On Monday 11 November 2002 10:34 am, Andre Poenitz wrote: > On Mon, Nov 11, 2002 at 10:30:10AM +0000, Angus Leeming wrote: > > > Well, I still consider the preview stuff as "highly experimental", so > > > unless there are hard crashs I'd better leave it as it is... > > > > Ummm. Highly experimental usually means "has been hacked together so it > > works, sort of, but is really unmaintainable". > > No, I mean "does not work out-of-the-box and does not work properly if > tweaked". The first is a matter of fact, and the second is a conceptal > problem... > > [I recently came to the conclusion that the only way to get correct > previews is to process the full file as there is no way to properly handle > included .tex or ERT in general. So what we could do is to run latex > every few seconds in the background and update previews afterwards. > This would, of course eat much more CPU time and produces a noticeale delay > between leaving a formula and update of the preview. But I really see no > other robust solution.] > > > Is that what you feel about my coding? > > Too early in the week to say ;-) > > Seriously, it's not the codeing, it's the concept.
I understand and agree with you here. However, given that we generate /some/ preview, we should be able to display it multiple times by copying-and-pasting. The bug lies in our implementation, not in the concept. Morever, since copying-and-pasting is a generic operation, the bug is likely to be present for all insets. The difference is that it is visually apparent for only a few. Angus
