[rtlnl] Extend _VALID_URL (#26549) (closes #25821)
[youtube-dl] / youtube_dl / extractor / rtlnl.py
index 0ab1eb69c8c82ae0bc04ec135971cd919b17a155..cf4dc85db2336c6672f981f4212d46f5cfe7af62 100644 (file)
+# coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import parse_duration
+from ..utils import (
+    int_or_none,
+    parse_duration,
+)
 
 
-class RtlXlIE(InfoExtractor):
-    IE_NAME = 'rtlxl.nl'
-    _VALID_URL = r'https?://www\.rtlxl\.nl/#!/[^/]+/(?P<uuid>[^/?]+)'
+class RtlNlIE(InfoExtractor):
+    IE_NAME = 'rtl.nl'
+    IE_DESC = 'rtl.nl and rtlxl.nl'
+    _VALID_URL = r'''(?x)
+        https?://(?:(?:www|static)\.)?
+        (?:
+            rtlxl\.nl/(?:[^\#]*\#!|programma)/[^/]+/|
+            rtl\.nl/(?:(?:system/videoplayer/(?:[^/]+/)+(?:video_)?embed\.html|embed)\b.+?\buuid=|video/)
+        )
+        (?P<id>[0-9a-f-]+)'''
 
-    _TEST = {
-        'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
-        'md5': 'cc16baa36a6c169391f0764fa6b16654',
+    _TESTS = [{
+        # new URL schema
+        'url': 'https://www.rtlxl.nl/programma/rtl-nieuws/0bd1384d-d970-3086-98bb-5c104e10c26f',
+        'md5': '490428f1187b60d714f34e1f2e3af0b6',
+        'info_dict': {
+            'id': '0bd1384d-d970-3086-98bb-5c104e10c26f',
+            'ext': 'mp4',
+            'title': 'RTL Nieuws',
+            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+            'timestamp': 1593293400,
+            'upload_date': '20200627',
+            'duration': 661.08,
+        },
+    }, {
+        # old URL schema
+        'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/82b1aad1-4a14-3d7b-b554-b0aed1b2c416',
+        'md5': '473d1946c1fdd050b2c0161a4b13c373',
         'info_dict': {
-            'id': '6e4203a6-0a5e-3596-8424-c599a59e0677',
+            'id': '82b1aad1-4a14-3d7b-b554-b0aed1b2c416',
             'ext': 'mp4',
-            'title': 'RTL Nieuws - Laat',
-            'description': 'md5:6b61f66510c8889923b11f2778c72dc5',
-            'timestamp': 1408051800,
-            'upload_date': '20140814',
-            'duration': 576.880,
+            'title': 'RTL Nieuws',
+            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+            'timestamp': 1461951000,
+            'upload_date': '20160429',
+            'duration': 1167.96,
         },
-    }
+        'skip': '404',
+    }, {
+        # best format available a3t
+        'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed/autoplay=false',
+        'md5': 'dea7474214af1271d91ef332fb8be7ea',
+        'info_dict': {
+            'id': '84ae5571-ac25-4225-ae0c-ef8d9efb2aed',
+            'ext': 'mp4',
+            'timestamp': 1424039400,
+            'title': 'RTL Nieuws - Nieuwe beelden Kopenhagen: chaos direct na aanslag',
+            'thumbnail': r're:^https?://screenshots\.rtl\.nl/(?:[^/]+/)*sz=[0-9]+x[0-9]+/uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed$',
+            'upload_date': '20150215',
+            'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.',
+        }
+    }, {
+        # empty synopsis and missing episodes (see https://github.com/ytdl-org/youtube-dl/issues/6275)
+        # best format available nettv
+        'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a/autoplay=false',
+        'info_dict': {
+            'id': 'f536aac0-1dc3-4314-920e-3bd1c5b3811a',
+            'ext': 'mp4',
+            'title': 'RTL Nieuws - Meer beelden van overval juwelier',
+            'thumbnail': r're:^https?://screenshots\.rtl\.nl/(?:[^/]+/)*sz=[0-9]+x[0-9]+/uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a$',
+            'timestamp': 1437233400,
+            'upload_date': '20150718',
+            'duration': 30.474,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # encrypted m3u8 streams, georestricted
+        'url': 'http://www.rtlxl.nl/#!/afl-2-257632/52a74543-c504-4cde-8aa8-ec66fe8d68a7',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.rtl.nl/system/videoplayer/derden/embed.html#!/uuid=bb0353b0-d6a4-1dad-90e9-18fe75b8d1f0',
+        'only_matching': True,
+    }, {
+        'url': 'http://rtlxl.nl/?_ga=1.204735956.572365465.1466978370#!/rtl-nieuws-132237/3c487912-023b-49ac-903e-2c5d79f8410f',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.rtl.nl/video/c603c9c2-601d-4b5e-8175-64f1e942dc7d/',
+        'only_matching': True,
+    }, {
+        'url': 'https://static.rtl.nl/embed/?uuid=1a2970fc-5c0b-43ff-9fdc-927e39e6d1bc&autoplay=false&publicatiepunt=rtlnieuwsnl',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        uuid = mobj.group('uuid')
-
+        uuid = self._match_id(url)
         info = self._download_json(
-            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
+            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=adaptive/' % uuid,
             uuid)
 
         material = info['material'][0]
-        episode_info = info['episodes'][0]
+        title = info['abstracts'][0]['name']
+        subtitle = material.get('title')
+        if subtitle:
+            title += ' - %s' % subtitle
+        description = material.get('synopsis')
 
-        progname = info['abstracts'][0]['name']
-        subtitle = material['title'] or info['episodes'][0]['name']
+        meta = info.get('meta', {})
 
         videopath = material['videopath']
-        f4m_url = 'http://manifest.us.rtl.nl' + videopath
+        m3u8_url = meta.get('videohost', 'http://manifest.us.rtl.nl') + videopath
+
+        formats = self._extract_m3u8_formats(
+            m3u8_url, uuid, 'mp4', m3u8_id='hls', fatal=False)
+        self._sort_formats(formats)
 
-        formats = self._extract_f4m_formats(f4m_url, uuid)
+        thumbnails = []
 
-        video_urlpart = videopath.split('/flash/')[1][:-4]
-        PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4'
+        for p in ('poster_base_url', '"thumb_base_url"'):
+            if not meta.get(p):
+                continue
 
-        formats.extend([
-            {
-                'url': PG_URL_TEMPLATE % ('a2m', video_urlpart),
-                'format_id': 'pg-sd',
-            },
-            {
-                'url': PG_URL_TEMPLATE % ('a3m', video_urlpart),
-                'format_id': 'pg-hd',
-            }
-        ])
+            thumbnails.append({
+                'url': self._proto_relative_url(meta[p] + uuid),
+                'width': int_or_none(self._search_regex(
+                    r'/sz=([0-9]+)', meta[p], 'thumbnail width', fatal=False)),
+                'height': int_or_none(self._search_regex(
+                    r'/sz=[0-9]+x([0-9]+)',
+                    meta[p], 'thumbnail height', fatal=False))
+            })
 
         return {
             'id': uuid,
-            'title': '%s - %s' % (progname, subtitle),
+            'title': title,
             'formats': formats,
             'timestamp': material['original_date'],
-            'description': episode_info['synopsis'],
+            'description': description,
             'duration': parse_duration(material.get('duration')),
+            'thumbnails': thumbnails,
         }