[ie/rbgtum] Fix extraction and support new URL format (#7690)

Authored by: simon300000

yt_dlp/extractor/_extractors.py

@@ -1601,6 +1601,7 @@ from .rbmaradio import RBMARadioIE
 from .rbgtum import (
     RbgTumIE,
     RbgTumCourseIE,
+    RbgTumNewCourseIE,
 )
 from .rcs import (
     RCSIE,

yt_dlp/extractor/rbgtum.py

@@ -1,10 +1,11 @@
 import re
 
 from .common import InfoExtractor
+from ..utils import parse_qs, remove_start, traverse_obj, ExtractorError
 
 
 class RbgTumIE(InfoExtractor):
-    _VALID_URL = r'https://live\.rbg\.tum\.de/w/(?P<id>.+)'
+    _VALID_URL = r'https://(?:live\.rbg\.tum\.de|tum\.live)/w/(?P<id>[^?#]+)'
     _TESTS = [{
         # Combined view
         'url': 'https://live.rbg.tum.de/w/cpp/22128',
@@ -35,16 +36,18 @@ class RbgTumIE(InfoExtractor):
             'title': 'Fachschaftsvollversammlung',
             'series': 'Fachschaftsvollversammlung Informatik',
         }
+    }, {
+        'url': 'https://tum.live/w/linalginfo/27102',
+        'only_matching': True,
     }, ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        m3u8 = self._html_search_regex(r'(https://.+?\.m3u8)', webpage, 'm3u8')
+        m3u8 = self._html_search_regex(r'"(https://[^"]+\.m3u8[^"]*)', webpage, 'm3u8')
-        lecture_title = self._html_search_regex(r'(?si)<h1.*?>(.*)</h1>', webpage, 'title')
+        lecture_title = self._html_search_regex(r'<h1[^>]*>([^<]+)</h1>', webpage, 'title', fatal=False)
-        lecture_series_title = self._html_search_regex(
+        lecture_series_title = remove_start(self._html_extract_title(webpage), 'TUM-Live | ')
-            r'(?s)<title\b[^>]*>\s*(?:TUM-Live\s\|\s?)?([^:]+):?.*?</title>', webpage, 'series')
 
         formats = self._extract_m3u8_formats(m3u8, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
 
@@ -57,9 +60,9 @@ class RbgTumIE(InfoExtractor):
 
 
 class RbgTumCourseIE(InfoExtractor):
-    _VALID_URL = r'https://live\.rbg\.tum\.de/course/(?P<id>.+)'
+    _VALID_URL = r'https://(?P<hostname>(?:live\.rbg\.tum\.de|tum\.live))/old/course/(?P<id>(?P<year>\d+)/(?P<term>\w+)/(?P<slug>[^/?#]+))'
     _TESTS = [{
-        'url': 'https://live.rbg.tum.de/course/2022/S/fpv',
+        'url': 'https://live.rbg.tum.de/old/course/2022/S/fpv',
         'info_dict': {
             'title': 'Funktionale Programmierung und Verifikation (IN0003)',
             'id': '2022/S/fpv',
@@ -69,7 +72,7 @@ class RbgTumCourseIE(InfoExtractor):
         },
         'playlist_count': 13,
     }, {
-        'url': 'https://live.rbg.tum.de/course/2022/W/set',
+        'url': 'https://live.rbg.tum.de/old/course/2022/W/set',
         'info_dict': {
             'title': 'SET FSMPIC',
             'id': '2022/W/set',
@@ -78,16 +81,62 @@ class RbgTumCourseIE(InfoExtractor):
             'noplaylist': False,
         },
         'playlist_count': 6,
+    }, {
+        'url': 'https://tum.live/old/course/2023/S/linalginfo',
+        'only_matching': True,
     }, ]
 
     def _real_extract(self, url):
-        course_id = self._match_id(url)
+        course_id, hostname, year, term, slug = self._match_valid_url(url).group('id', 'hostname', 'year', 'term', 'slug')
-        webpage = self._download_webpage(url, course_id)
+        meta = self._download_json(
+            f'https://{hostname}/api/courses/{slug}/', course_id, fatal=False,
+            query={'year': year, 'term': term}) or {}
+        lecture_series_title = meta.get('Name')
+        lectures = [self.url_result(f'https://{hostname}/w/{slug}/{stream_id}', RbgTumIE)
+                    for stream_id in traverse_obj(meta, ('Streams', ..., 'ID'))]
 
-        lecture_series_title = self._html_search_regex(r'(?si)<h1.*?>(.*)</h1>', webpage, 'title')
+        if not lectures:
+            webpage = self._download_webpage(url, course_id)
+            lecture_series_title = remove_start(self._html_extract_title(webpage), 'TUM-Live | ')
+            lectures = [self.url_result(f'https://{hostname}{lecture_path}', RbgTumIE)
+                        for lecture_path in re.findall(r'href="(/w/[^/"]+/[^/"]+)"', webpage)]
 
-        lecture_urls = []
+        return self.playlist_result(lectures, course_id, lecture_series_title)
-        for lecture_url in re.findall(r'(?i)href="/w/(.+)(?<!/cam)(?<!/pres)(?<!/chat)"', webpage):
-            lecture_urls.append(self.url_result('https://live.rbg.tum.de/w/' + lecture_url, ie=RbgTumIE.ie_key()))
 
-        return self.playlist_result(lecture_urls, course_id, lecture_series_title)
+
+class RbgTumNewCourseIE(InfoExtractor):
+    _VALID_URL = r'https://(?P<hostname>(?:live\.rbg\.tum\.de|tum\.live))/\?'
+    _TESTS = [{
+        'url': 'https://live.rbg.tum.de/?year=2022&term=S&slug=fpv&view=3',
+        'info_dict': {
+            'title': 'Funktionale Programmierung und Verifikation (IN0003)',
+            'id': '2022/S/fpv',
+        },
+        'params': {
+            'noplaylist': False,
+        },
+        'playlist_count': 13,
+    }, {
+        'url': 'https://live.rbg.tum.de/?year=2022&term=W&slug=set&view=3',
+        'info_dict': {
+            'title': 'SET FSMPIC',
+            'id': '2022/W/set',
+        },
+        'params': {
+            'noplaylist': False,
+        },
+        'playlist_count': 6,
+    }, {
+        'url': 'https://tum.live/?year=2023&term=S&slug=linalginfo&view=3',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        query = parse_qs(url)
+        errors = [key for key in ('year', 'term', 'slug') if not query.get(key)]
+        if errors:
+            raise ExtractorError(f'Input URL is missing query parameters: {", ".join(errors)}')
+        year, term, slug = query['year'][0], query['term'][0], query['slug'][0]
+        hostname = self._match_valid_url(url).group('hostname')
+
+        return self.url_result(f'https://{hostname}/old/course/{year}/{term}/{slug}', RbgTumCourseIE)