When opening a new window we ignore windows in full screen or on another space when searching for a window we can cascade to.
Full screen status of a window is restored after a relaunch and the bigger frame (when in full screen mode) is not stored in the session data or user defaults (as it should not be used).
Closes issue #667.
The problem is that during program termination all data with static storage is destructed but that may happen before the threads have finished their work.
This ensures that we restore the selection for the currently visible folder on relaunch. Selections in other folders is lost, see previous commit for more info.
This means the set of selected file browser items is reset on relaunch — I’ve found that restoring selection when navigating to a folder was often not really desired.
Previously we would reload the changed entries, if they were all visible, falling back on a full reload if not.
If items are not visible there should be no reason to do a full reload — I imagine the thinking was that the NSOutlineView may have previously fetched (and cached) one of the non-visible items (since one of its ancestors could just be collapsed). Since this is just a display attribute, it shouldn’t matter, even if the NSOutlineView would work like this.
This is required to have the object respond to various action methods such as cut:, copy:, paste:, and delete: — ideally though the class would handle it by itself, as OFB responsibility is bleeding into the document controller (though long-term I think it makes sense to instead let the document controller implement the OFB action methods).
Preview can now be initiated with multiple items and while the preview panel is showing, one can use arrow up/down to switch to other items (in the file browser).
This is a feature I have never used myself and it was somewhat limited by having the recent locations unique for each window.
If the feature should be revived it should be via a globally shared list of recent locations (and should persist across relaunch).
Presently this is non-functional but removing the OakStatusBar.h dependency makes it easier to convert to ARC, so that’ll be the next step, before actually making OFBHeaderView functional.
The problem with hg is that running ‘hg status’ may trigger fsevents, which would normally have us call ‘hg status’ again. To avoid this, we check if there actually are changes compared to the last time we ran the SCM driver. This has some overhead, as we need to scan the disk to produce the fingerprint.
Additionally the throttling has been improved. Previously we would delay a status fetch if it had been less than 3 seconds since the last request, but during this “penalty wait” more requests could still be submitted (in case of disk activity) and would be queued for later execution — now such requests are dropped.
When we quit, objects with static storage are destroyed. Since the order is undefined, we basically cannot reference other objects in destructors, unless we own them or they are reference counted.
One case that seems to be an actual problem is the oak::process_t type — based on crash logs this often seems to be destroyed last, and it unregisters itself in its destructor, but using an object that it previously did not own (but obtained via a sort of factory function).
