Use File Node API for more download cases, skip unreadable files#18433
Use File Node API for more download cases, skip unreadable files#18433
Conversation
|
mh, some integration tests for download throw 404s and i can reproduce locally :/ |
|
Lets restart them first |
|
Nope we'll have to dive into that.. |
|
yeah, no surprise it fails for me, too… I'll dive, either a bit later, or tomorrow morning. |
|
first arg has to be full path, not just a name. works one folder down, but not deeper structures. |
|
Passes now (apart from those MySQL timeouts) |
| * @throws InvalidPathException | ||
| */ | ||
| public function addDirRecursive($dir, $internalDir='') { | ||
| public function addDirRecursive(string $dir, string $internalDir = ''): void { |
There was a problem hiding this comment.
passing the Node/FileInfo instead of the path here would be nice as it would cut the number of filecache queries in half
There was a problem hiding this comment.
Passing the Node would be more helpful, I agree and had the same thought. I discarded it to have a simpler and smaller changeset to backport.
|
/backport to stable17 |
|
/backport to stable16 |
|
/backport to stable15 |
Signed-off-by: Julius Härtl <jus@bitgrid.net>
Signed-off-by: Arthur Schiwon <blizzz@arthur-schiwon.de>
* avoids a 403, but enables download of resources that are not restricted * single file downloads still cause 403 Signed-off-by: Arthur Schiwon <blizzz@arthur-schiwon.de>
9cd969b to
af91efd
Compare
|
backport to stable17 in #18488 |
|
backport to stable16 in #18489 |
|
The backport to stable15 failed. Please do this backport manually. |
|
15 is EOL now |
Follow up to #17822