On Thu, Jun 29, 2006 at 10:05:50PM +0300, Martin Vermeer wrote: > On Thu, 29 Jun 2006 13:54:09 +0200 Jean-Marc Lasgouttes <[EMAIL PROTECTED]> > wrote: > > > >>>>> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: > > > > > I suspect you are now referring to a different bug. I, to the > > > note-within-inset extends-past-cell-right-edge one. > > > > Actually, I believe it is the same bug: > > wide-collapsable-inset-always-uses-the-whole-screen-enven-when-it-does-n > > ot-need-to. > > (and it depends whether the cursor is in the inset; that's why I tried > > to restrict Wide() setting). > > I see. I believe you are on an impossible mission. And I disagree: your "bug" > isn't fixable within the context of the speedup patch.
To be more precise: the wide collapsable inset uses the "whole screen" (i.e., the maximum space made available to it by its surroundings) in order to occupy a _fixed width_ (i.e., a width invariant for the inset's content). It is this that makes "parsimonious" updating of only single rows (within the inset) possible without messing up frame or background colour rectangle. And _of course_ you see this only when the cursor is in the inset: it is the typing of the inset's content that we want to speed up in this way! - Martin
pgpeNxsJOr83N.pgp
Description: PGP signature
