Old Extractors left behind:
VLivePlaylistIE
YoutubeSearchURLIE
YoutubeShowIE
YoutubeFavouritesIE
If removing old extractors, make corresponding changes in
docs/supportedsites.md
youtube_dlc/extractor/extractors.py
Not merged:
.github/ISSUE_TEMPLATE/1_broken_site.md
.github/ISSUE_TEMPLATE/2_site_support_request.md
.github/ISSUE_TEMPLATE/3_site_feature_request.md
.github/ISSUE_TEMPLATE/4_bug_report.md
.github/ISSUE_TEMPLATE/5_feature_request.md
test/test_all_urls.py
youtube_dlc/version.py
Changelog
deobfuscates the video URL using a reverse engineered version of KVS
player's algorithm. This was tested against version 4.0.4, 5.0.1,
5.1.1.4 and 5.2.0.4 of the player and a warning will be issued if the
major version changes.
We already had a few copies of Polymer-style pagination handling logic
for certain circumstances, but now we're forced into using it for all
playlists since we can no longer disable Polymer. Refactor the logic to
move it to the parent class for all entry lists (including e.g. search
results, feeds, and list of playlists), and generify a bit to cover the
child classes' use cases.
Not my changes, but from @franhp that didn't get merged on yt-dl on time
It supports BTCC new pages' schema from 2019 an on (/articles/ instead of /races/)
live_chat_continuation['continuations'][0]['liveChatReplayContinuationData']['continuation'] can not exist.
So catch the KeyError.
Traceback:
$ tubeup 'https://youtube.com/watch?v=JyE9OF03cao'
[debug] Encodings: locale UTF-8, fs utf-8, out UTF-8, pref UTF-8
[debug] youtube-dlc version 2020.10.25
[debug] Python version 3.7.3 (CPython) - Linux-5.8.0-0.bpo.2-amd64-x86_64-with-debian-10.6
[debug] exe versions: ffmpeg 3.3.9, ffprobe 3.3.9
[debug] Proxy map: {}
There are no annotations to write.
[download] 452.59KiB at 615.35KiB/s (00:01)ERROR: 'liveChatReplayContinuationData'
Traceback (most recent call last):
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 846, in extract_info
return self.process_ie_result(ie_result, download, extra_info)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 901, in process_ie_result
return self.process_video_result(ie_result, download=download)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 1696, in process_video_result
self.process_info(new_info)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 1894, in process_info
dl(sub_filename, sub_info, subtitle=True)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 1866, in dl
return fd.download(name, info, subtitle)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/downloader/common.py", line 375, in download
return self.real_download(filename, info_dict)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/downloader/youtube_live_chat.py", line 85, in real_download
continuation_id = live_chat_continuation['continuations'][0]['liveChatReplayContinuationData']['continuation']
KeyError: 'liveChatReplayContinuationData'
This can happen if another software is using yt-dlc'API (ie: tubeup).
The stack trace would be:
$ tubeup 'https://youtube.com/watch?v=JyE9OF03cao'
[debug] Encodings: locale UTF-8, fs utf-8, out UTF-8, pref UTF-8
[debug] youtube-dlc version 2020.10.25
[debug] Python version 3.7.3 (CPython) - Linux-5.8.0-0.bpo.2-amd64-x86_64-with-debian-10.6
[debug] exe versions: ffmpeg 3.3.9, ffprobe 3.3.9
[debug] Proxy map: {}
There are no annotations to write.
ERROR: '>' not supported between instances of 'NoneType' and 'int'
Traceback (most recent call last):
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 846, in extract_info
return self.process_ie_result(ie_result, download, extra_info)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 901, in process_ie_result
return self.process_video_result(ie_result, download=download)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 1696, in process_video_result
self.process_info(new_info)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 1894, in process_info
dl(sub_filename, sub_info, subtitle=True)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/YoutubeDL.py", line 1866, in dl
return fd.download(name, info, subtitle)
File "/mnt/data2/Backup/Wiki/.local/lib/python3.7/site-packages/youtube_dlc/downloader/common.py", line 367, in download
if self.params.get('sleep_interval_subtitles') > 0:
TypeError: '>' not supported between instances of 'NoneType' and 'int'
This fix had been proposed on yt-dl for a lengthy period of time but was never merged. It has been thoroughly tested but a large section of the community.
In the event when there are no available sources due to DRM controlled sources, return a DRM error and don't proceed with trying.
#28 reports that an ExtractorError "No video formats found". Which is true, because the formats list is empty, however it's empty because they are all locked. This provides a more informative message to the end-user.
# TESTING
Tried the URL provided in #28 and confirmed a DRM messages is returned.
Fix a problem introduced in 320724f964 where is extracted the ID from the url with self._match_id but the problem is that ID is not always present in the url passed so the title should be extracted as proposed by the fix (and like is done in _real_extract (see line 337))
