* A repair step that inserts a background job for each user
* Each background job will delete for 15 seconds if it takes longer we
reschedule. This is done so instances that don't use the system cron
won't time out.
* Added tests
Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
Whenever a group share is created after a direct share, the stime order
needs to be properly considered in the repair routine, considering that
the direct user share is appended to the $subShares array and breaking
its order.
Pick the most recent subshare that has no parenthesis from duplication
which should match whichever name the user picked last. If all
subshares have duplicate parenthesis names, use the least recent group
share's target instead.
The repair step was a bit overeager to skip repairing so it missed the
case where a group share exists without subshares but with an
additional direct user share.
Fixes#22978
On some older installations the permissions for the userRoot and the
avatars are not correct. This breaks since we now use the Node API in
the avatar code.
This repair job makes sure that the permissions are set correctly.
* Unit tests added
* Move a-b to PSR-4
* Move c-d to PSR-4
* Move e+g to PSR-4
* Move h-l to PSR-4
* Move m-r to PSR-4
* Move s-u to PSR-4
* Move files/ to PSR-4
* Move remaining tests to PSR-4
* Remove Test\ from old autoloader