The shibboleth implementation no longer maintains its own QNAM.
Instead, MirallAccessManager now holds a custom QNAM implementation
which saves cookies to a file on disk.
This patch also reduces some complexity wrt the browser window,
which used to be deleted via a roundtrip to its callee, which
is not longer required.
Fixes#1764 and Enterprise bug #165
Going forward, AbstractCredentials::getQNAM() could maybe removed entirely.
bubbleUpSyncResult is only called from slotSyncFinished, so if _engine
is invalid there, it is also invalid in slotSyncFinished
This reverts commit aee7515d42.
In case two renames are done on the same file/folder very quickly we
lost the information that the second operation was also a rename. That
was because we tried to get the inode value from a stat on the file once
the first rename was finished. But at that point, the file was already
gone because of the second rename.
Now the original inode is kept and written to db in case the file can
not be stat'ed.
This fixes bug #1737
We don't include them in the total amout of files so it should
not be included in the progress either.
Also, for some reasons directories seems to be 16kB on windows
Fixup commit 9193286fc1
Use the right URL for the PROPPATCH and HEAD, do not let other
jobs start when we wait for the UpdateMTimeAndETagJob
Update the etag and the fileid of the real item.
Before, we would only set up a file system watcher when we read the
config at startup. But we also need to do it in the other case when
the user configure new folder to watch
We can't use the quota job for that as it needs the credidentials and therefore
may re-enter the credidential code when we are currently trying to fetch the credentials.
The quotainfo.cpp part of this patch is basically a revert of d836b80153
We fetch the id from the server, but don't save them in the database.
I Could have used INSTRUCTION_UPDATED for that, but then i would need to update the
reconcile algorithm to take in account the fact that UPDATED is possible there.
Instead, use should_update_etag which means the db is going to be written again
Remove reference to old instruction _UPDATED and _DELETED which does not make sens with
the new propagator
Improve the test to test this case, and that etags are properly writen to the DB
when there is a fake conflict
This can cause the crash because the _engine member is deleted in the
sync finished slot. The solution is to store the stopWatch object
before the engine is destroyed.
Fixes bug #1675
