Browse Source

Improve chapter sanitization

pukkandan 2 years ago
parent
commit
a3976e0760
2 changed files with 10 additions and 7 deletions
  1. 7 2
      yt_dlp/YoutubeDL.py
  2. 3 5
      yt_dlp/extractor/youtube.py

+ 7 - 2
yt_dlp/YoutubeDL.py

@@ -2377,13 +2377,18 @@ class YoutubeDL:
             self.report_warning('"duration" field is negative, there is an error in extractor')
 
         chapters = info_dict.get('chapters') or []
+        if chapters and chapters[0].get('start_time'):
+            chapters.insert(0, {'start_time': 0})
+
         dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
-        for prev, current, next_ in zip(
-                (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)):
+        for idx, (prev, current, next_) in enumerate(zip(
+                (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
             if current.get('start_time') is None:
                 current['start_time'] = prev.get('end_time')
             if not current.get('end_time'):
                 current['end_time'] = next_.get('start_time')
+            if not current.get('title'):
+                current['title'] = f'<Untitled Chapter {idx}>'
 
         if 'playlist' not in info_dict:
             # It isn't part of a playlist

+ 3 - 5
yt_dlp/extractor/youtube.py

@@ -2764,17 +2764,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         if not strict:
             chapter_list.sort(key=lambda c: c['start_time'] or 0)
 
-        chapters = [{'start_time': 0, 'title': '<Untitled>'}]
+        chapters = [{'start_time': 0}]
         for idx, chapter in enumerate(chapter_list):
-            if chapter['start_time'] is None or not chapter['title']:
+            if chapter['start_time'] is None:
                 self.report_warning(f'Incomplete chapter {idx}')
             elif chapters[-1]['start_time'] <= chapter['start_time'] <= duration:
-                chapters[-1]['end_time'] = chapter['start_time']
                 chapters.append(chapter)
             else:
                 self.report_warning(f'Invalid start time for chapter "{chapter["title"]}"')
-        chapters[-1]['end_time'] = duration
-        return chapters if len(chapters) > 1 and chapters[1]['start_time'] else chapters[1:]
+        return chapters[1:]
 
     def _extract_comment(self, comment_renderer, parent=None):
         comment_id = comment_renderer.get('commentId')