On 04/21/2012 05:53 AM, Till Oliver Knoll wrote: > .... As a proof of concept, a naive "user settings aware" Qt implementation > could be > > - When initialising QApplication and > - When we detect running on KDE > - Dynamically link to "KDE Core" library (if not already done so anyway) > - For each standard QKeySequence: > * Instantiate the corresponding KAction via KStandardAction > * Get the shortcut (QKeySequence) set for the KAction and initialise the > standard QKeySequence in Qt with it > > > Off course instantiating n objects just for getting a shortcut is totally > nonsense, let alone that linking against a library (KDE Core) which on its > turn depends on Qt itself is not such a brilliant idea... > > Heck, but via clever plugin mechanisms Qt also manages to call the standard > KDE file dialog today, so I'm pretty sure there is a way to query the KDE > shortcut settings, too, in some way :)
I wouldn't want to do this, even as a 'POC'. Linking Qt's keyboard shortcuts to KDE would be a fundamental error WRT the layered design and dependencies standards of both Projects. (Keyboard input is pretty fundamental.) Now, if I may return to my back to my own fundamental question: Thiago, if I submit a change for 5.0 (where we WON'T have the BC problem with adding symbols to the keyboard shortcuts enumeration), will you need to reject it as "too late for 5.0"? Or can it be accepted if the reviewers (MS-O, BH-MBP, and maybe one other) Approve the Change? _______________________________________________ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
