On Mon, Jun 19, 2006 at 04:16:20PM +0200, Jean-Marc Lasgouttes wrote: > >>>>> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: > > Martin> On Mon, 2006-06-19 at 12:52 +0200, Jean-Marc Lasgouttes wrote: > >> >>>>> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: > >> > Martin> Here's the patch you've been looking for... it simplifies > Martin> things which is always a good sign. > >> Instead of using Wide(), isn't it possible to use display()? > > Martin> No, it is a different thing. An inset is either display(), or > Martin> it is not. Not like Wide() where you poke the value from the > Martin> rowpainter. > > And what are the situations where the two informations do not match?
Your question presupposes that these two mechanisms would have anything in common. They have not. display() is a property of the inset (or in math, a mode of the inset). Wide() is a status element of text insets only that depends on its position within he surrounding row and that of the cursor relative to it. So a numerically accurate but conceptually misleading answer would be "all kinds of situations that happen all the time". > Martin> OK. I'll commit presently to trunk then. What about 1.4? > > What is the UI effect? Back to old editing mode? What do you mean? As Bennett writes, speed is OK also with this patch. The visual effect is that you have real boxes within real boxes always, also for "wide" insets. The "wideness" now means: extends to its theoretical maximum right edge, defined by the surrounding text. (This is mi.base.textwidth, becoming text.maxwidth_ inside the lyxtext of the textinset. See insetcollapsable.) - Martin
pgpoTHGfWvoCi.pgp
Description: PGP signature
