| Age | Commit message (Collapse) | Author |
|---|
|
|
|
playlists (closes #27833)
|
|
|
|
|
|
|
|
|
|
|
|
Co-authored-by: remitamine <remitamine@gmail.com>
|
|
|
|
multiple + operators (closes #27803)
|
|
Mixcloud IDs are generated as `username_slug` when the full ID dict has been
downloaded. When downloading a list (e.g. uploads, favorites, ...), the temporary
ID is just the `slug`. This made e.g. archive file usage require the download
of stream metadata before the download can be rejected as already downloaded.
This commit attempts to get the uploader username during the GraphQL query, so the
temporary IDs are generated similarly.
|
|
|
|
|
|
Closes #26963.
|
|
#27588)(closes #27604)
|
|
|
|
|
|
|
|
JSON objects with an empty "files" tag seem to be a valid PeerTube API
response. In those cases the "files" arrays contained in the
"streamingPlaylists" members can be used instead.
closes #26002
closes #27586
|
|
|
|
|
|
|
|
closes #27698
|
|
|
|
|
|
|
|
|
|
#17650)(closes #27629)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Co-authored-by: Sergey M. <dstftw@gmail.com>
|
|
|
|
|
|
#27450)
|
|
Less than a week old videos use a '20h ago' or '1d ago' format.
I kept the support for 'Ago' with uppercase start at is was already in the code.
|
|
Tests truncate the download to 10241 bytes, which is not divisible by 16
and cannot be decrypted. Tests don't really care about the decrypted
content, just that the data they retrieved is the expected data.
Therefore, it's fine to just return the encrypted data to tests.
See: #27621 and #27620
|
|
the description has no HTML tags now.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
