Merge 3e9b4fdb60 into f919729538

yt_dlp/extractor/frontendmasters.py

@@ -16,11 +16,20 @@ class FrontendMastersBaseIE(InfoExtractor):
 
     _NETRC_MACHINE = 'frontendmasters'
 
-    _QUALITIES = {
+    def _get_subtitles(self, lesson_data, course_data):
-        'low': {'width': 480, 'height': 360},
+        captions_base = "https://captions.frontendmasters.com/assets/courses/"
-        'mid': {'width': 1280, 'height': 720},
+        lesson_slug = lesson_data.get('slug')
-        'high': {'width': 1920, 'height': 1080},
+        lesson_index = lesson_data.get('index')
-    }
+        date_published = course_data.get('datePublished')
+        course_slug = course_data.get('slug')
+
+        subtitles_url = f'{captions_base}{date_published}-{course_slug}/{lesson_index}-{lesson_slug}.vtt'
+
+        return {
+            'en': [{
+                'url': subtitles_url
+            }]
+        }
 
     def _perform_login(self, username, password):
         login_page = self._download_webpage(
@@ -72,7 +81,7 @@ def _extract_chapters(course):
         return chapters
 
     @staticmethod
-    def _extract_lesson(chapters, lesson_id, lesson):
+    def _extract_lesson(chapters, lesson_id, lesson, subtitles):
         title = lesson.get('title') or lesson_id
         display_id = lesson.get('slug')
         description = lesson.get('description')
@@ -109,6 +118,7 @@ def _extract_lesson(chapters, lesson_id, lesson):
             'duration': duration,
             'chapter': chapter,
             'chapter_number': chapter_number,
+            'subtitles': subtitles
         }
 
 
@@ -132,43 +142,28 @@ def _real_extract(self, url):
         lesson_id = self._match_id(url)
 
         source_url = f'{self._API_BASE}/video/{lesson_id}/source'
-
+        headers = {
-        formats = []
+            'Referer': 'https://frontendmasters.com/',
-        for ext in ('webm', 'mp4'):
-            for quality in ('low', 'mid', 'high'):
-                resolution = self._QUALITIES[quality].copy()
-                format_id = f'{ext}-{quality}'
-                format_url = self._download_json(
-                    source_url, lesson_id,
-                    f'Downloading {format_id} source JSON', query={
-                        'f': ext,
-                        'r': resolution['height'],
-                    }, headers={
-                        'Referer': url,
-                    }, fatal=False)['url']
-
-                if not format_url:
-                    continue
-
-                f = resolution.copy()
-                f.update({
-                    'url': format_url,
-                    'ext': ext,
-                    'format_id': format_id,
-                })
-                formats.append(f)
-
-        subtitles = {
-            'en': [{
-                'url': f'{self._API_BASE}/transcripts/{lesson_id}.vtt',
-            }],
         }
+        cookies = self._get_cookies("https://frontendmasters.com/")
+        fem_auth_mod = cookies.get('fem_auth_mod')
+        if fem_auth_mod:
+            headers['Cookie'] = f'fem_auth_mod={fem_auth_mod.value}'
+
+        json_response = self._download_json(
+            source_url,
+            'Downloading source JSON', query={
+                'f': 'm3u8'
+            }, headers=headers)
+
+        m3u8_url = json_response.get('url')
+
+        formats = self._extract_m3u8_formats(m3u8_url, lesson_id)
 
         return {
             'id': lesson_id,
             'title': lesson_id,
             'formats': formats,
-            'subtitles': subtitles,
         }
 
 
@@ -203,8 +198,9 @@ def _real_extract(self, url):
             for video_id, data in course['lessonData'].items()
             if data.get('slug') == lesson_name)
 
+        subtitles = self.extract_subtitles(lesson, course)
         chapters = self._extract_chapters(course)
-        return self._extract_lesson(chapters, lesson_id, lesson)
+        return self._extract_lesson(chapters, lesson_id, lesson, subtitles)
 
 
 class FrontendMastersCourseIE(FrontendMastersPageBaseIE):
@@ -239,9 +235,10 @@ def _real_extract(self, url):
         for lesson in lessons:
             lesson_name = lesson.get('slug')
             lesson_id = lesson.get('hash') or lesson.get('statsId')
+            subtitles = self.extract_subtitles(lesson, course)
             if not lesson_id or not lesson_name:
                 continue
-            entries.append(self._extract_lesson(chapters, lesson_id, lesson))
+            entries.append(self._extract_lesson(chapters, lesson_id, lesson, subtitles))
 
         title = course.get('title')
         description = course.get('description')