On Tue, Feb 03, 2004 at 10:31:08AM +0100, Alfredo Braunstein wrote: > Andre Poenitz wrote: > > > On Tue, Feb 03, 2004 at 10:16:37AM +0100, Alfredo Braunstein wrote: > >> Andre Poenitz wrote: > >> > >> > That's the goal (and that's why I want to drop that current 'math undo > >> > works only on the containing text paragraph and makes the cursor always > >> > leave the inset - which is a pain to use). > >> > >> You could also store an LCursor with every undo instance > > > > Not really because of the inset_ 'cache' field in CursorSlice. When I am > > undoing a 'remove', there's no guarantee that the re-created insets get > > the same adress. > > But it's only a cache... the information is there to rebuild the correct > insets pointers, right?
Yes. But I am to lazy to code up that information re-build [which might not be difficult after all, but...] Andre'
