Compare commits

...

2 Commits

Author SHA1 Message Date
jhwgh1968
4b97b18050 [Substack] test multi-embed page of "wrong" type 2025-04-24 17:59:06 -05:00
jhwgh1968
e89132c678 [Substack] allow multiple embeds regardless of post type 2025-04-24 17:58:09 -05:00

View File

@ -86,6 +86,51 @@ class SubstackIE(InfoExtractor):
'duration': 25.939592,
'uploader_id': '535611',
},
}, {
# A "podcast" page which contains embedded video
'url': 'https://substack.com/home/post/p-161501224',
'info_dict': {
'id': '161501224',
'title': 'Videos for Global Medical Tyranny Just Got Real | Daily Pulse',
},
'playlist': [{
'info_dict': {
'id': '161501224',
'ext': 'mp3',
'title': 'Global Medical Tyranny Just Got Real | Daily Pulse',
'uploader_id': '975571',
'uploader': 'The Vigilant Fox',
'thumbnail': r're:https://substack.*video.*/video_upload/post/.+\.png',
'description': 'The news you werent supposed to see.',
},
}, {
'info_dict': {
'id': '95b4b5a7-3873-465d-a6f4-8d3c499f29f2',
'ext': 'mp4',
'title': 'WHO.mp4',
'uploader_id': '82027648',
'duration': 593.3333,
'thumbnail': r're:https://substack.*video.*/video_upload/post/.+\.png',
},
}, {
'info_dict': {
'id': '235db8bf-c6e8-439a-b51a-eb7566fc7ac1',
'ext': 'mp4',
'title': 'The Vigilant Fox\'s Video - Apr 16, 2025-VEED.mp4',
'uploader_id': '82027648',
'duration': 498.86667,
'thumbnail': r're:https://substack.*video.*/video_upload/post/.+\.png',
},
}, {
'info_dict': {
'id': 'ef3fe4c1-d748-4665-ba9c-33396008e75b',
'ext': 'mp4',
'title': 'The Vigilant Fox\'s Video - Apr 16, 2025-VEED (1).mp4',
'uploader_id': '82027648',
'duration': 410.53333,
'thumbnail': r're:https://substack.*video.*/video_upload/post/.+\.png',
},
}],
}]
@classmethod
@ -133,7 +178,9 @@ class SubstackIE(InfoExtractor):
canonical_url = urllib.parse.urlparse(url)._replace(netloc=domain).geturl()
post_type = webpage_info['post']['type']
post_title = traverse_obj(webpage_info, ('post', 'title'))
formats, subtitles = [], {}
all_results = []
result_info = {
'id': str(webpage_info['post']['id']),
'subtitles': subtitles,
@ -144,7 +191,9 @@ class SubstackIE(InfoExtractor):
'uploader_id': str_or_none(traverse_obj(webpage_info, ('post', 'publication_id'))),
'webpage_url': canonical_url,
}
# handle specific post types which are based on media
if post_type == 'podcast':
podcast_result = dict(result_info)
fmt = {'url': webpage_info['post']['podcast_url']}
if not determine_ext(fmt['url'], default_ext=None):
# The redirected format URL expires but the original URL doesn't,
@ -153,21 +202,21 @@ class SubstackIE(InfoExtractor):
HEADRequest(fmt['url']), display_id,
'Resolving podcast file extension',
'Podcast URL is invalid').url)
result_info['formats'] = [fmt]
return result_info
elif post_type == 'video':
podcast_result['formats'] = [fmt]
all_results.append(podcast_result)
if post_type == 'video':
video_result = dict(result_info)
formats, subtitles = self._extract_video_formats(
webpage_info['post']['videoUpload']['id'], canonical_url)
result_info.update({
video_result.update({
'formats': formats,
'subtitles': subtitles,
})
return result_info
elif post_type == 'newsletter':
results = []
all_results.append(video_result)
# search for embedded players on the page
found_items = []
post_id = str(webpage_info['post']['id'])
post_title = traverse_obj(webpage_info, ('post', 'title'))
assert result_info['uploader_id'] is not None, 'newsletter posted without user_id'
video_players = re.finditer(
@ -230,12 +279,10 @@ class SubstackIE(InfoExtractor):
'uploader_id': str_or_none(json_vid_data.get('user_id', None)),
'uploader': None, # video uploader username is not included
})
results.append(new_result)
all_results.append(new_result)
if len(results) > 0:
if len(all_results) > 0:
playlist_title = f'Videos for {post_title}'
return self.playlist_result(results, post_id, playlist_title)
return self.playlist_result(all_results, post_id, playlist_title)
else:
self.raise_no_formats(f'Page type "{post_type}" contains no supported embeds')
self.raise_no_formats(f'Page type "{post_type}" is not supported')