Merge b663bba81d into c5098961b0

youtube_dl/extractor/extractors.py

@@ -672,7 +672,11 @@ from .massengeschmacktv import MassengeschmackTVIE
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
 from .medaltv import MedalTVIE
-from .mediaset import MediasetIE
+from .mediaset import (
+    MediasetIE,
+    MediasetClipIE,
+    MediasetShowIE,
+)
 from .mediasite import (
     MediasiteIE,
     MediasiteCatalogIE,

youtube_dl/extractor/mediaset.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import functools
 import re
 
 from .theplatform import ThePlatformBaseIE
@@ -10,7 +11,11 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    GeoRestrictedError,
     int_or_none,
+    OnDemandPagedList,
+    try_get,
+    urljoin,
     update_url_query,
 )
 
@@ -30,37 +35,110 @@ class MediasetIE(ThePlatformBaseIE):
                     '''
     _TESTS = [{
         # full episode
-        'url': 'https://www.mediasetplay.mediaset.it/video/hellogoodbye/quarta-puntata_FAFU000000661824',
+        'url': 'https://www.mediasetplay.mediaset.it/video/mrwronglezionidamore/episodio-1_F310575103000102',
-        'md5': '9b75534d42c44ecef7bf1ffeacb7f85d',
+        'md5': 'a7e75c6384871f322adb781d3bd72c26',
         'info_dict': {
-            'id': 'FAFU000000661824',
+            'id': 'F310575103000102',
             'ext': 'mp4',
-            'title': 'Quarta puntata',
+            'title': 'Episodio 1',
-            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+            'description': 'md5:e8017b7d7194e9bfb75299c2b8d81e02',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 1414.26,
+            'duration': 2682.0,
-            'upload_date': '20161107',
+            'upload_date': '20210530',
-            'series': 'Hello Goodbye',
+            'series': 'Mr Wrong - Lezioni d\'amore',
-            'timestamp': 1478532900,
+            'timestamp': 1622413946,
-            'uploader': 'Rete 4',
+            'uploader': 'Canale 5',
-            'uploader_id': 'R4',
+            'uploader_id': 'C5',
+            'season': 'Season 1',
+            'episode': 'Episode 1',
+            'season_number': 1,
+            'episode_number': 1,
+            'chapters': [{'start_time': 0.0, 'end_time': 439.88}, {'start_time': 439.88, 'end_time': 1685.84}, {'start_time': 1685.84, 'end_time': 2682.0}],
         },
+        'skip': 'Geo restricted',
     }, {
         'url': 'https://www.mediasetplay.mediaset.it/video/matrix/puntata-del-25-maggio_F309013801000501',
-        'md5': '288532f0ad18307705b01e581304cd7b',
+        'md5': '1276f966ac423d16ba255ce867de073e',
         'info_dict': {
             'id': 'F309013801000501',
             'ext': 'mp4',
             'title': 'Puntata del 25 maggio',
-            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+            'description': 'md5:ee2e456e3eb1dba5e814596655bb5296',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 6565.007,
+            'duration': 6565.008,
-            'upload_date': '20180526',
+            'upload_date': '20200903',
             'series': 'Matrix',
-            'timestamp': 1527326245,
+            'timestamp': 1599172492,
             'uploader': 'Canale 5',
             'uploader_id': 'C5',
+            'season': 'Season 5',
+            'episode': 'Episode 5',
+            'season_number': 5,
+            'episode_number': 5,
+            'chapters': [{'start_time': 0.0, 'end_time': 3409.08}, {'start_time': 3409.08, 'end_time': 6565.008}],
         },
+    }, {
+        'url': 'https://www.mediasetplay.mediaset.it/video/cameracafe5/episodio-69-pezzo-di-luna_F303843101017801',
+        'md5': 'd1650ac9ff944f185556126a736df148',
+        'info_dict': {
+            'id': 'F303843101017801',
+            'ext': 'mp4',
+            'title': 'Episodio 69 - Pezzo di luna',
+            'description': 'md5:7c32c8ec4118b72588b9412f11353f73',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 263.008,
+            'upload_date': '20200902',
+            'series': 'Camera Café 5',
+            'timestamp': 1599064700,
+            'uploader': 'Italia 1',
+            'uploader_id': 'I1',
+            'season': 'Season 5',
+            'episode': 'Episode 178',
+            'season_number': 5,
+            'episode_number': 178,
+            'chapters': [{'start_time': 0.0, 'end_time': 261.88}, {'start_time': 261.88, 'end_time': 263.008}],
+        },
+        'skip': 'Geo restricted',
+    }, {
+        'url': 'https://www.mediasetplay.mediaset.it/video/cameracafe5/episodio-51-tu-chi-sei_F303843107000601',
+        'md5': '567e9ad375b7a27a0e370650f572a1e3',
+        'info_dict': {
+            'id': 'F303843107000601',
+            'ext': 'mp4',
+            'title': 'Episodio 51 - Tu chi sei?',
+            'description': 'md5:42ef006e56824cc31787a547590923f4',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 367.021,
+            'upload_date': '20200902',
+            'series': 'Camera Café 5',
+            'timestamp': 1599069817,
+            'uploader': 'Italia 1',
+            'uploader_id': 'I1',
+            'season': 'Season 5',
+            'episode': 'Episode 6',
+            'season_number': 5,
+            'episode_number': 6,
+            'chapters': [{'start_time': 0.0, 'end_time': 358.68}, {'start_time': 358.68, 'end_time': 367.021}],
+        },
+        'skip': 'Geo restricted',
+    }, {
+        # movie
+        'url': 'https://www.mediasetplay.mediaset.it/movie/selvaggi/selvaggi_F006474501000101',
+        'md5': '720440187a2ae26af8148eb9e6b901ed',
+        'info_dict': {
+            'id': 'F006474501000101',
+            'ext': 'mp4',
+            'title': 'Selvaggi',
+            'description': 'md5:cfdedbbfdd12d4d0e5dcf1fa1b75284f',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 5233.01,
+            'upload_date': '20210729',
+            'timestamp': 1627594716,
+            'uploader': 'Cine34',
+            'uploader_id': 'B6',
+            'chapters': [{'start_time': 0.0, 'end_time': 1938.56}, {'start_time': 1938.56, 'end_time': 5233.01}],
+        },
+        'skip': 'Geo restricted',
     }, {
         # clip
         'url': 'https://www.mediasetplay.mediaset.it/video/gogglebox/un-grande-classico-della-commedia-sexy_FAFU000000661680',
@@ -131,6 +209,22 @@ class MediasetIE(ThePlatformBaseIE):
             video.attrib['src'] = re.sub(r'(https?://vod05)t(-mediaset-it\.akamaized\.net/.+?.mpd)\?.+', r'\1\2', video.attrib['src'])
         return super(MediasetIE, self)._parse_smil_formats(smil, smil_url, video_id, namespace, f4m_params, transform_rtmp_url)
 
