Abdelrazak Younes wrote: > I have gone too fast here... it is still needed apparently but it was > not not done correctly, hence the crash. I will clean this up at put the > CoordCache update at the correct place. I am going to do this on demand > instead of everytime. And the correct place is when you change screen > display because of a FitCursor.
And when a paragraph gets one more line because you entered something. Or is that handled by fitCursor? >> Do you say that fontmetrics and paragraph breaking in math is too slow? > > If you look at my test document, this is no surprise really. Considering > what we ask mathed to do, I'd say it's pretty fast. The problem is that > we updated the whole formula before at each key stroke whereas it was > not needed at all! OK, that is what I thought, too. Your original post did rather look like "mathed metrics is too slow on itself", and that would have surprised me. >> I >> would not understand that, because it is (according to Andre) basically >> unchanged from 1.3 (and there it was quick enough). > > The code that I just removed appeared in 1.4 I think. Yes. Georg
