[rtvs] Fix extractor (#2795)

Closes #2758

Authored by: Bricio

yt_dlp/extractor/rtvs.py

@@ -1,11 +1,19 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 
+from ..utils import (
+    parse_duration,
+    traverse_obj,
+    unified_timestamp,
+)
+
 
 class RTVSIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rtvs\.sk/(?:radio|televizia)/archiv/\d+/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?rtvs\.sk/(?:radio|televizia)/archiv(?:/\d+)?/(?P<id>\d+)/?(?:[#?]|$)'
     _TESTS = [{
         # radio archive
         'url': 'http://www.rtvs.sk/radio/archiv/11224/414872',
@@ -13,23 +21,37 @@ class RTVSIE(InfoExtractor):
         'info_dict': {
             'id': '414872',
             'ext': 'mp3',
-            'title': 'Ostrov pokladov 1 časť.mp3'
+            'title': 'Ostrov pokladov 1 časť.mp3',
-        },
+            'duration': 2854,
-        'params': {
+            'thumbnail': 'https://www.rtvs.sk/media/a501/image/file/2/0000/b1R8.rtvs.jpg',
-            'skip_download': True,
+            'display_id': '135331',
         }
     }, {
         # tv archive
         'url': 'http://www.rtvs.sk/televizia/archiv/8249/63118',
-        'md5': '85e2c55cf988403b70cac24f5c086dc6',
         'info_dict': {
             'id': '63118',
             'ext': 'mp4',
             'title': 'Amaro Džives - Náš deň',
-            'description': 'Galavečer pri príležitosti Medzinárodného dňa Rómov.'
+            'description': 'Galavečer pri príležitosti Medzinárodného dňa Rómov.',
-        },
+            'thumbnail': 'https://www.rtvs.sk/media/a501/image/file/2/0031/L7Qm.amaro_dzives_png.jpg',
-        'params': {
+            'timestamp': 1428555900,
-            'skip_download': True,
+            'upload_date': '20150409',
+            'duration': 4986,
+        }
+    }, {
+        # tv archive
+        'url': 'https://www.rtvs.sk/televizia/archiv/18083?utm_source=web&utm_medium=rozcestnik&utm_campaign=Robin',
+        'info_dict': {
+            'id': '18083',
+            'ext': 'mp4',
+            'title': 'Robin',
+            'description': 'md5:2f70505a7b8364491003d65ff7a0940a',
+            'timestamp': 1636652760,
+            'display_id': '307655',
+            'duration': 831,
+            'upload_date': '20211111',
+            'thumbnail': 'https://www.rtvs.sk/media/a501/image/file/2/0916/robin.jpg',
         }
     }]
 
@@ -37,11 +59,31 @@ class RTVSIE(InfoExtractor):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
+        iframe_id = self._search_regex(
+            r'<iframe[^>]+id\s*=\s*"player_[^_]+_([0-9]+)"', webpage, 'Iframe ID')
+        iframe_url = self._search_regex(
+            fr'<iframe[^>]+id\s*=\s*"player_[^_]+_{re.escape(iframe_id)}"[^>]+src\s*=\s*"([^"]+)"', webpage, 'Iframe URL')
 
-        playlist_url = self._search_regex(
+        webpage = self._download_webpage(iframe_url, video_id, 'Downloading iframe')
-            r'playlist["\']?\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
+        json_url = self._search_regex(r'var\s+url\s*=\s*"([^"]+)"\s*\+\s*ruurl', webpage, 'json URL')
-            'playlist url', group='url')
+        data = self._download_json(f'https:{json_url}b=mozilla&p=win&v=97&f=0&d=1', video_id)
 
-        data = self._download_json(
+        if data.get('clip'):
-            playlist_url, video_id, 'Downloading playlist')[0]
+            data['playlist'] = [data['clip']]
-        return self._parse_jwplayer_data(data, video_id=video_id)
+
+        if traverse_obj(data, ('playlist', 0, 'sources', 0, 'type')) == 'audio/mp3':
+            formats = [{'url': traverse_obj(data, ('playlist', 0, 'sources', 0, 'src'))}]
+        else:
+            formats = self._extract_m3u8_formats(traverse_obj(data, ('playlist', 0, 'sources', 0, 'src')), video_id)
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': iframe_id,
+            'title': traverse_obj(data, ('playlist', 0, 'title')),
+            'description': traverse_obj(data, ('playlist', 0, 'description')),
+            'duration': parse_duration(traverse_obj(data, ('playlist', 0, 'length'))),
+            'thumbnail': traverse_obj(data, ('playlist', 0, 'image')),
+            'timestamp': unified_timestamp(traverse_obj(data, ('playlist', 0, 'datetime_create'))),
+            'formats': formats
+        }