+    def _check_drm_formats(self, tp_formats, video_id):
+        has_nondrm, drm_manifest = False, ''
+        for f in tp_formats:
+            if '_sampleaes/' in (f.get('manifest_url') or ''):
+                drm_manifest = drm_manifest or f['manifest_url']
+                f['has_drm'] = True
+            if not has_nondrm and not f.get('has_drm') and f.get('manifest_url'):
+                has_nondrm = True
+
+        nodrm_manifest = re.sub(r'_sampleaes/(\w+)_fp_', r'/\1_no_', drm_manifest)
+        if has_nondrm or nodrm_manifest == drm_manifest:
+            return
+
+        tp_formats.extend(self._extract_m3u8_formats(
+            nodrm_manifest, video_id, m3u8_id='hls', fatal=False) or [])
+
     def _real_extract(self, url):
         guid = self._match_id(url)
         tp_path = 'PR1GhC/media/guid/2702976343/' + guid
@@ -138,49 +232,164 @@ class MediasetIE(ThePlatformBaseIE):
 
         formats = []
         subtitles = {}
-        first_e = None
+        first_e = geo_e = None
-        for asset_type in ('SD', 'HD'):
+        asset_type = 'geoNo:HD,browser,geoIT|geoNo:HD,geoIT|geoNo:SD,browser,geoIT|geoNo:SD,geoIT|geoNo|HD|SD'
-            # TODO: fixup ISM+none manifest URLs
+        # TODO: fixup ISM+none manifest URLs
-            for f in ('MPEG4', 'MPEG-DASH+none', 'M3U+none'):
+        for f in ('MPEG4', 'M3U'):
-                try:
+            try:
-                    tp_formats, tp_subtitles = self._extract_theplatform_smil(
+                tp_formats, tp_subtitles = self._extract_theplatform_smil(
-                        update_url_query('http://link.theplatform.%s/s/%s' % (self._TP_TLD, tp_path), {
+                    update_url_query('http://link.theplatform.%s/s/%s' % (self._TP_TLD, tp_path), {
-                            'mbr': 'true',
+                        'mbr': 'true',
-                            'formats': f,
+                        'formats': f,
-                            'assetTypes': asset_type,
+                        'assetTypes': asset_type,
-                        }), guid, 'Downloading %s %s SMIL data' % (f.split('+')[0], asset_type))
+                    }), guid, 'Downloading %s SMIL data' % (f.split('+')[0]))
-                except ExtractorError as e:
+            except ExtractorError as e:
-                    if not first_e:
+                if not first_e:
-                        first_e = e
+                    first_e = e
-                    break
+                if not geo_e and isinstance(e, GeoRestrictedError):
-                for tp_f in tp_formats:
+                    geo_e = e
-                    tp_f['quality'] = 1 if asset_type == 'HD' else 0
+                continue
-                formats.extend(tp_formats)
+            self._check_drm_formats(tp_formats, guid)
-                subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+            formats.extend(tp_formats)
-        if first_e and not formats:
+            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+
+        # check for errors and report them
+        if (first_e or geo_e) and not formats:
+            if geo_e:
+                raise geo_e
+            if 'None of the available releases match' in first_e.message:
+                raise ExtractorError('No non-DRM formats available', cause=first_e)
             raise first_e
+
         self._sort_formats(formats)
 
