[ie/lrt] Fix extractors (#14193)

Closes #13501
Authored by: seproDev

yt_dlp/extractor/lrt.py

@@ -1,22 +1,14 @@
 from .common import InfoExtractor
 from ..utils import (
     clean_html,
-    merge_dicts,
-    traverse_obj,
     unified_timestamp,
     url_or_none,
     urljoin,
 )
+from ..utils.traversal import traverse_obj
 
 
-class LRTBaseIE(InfoExtractor):
+class LRTStreamIE(InfoExtractor):
-    def _extract_js_var(self, webpage, var_name, default=None):
-        return self._search_regex(
-            fr'{var_name}\s*=\s*(["\'])((?:(?!\1).)+)\1',
-            webpage, var_name.replace('_', ' '), default, group=2)
-
-
-class LRTStreamIE(LRTBaseIE):
     _VALID_URL = r'https?://(?:www\.)?lrt\.lt/mediateka/tiesiogiai/(?P<id>[\w-]+)'
     _TESTS = [{
         'url': 'https://www.lrt.lt/mediateka/tiesiogiai/lrt-opus',
@@ -31,86 +23,110 @@ class LRTStreamIE(LRTBaseIE):
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-        streams_data = self._download_json(self._extract_js_var(webpage, 'tokenURL'), video_id)
+
+        # TODO: Use _search_nextjs_v13_data once fixed
+        get_stream_url = self._search_regex(
+            r'\\"get_streams_url\\":\\"([^"]+)\\"', webpage, 'stream URL')
+        streams_data = self._download_json(get_stream_url, video_id)
 
         formats, subtitles = [], {}
         for stream_url in traverse_obj(streams_data, (
-                'response', 'data', lambda k, _: k.startswith('content')), expected_type=url_or_none):
+                'response', 'data', lambda k, _: k.startswith('content'), {url_or_none})):
-            fmts, subs = self._extract_m3u8_formats_and_subtitles(stream_url, video_id, 'mp4', m3u8_id='hls', live=True)
+            fmts, subs = self._extract_m3u8_formats_and_subtitles(
+                stream_url, video_id, 'mp4', m3u8_id='hls', live=True)
             formats.extend(fmts)
             subtitles = self._merge_subtitles(subtitles, subs)
 
-        stream_title = self._extract_js_var(webpage, 'video_title', 'LRT')
         return {
             'id': video_id,
             'formats': formats,
             'subtitles': subtitles,
             'is_live': True,
-            'title': f'{self._og_search_title(webpage)} - {stream_title}',
+            'title': self._og_search_title(webpage),
         }
 
