On Wed, Oct 03, 2007 at 10:02:03AM +0200, Jean-Marc Lasgouttes wrote: > Richard Heck <[EMAIL PROTECTED]> writes: > > > Andre Poenitz wrote: > >> Is there a particular reason we keep the dialogs around instead of > >> re-build them each time they are shown? > >> > >> Do they keep important state? > > Isn't there a speed argument for big things like prefs or document > settings?
I don't think so. I've been creating and destroying "big" dialogs for years now and have not had performance issues due to that. > Does it hurt to keep the dialogs? It certainly makes the code a bit more complicated. Probably not much. Btw, am I the only one finding the following behaviour "funny": On dialog apply (in the frontend) in most cass we call LFUN_INSET_APPLY in LyXFunc with the name of the dialog, the the core calls the frontend for a mapping from name to Inset *, and then is does the actual work. I wonder how _that_ is simpler than calling a core function with the Inset* directly. [Apart from that I am not even sure whether it makes sense to stor the Inset * at all. One could e.g. take some "suitable" inset close to the cursor position and get rid of that hassle of keeping the inset pointers in the frontend up-to-date.] Andre'