-        fields = []
-        for templ, repls in (('tvSeason%sNumber', ('', 'Episode')), ('mediasetprogram$%s', ('brandTitle', 'numberOfViews', 'publishInfo'))):
-            fields.extend(templ % repl for repl in repls)
         feed_data = self._download_json(
-            'https://feed.entertainment.tv.theplatform.eu/f/PR1GhC/mediaset-prod-all-programs/guid/-/' + guid,
+            'https://feed.entertainment.tv.theplatform.eu/f/PR1GhC/mediaset-prod-all-programs-v2/guid/-/' + guid,
-            guid, fatal=False, query={'fields': ','.join(fields)})
+            guid, fatal=False)
         if feed_data:
             publish_info = feed_data.get('mediasetprogram$publishInfo') or {}
+            thumbnails = feed_data.get('thumbnails') or {}
+            thumbnail = None
+            for key, value in thumbnails.items():
+                if key.startswith('image_keyframe_poster-'):
+                    thumbnail = value.get('url')
+                    break
+
             info.update({
-                'episode_number': int_or_none(feed_data.get('tvSeasonEpisodeNumber')),
+                'description': info.get('description') or feed_data.get('description') or feed_data.get('longDescription'),
-                'season_number': int_or_none(feed_data.get('tvSeasonNumber')),
-                'series': feed_data.get('mediasetprogram$brandTitle'),
                 'uploader': publish_info.get('description'),
                 'uploader_id': publish_info.get('channel'),
                 'view_count': int_or_none(feed_data.get('mediasetprogram$numberOfViews')),
+                'thumbnail': thumbnail,
             })
 
