list',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/jahr/',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return (False if TVNowIE.suitable(url) or TVNowListIE.suitable(url)
+                else super(TVNowShowIE, cls).suitable(url))
+
+    def _real_extract(self, url):
+        base_url, show_id = re.match(self._VALID_URL, url).groups()
+
+        list_info = self._extract_list_info(show_id, show_id)
+
+        entries = []
+        for season_info in list_info['formatTabs']['items']:
+            season_url = season_info.get('seoheadline')
+            if not season_url:
+                continue
+            season_id = season_info.get('id')
+            entries.append(self.url_result(
+                '%s/list/%s' % (base_url, season_url), TVNowListIE.ie_key(),
+                compat_str(season_id) if season_id else None,
+                season_info.get('headline')))
+
+        return self.playlist_result(entries, show_id, list_info.get('title'))