[ie/tvnoe] Rework Extractor

yt_dlp/extractor/tvnoe.py

@@ -1,46 +1,80 @@
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     clean_html,
-    get_element_by_class,
+    extract_attributes,
     js_to_json,
+    mimetype2ext,
+    unified_strdate,
+    urljoin,
 )
+from ..utils.traversal import find_element, traverse_obj
 
 
 class TVNoeIE(InfoExtractor):
-    _WORKING = False
+    IE_NAME = 'tvnoe'
-    _VALID_URL = r'https?://(?:www\.)?tvnoe\.cz/video/(?P<id>[0-9]+)'
+    IE_DESC = 'Televize Noe'
-    _TEST = {
+
-        'url': 'http://www.tvnoe.cz/video/10362',
+    _VALID_URL = r'https?://(?:www\.)?tvnoe\.cz/porad/(?P<id>[\w-]+)'
-        'md5': 'aee983f279aab96ec45ab6e2abb3c2ca',
+    _TESTS = [{
+        'url': 'https://www.tvnoe.cz/porad/43216-outdoor-films-s-mudr-tomasem-kempnym-pomahat-potrebnym-nejen-u-nas',
         'info_dict': {
-            'id': '10362',
+            'id': '43216-outdoor-films-s-mudr-tomasem-kempnym-pomahat-potrebnym-nejen-u-nas',
             'ext': 'mp4',
-            'series': 'Noční univerzita',
+            'title': 'Pomáhat potřebným nejen u nás',
-            'title': 'prof. Tomáš Halík, Th.D. - Návrat náboženství a střet civilizací',
+            'description': 'md5:78b538ee32f7e881ec23b9c278a0ff3a',
-            'description': 'md5:f337bae384e1a531a52c55ebc50fff41',
+            'release_date': '20250531',
+            'series': 'Outdoor Films s MUDr. Tomášem Kempným',
+            'thumbnail': r're:https?://www\.tvnoe\.cz/.+\.jpg',
         },
-    }
+    }, {
+        'url': 'https://www.tvnoe.cz/porad/43205-zamysleni-tomase-halika-7-nedele-velikonocni',
+        'info_dict': {
+            'id': '43205-zamysleni-tomase-halika-7-nedele-velikonocni',
+            'ext': 'mp4',
+            'title': '7. neděle velikonoční',
+            'description': 'md5:6bb9908efc59abe60e1c8c7c0e9bb6cd',
+            'release_date': '20250531',
+            'series': 'Zamyšlení Tomáše Halíka',
+            'thumbnail': r're:https?://www\.tvnoe\.cz/.+\.jpg',
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
+        player = self._search_json(
+            r'var\s+INIT_PLAYER\s*=\s*', webpage, 'init player',
+            video_id, transform_source=js_to_json, fatal=True)
 
-        iframe_url = self._search_regex(
+        formats = []
-            r'<iframe[^>]+src="([^"]+)"', webpage, 'iframe URL')
+        for source in traverse_obj(player, ('tracks', ..., lambda _, v: v['src'])):
+            src_url = source['src']
+            ext = mimetype2ext(source.get('type'))
+            if ext == 'm3u8':
+                fmts = self._extract_m3u8_formats(
+                    src_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
+            elif ext == 'mpd':
+                fmts = self._extract_mpd_formats(
+                    src_url, video_id, mpd_id='dash', fatal=False)
+            else:
+                self.report_warning(f'Unsupported stream type: {ext}')
+                continue
+            formats.extend(fmts)
 
-        ifs_page = self._download_webpage(iframe_url, video_id)
+        return {
-        jwplayer_data = self._find_jwplayer_data(
-            ifs_page, video_id, transform_source=js_to_json)
-        info_dict = self._parse_jwplayer_data(
-            jwplayer_data, video_id, require_title=False, base_url=iframe_url)
-
-        info_dict.update({
             'id': video_id,
-            'title': clean_html(get_element_by_class(
+            'description': clean_html(self._search_regex(
-                'field-name-field-podnazev', webpage)),
+                r'<p\s+class="">(.+)</p>', webpage, 'description', default='')),
-            'description': clean_html(get_element_by_class(
+            'formats': formats,
-                'field-name-body', webpage)),
+            **traverse_obj(webpage, {
-            'series': clean_html(get_element_by_class('title', webpage)),
+                'title': ({find_element(tag='h2')}, {clean_html}),
-        })
+                'release_date': (
-
+                    {re.compile(r'\d{1,2}\.\d{1,2}\.\d{4}').findall}, filter, ..., {unified_strdate}, any),
-        return info_dict
+                'series': ({find_element(tag='h1')}, {clean_html}),
+                'thumbnail': (
+                    {find_element(id='player-live', html=True)}, {extract_attributes},
+                    'poster', {urljoin('https://www.tvnoe.cz/')}),
+            }),
+        }