+            if feed_data.get('programType') == 'episode':
+                info.update({
+                    'episode_number': int_or_none(
+                        feed_data.get('tvSeasonEpisodeNumber')),
+                    'season_number': int_or_none(
+                        feed_data.get('tvSeasonNumber')),
+                    'series': feed_data.get('mediasetprogram$brandTitle'),
+                })
+
         info.update({
             'id': guid,
             'formats': formats,
             'subtitles': subtitles,
         })
         return info
+
+
+class MediasetClipIE(MediasetIE):
+    _VALID_URL = r'https?://(?:www\.)?\w+\.mediaset\.it/video/(?:[^/]+/)*[\w-]+_(?P<id>\d+)\.s?html?'
+    _TESTS = [{
+        'url': 'https://www.grandefratello.mediaset.it/video/ventinovesima-puntata_27071.shtml',
+        'info_dict': {
+            'id': 'F310293901002901',
+            'ext': 'mp4',
+        },
+        'skip': 'Geo restricted, DRM content',
+    }]
+
+    def _real_extract(self, url):
+        clip_id = self._match_id(url)
+        webpage = self._download_webpage(url, clip_id)
+        guid = self._search_regex(
+            (r'''var\s*_onplay_guid\s*=\s*(?P<q>'|"|\b)(?P<guid>[\dA-Z]{16,})(?P=q)\s*;''',
+             r'\bGUID\s+(?P<guid>[\dA-Z]{16,})\b', ),
+            webpage, 'clip GUID', group='guid')
+        return self.url_result('mediaset:%s' % guid, ie='Mediaset', video_id=clip_id)
+
+
+class MediasetShowIE(MediasetIE):
+    _VALID_URL = r'''(?x)
+                    (?:
+                        https?://
+                            (?:(?:www|static3)\.)?mediasetplay\.mediaset\.it/
+                            (?:
+                                (?:fiction|programmi-tv|serie-tv)/(?:.+?/)?
+                                    (?:[a-z-]+)_SE(?P<id>\d{12})
+                                    (?:,ST(?P<st>\d{12}))?
+                                    (?:,sb(?P<sb>\d{9}))?$
+                            )
+                    )
+                    '''
+    _TESTS = [{
+        # TV Show webpage (general webpage)
+        'url': 'https://www.mediasetplay.mediaset.it/programmi-tv/leiene/leiene_SE000000000061',
+        'info_dict': {
+            'id': '000000000061',
+            'title': 'Le Iene',
+        },
+        'playlist_mincount': 7,
+    }, {
+        # TV Show webpage (specific season)
+        'url': 'https://www.mediasetplay.mediaset.it/programmi-tv/leiene/leiene_SE000000000061,ST000000002763',
+        'info_dict': {
+            'id': '000000002763',
+            'title': 'Le Iene',
+        },
+        'playlist_mincount': 7,
+    }, {
+        # TV Show specific playlist (with multiple pages)
+        'url': 'https://www.mediasetplay.mediaset.it/programmi-tv/leiene/iservizi_SE000000000061,ST000000002763,sb100013375',
+        'info_dict': {
+            'id': '100013375',
+            'title': 'I servizi',
+        },
+        'playlist_mincount': 50,
+    }]
+
+    _BY_SUBBRAND = 'https://feed.entertainment.tv.theplatform.eu/f/PR1GhC/mediaset-prod-all-programs-v2?byCustomValue={subBrandId}{%s}&sort=:publishInfo_lastPublished|desc,tvSeasonEpisodeNumber|desc&range=%d-%d'
+    _PAGE_SIZE = 25
+    _match_valid_url = lambda s, u: re.match(s._VALID_URL, u)
+
+    def _fetch_page(self, sb, page):
+        lower_limit = page * self._PAGE_SIZE + 1
+        upper_limit = lower_limit + self._PAGE_SIZE - 1
+        content = self._download_json(
+            self._BY_SUBBRAND % (sb, lower_limit, upper_limit), sb)
+        for entry in content.get('entries') or []:
+            res = self.url_result('mediaset:' + entry['guid'])
+            if res:
+                res['playlist_title'] = entry['mediasetprogram$subBrandDescription']
+            yield res
+
+    def _real_extract(self, url):
+
+        playlist_id, st, sb = self._match_valid_url(url).group('id', 'st', 'sb')
+        if not sb:
+            page = self._download_webpage(url, st or playlist_id)
+            entries = [self.url_result(urljoin('https://www.mediasetplay.mediaset.it', url))
+                       for url in re.findall(r'href="([^<>=]+SE\d{12},ST\d{12},sb\d{9})">[^<]+<', page)]
+            title = (self._html_search_regex(r'(?s)<h1[^>]*>(.+?)</h1>', page, 'title', default=None)
+                     or self._og_search_title(page))
+            return self.playlist_result(entries, st or playlist_id, title)
+
+        entries = OnDemandPagedList(
+            functools.partial(self._fetch_page, sb),
+            self._PAGE_SIZE)
+        # slice explicitly, as no __getitem__ in OnDemandPagedList yet
+        title = try_get(entries, lambda x: x.getslice(0, 1)[0]['playlist_title'])
+
+        return self.playlist_result(entries, sb, title)

youtube_dl/utils.py

@@ -4106,6 +4106,12 @@ class PagedList(object):
         # This is only useful for tests
         return len(self.getslice())
 
+    def _getslice(self, start, end):
+        raise NotImplementedError('This method must be implemented by subclasses')
+
+    def getslice(self, start=0, end=None):
+        return list(self._getslice(start, end))
+
 
 class OnDemandPagedList(PagedList):
     def __init__(self, pagefunc, pagesize, use_cache=True):
@@ -4115,11 +4121,12 @@ class OnDemandPagedList(PagedList):
         if use_cache:
             self._cache = {}
 
-    def getslice(self, start=0, end=None):
+    def _getslice(self, start=0, end=None):
-        res = []
+        firstpage = start // self._pagesize
-        for pagenum in itertools.count(start // self._pagesize):
+        nextfirstid = firstpage * self._pagesize
-            firstid = pagenum * self._pagesize
+        for pagenum in itertools.count(firstpage):
-            nextfirstid = pagenum * self._pagesize + self._pagesize
+            firstid = nextfirstid
+            nextfirstid += self._pagesize
             if start >= nextfirstid:
                 continue
 
@@ -4132,18 +4139,19 @@ class OnDemandPagedList(PagedList):
                 self._cache[pagenum] = page_results
 
             startv = (
-                start % self._pagesize
+                start - firstid
                 if firstid <= start < nextfirstid
                 else 0)
 
             endv = (
-                ((end - 1) % self._pagesize) + 1
+                end - firstid
                 if (end is not None and firstid <= end <= nextfirstid)
                 else None)
 
             if startv != 0 or endv is not None:
                 page_results = page_results[startv:endv]
-            res.extend(page_results)
+            for item in page_results:
+                yield item
 
             # A little optimization - if current page is not "full", ie. does
             # not contain page_size videos then we can assume that this page
@@ -4156,7 +4164,7 @@ class OnDemandPagedList(PagedList):
             # break out early as well
             if end == nextfirstid:
                 break
-        return res
+        return
 
 
 class InAdvancePagedList(PagedList):