On Tue, Nov 18, 2003 at 10:01:35AM +0100, Alfredo Braunstein wrote: > Andre Poenitz wrote: > > > On Tue, Nov 18, 2003 at 09:19:16AM +0100, Alfredo Braunstein wrote: > >> - shouldn't these be handled by the BufferView? > > > > Maybe. I am not sure, though. > > > >> As opposed to right/left, they (should/used to) ignore the document > >> structure. > > > > Getting up/down (sort of) right was _the_ problem when glueing the > > mathed shards together. It currently is some mixture based on inset > > nesting and setting by coordinate. As the outer world is a bit simpler > > I'd expect we could get away with a pure nesting based solution. > > > >> Right now they enter in insets ignoring it, but exit insets losing the x > >> coordinate (by some dispatchresult flag I presume). > > > > The target x is BufferView::target_x_. Maybe its not updated often > > enough... > > Mmm... but we don't use x_target_ at all IIUC. We simply return with > DISPATCHED_RIGHT/LEFT when exiting and inset.
The containing inset should have a look at BufferView::target_x_ when dispatching this. > This is why I think that we should handle it directly in the BufferView. This is one option, which should work with the kind of rectangular structure we have in the outer world. It's not sufficient for math, though. But we don't care about that right now... > [I think that the problem with target_x_ is that it is updated *too much*, > and thus it is completely in-sync with the cursor (and thus useless). I'll > have a look at it.] Might be true... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)
