we should return false from
BulkPropagatorDownloadJob::scheduleSelfOrChild() because we have no
child jobs that are not done and waiting to be started
ensure we update the job status only once (even in case of errors)
ensure we update the state of the items once and only once
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
filter new placeholder items by invalid modtime and only proceed with
creating only the valid ones
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
use the proper VFS plug-in API to request all files to be cerated as a
single batch
will now need a proper implementation using the batch Windows CfApi
function
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
should allow to speed API calls to Windows CfApi by reducing their
numbers by asking for a list of new placeholders via a single call
instead of one for each new item
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
this code does not make sense
I tested manually and this code is never called
I doubt this code is needed or useful
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
try to use const and auto as much as possible
remove useless constructor argument that is always using the default value
makes code cleaner
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
no need to carry some hack when we can use the same standard and cross
platform feature directly from Qt code
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
originally using qFatal or qCFatal should have written a crash dump log
file
bring it back and ensure the crash would happen via the default Qt log
message handler
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
when a build is configured without the network proxy feature, ensure we
do not show any UI related to network proxy settings in the new account
wizard
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
Some callbacks do not contain a valid value for the `ProcessInfo` member
despite `CF_CONNECT_FLAG_REQUIRE_PROCESS_INFO` being set.
Signed-off-by: Jyrki Gadinger <nilsding@nilsding.org>
The `hydrationRequestReady` signal could be emitted before it was
connected to the event loop, resulting in the data transfer never being
started.
Signed-off-by: Jyrki Gadinger <nilsding@nilsding.org>
will migrate invalid ietm encrypted status in the client DB when
migrating from older releases
that should ensure working synchronization by having correct tracking
of which items are encrypted
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
will stop failing to decode using private/publy key pair when
downloading encrypted content (i.e. symetric keys are encrypted using
asymetric encryption)
we improperly use a certificate hash to find the key while the private
key is the current one of the user downloading the files
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
main issue is with child folders that were misbehaving once the initial
upload was done
Signed-off-by: Matthieu Gallien <matthieu.gallien@nextcloud.com>