 
-class LRTVODIE(LRTBaseIE):
+class LRTVODIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?lrt\.lt(?P<path>/mediateka/irasas/(?P<id>[0-9]+))'
+    _VALID_URL = [
+        r'https?://(?:(?:www|archyvai)\.)?lrt\.lt/mediateka/irasas/(?P<id>[0-9]+)',
+        r'https?://(?:(?:www|archyvai)\.)?lrt\.lt/mediateka/video/[^?#]+\?(?:[^#]*&)?episode=(?P<id>[0-9]+)',
+    ]
     _TESTS = [{
         # m3u8 download
         'url': 'https://www.lrt.lt/mediateka/irasas/2000127261/greita-ir-gardu-sicilijos-ikvepta-klasikiniu-makaronu-su-baklazanais-vakariene',
         'info_dict': {
             'id': '2000127261',
             'ext': 'mp4',
-            'title': 'Greita ir gardu: Sicilijos įkvėpta klasikinių makaronų su baklažanais vakarienė',
+            'title': 'Nustebinkite svečius klasikiniu makaronų su baklažanais receptu',
             'description': 'md5:ad7d985f51b0dc1489ba2d76d7ed47fa',
-            'duration': 3035,
+            'timestamp': 1604086200,
-            'timestamp': 1604079000,
             'upload_date': '20201030',
             'tags': ['LRT TELEVIZIJA', 'Beatos virtuvė', 'Beata Nicholson', 'Makaronai', 'Baklažanai', 'Vakarienė', 'Receptas'],
             'thumbnail': 'https://www.lrt.lt/img/2020/10/30/764041-126478-1287x836.jpg',
+            'channel': 'Beatos virtuvė',
         },
     }, {
-        # direct mp3 download
+        # audio download
-        'url': 'http://www.lrt.lt/mediateka/irasas/1013074524/',
+        'url': 'https://www.lrt.lt/mediateka/irasas/1013074524/kita-tema',
-        'md5': '389da8ca3cad0f51d12bed0c844f6a0a',
+        'md5': 'fc982f10274929c66fdff65f75615cb0',
         'info_dict': {
             'id': '1013074524',
-            'ext': 'mp3',
+            'ext': 'mp4',
-            'title': 'Kita tema 2016-09-05 15:05',
+            'title': 'Kita tema',
             'description': 'md5:1b295a8fc7219ed0d543fc228c931fb5',
-            'duration': 3008,
+            'channel': 'Kita tema',
-            'view_count': int,
+            'timestamp': 1473087900,
-            'like_count': int,
+            'upload_date': '20160905',
+        },
+    }, {
+        'url': 'https://www.lrt.lt/mediateka/video/auksinis-protas-vasara?episode=2000420320&season=%2Fmediateka%2Fvideo%2Fauksinis-protas-vasara%2F2025',
+        'info_dict': {
+            'id': '2000420320',
+            'ext': 'mp4',
+            'title': 'Kuris senovės romėnų poetas aprašė Narcizo mitą?',
+            'description': 'Intelektinė viktorina. Ved. Arūnas Valinskas ir Andrius Tapinas.',
+            'channel': 'Auksinis protas. Vasara',
+            'thumbnail': 'https://www.lrt.lt/img/2025/06/09/2094343-987905-1287x836.jpg',
+            'tags': ['LRT TELEVIZIJA', 'Auksinis protas'],
+            'timestamp': 1749851040,
+            'upload_date': '20250613',
         },
+    }, {
+        'url': 'https://archyvai.lrt.lt/mediateka/video/ziniu-riteriai-ir-damos?episode=49685&season=%2Fmediateka%2Fvideo%2Fziniu-riteriai-ir-damos%2F2013',
+        'only_matching': True,
+    }, {
+        'url': 'https://archyvai.lrt.lt/mediateka/irasas/2000077058/panorama-1989-baltijos-kelias',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        path, video_id = self._match_valid_url(url).group('path', 'id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        media_url = self._extract_js_var(webpage, 'main_url', path)
+        # TODO: Use _search_nextjs_v13_data once fixed
-        media = self._download_json(self._extract_js_var(
+        canonical_url = (
-            webpage, 'media_info_url',
+            self._search_regex(r'\\"(?:article|data)\\":{[^}]*\\"url\\":\\"(/[^"]+)\\"', webpage, 'content URL', fatal=False)
-            'https://www.lrt.lt/servisai/stream_url/vod/media_info/'),
+            or self._search_regex(r'<link\s+rel="canonical"\s*href="(/[^"]+)"', webpage, 'canonical URL'))
-            video_id, query={'url': media_url})
+
+        media = self._download_json(
+            'https://www.lrt.lt/servisai/stream_url/vod/media_info/',
+            video_id, query={'url': canonical_url})
         jw_data = self._parse_jwplayer_data(
             media['playlist_item'], video_id, base_url=url)
 
-        json_ld_data = self._search_json_ld(webpage, video_id)
+        return {
-
+            **jw_data,
-        tags = []
+            **traverse_obj(media, {
-        for tag in (media.get('tags') or []):
+                'id': ('id', {str}),
-            tag_name = tag.get('name')
+                'title': ('title', {str}),
-            if not tag_name:
+                'description': ('content', {clean_html}),
-                continue
+                'timestamp': ('date', {lambda x: x.replace('.', '/')}, {unified_timestamp}),
-            tags.append(tag_name)
+                'tags': ('tags', ..., 'name', {str}),
-
+            }),
-        clean_info = {
-            'description': clean_html(media.get('content')),
-            'tags': tags,
         }
 
-        return merge_dicts(clean_info, jw_data, json_ld_data)
-
 
-class LRTRadioIE(LRTBaseIE):
+class LRTRadioIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?lrt\.lt/radioteka/irasas/(?P<id>\d+)/(?P<path>[^?#/]+)'
     _TESTS = [{
         # m3u8 download