Re: Handling markDirty

2003-03-19 Thread Jean-Marc Lasgouttes
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> On Tue, Mar 18, 2003 at 07:51:23PM +0100, Andre Poenitz wrote: >> In that case a flag in the FuncRequest is in order I think... John> Actually we already have ReadOnly|NoBuffer, so I think I can do John> it anyway. What do you mean? Th

Re: Handling markDirty

2003-03-19 Thread Andre Poenitz
On Tue, Mar 18, 2003 at 07:52:27PM +0100, Lars Gullik Bjønnes wrote: > What dirty state are you tinking of André? > > Afaik what John is talking abut is that the buffer changed, not that > something needs a redraw. I understood that now. I was thinking of the 'NEED_VERY_LITTLE_REFRESH', 'NEED_MO

Re: Handling markDirty

2003-03-18 Thread Lars Gullik Bjønnes
Andre Poenitz <[EMAIL PROTECTED]> writes: | On Tue, Mar 18, 2003 at 06:18:09PM +, John Levon wrote: | > Can't we just add a flag in LyXAction to each dirtying lfun, and handle | > this in dispatch ? | | As almost every lfun has the potential of changing something (even cursor | movement can d

Re: Handling markDirty

2003-03-18 Thread John Levon
On Tue, Mar 18, 2003 at 07:51:23PM +0100, Andre Poenitz wrote: > In that case a flag in the FuncRequest is in order I think... Actually we already have ReadOnly|NoBuffer, so I think I can do it anyway. > [In the long run we'd need a more generic concept to return values from > dispatch()...] R

Re: Handling markDirty

2003-03-18 Thread Andre Poenitz
On Tue, Mar 18, 2003 at 06:46:04PM +, John Levon wrote: > I'm talking about the buffer's dirtiness not redraws. Oops. Sorry. In that case a flag in the FuncRequest is in order I think... [In the long run we'd need a more generic concept to return values from dispatch()...] Andre' -- Thos

Re: Handling markDirty

2003-03-18 Thread John Levon
On Tue, Mar 18, 2003 at 07:32:24PM +0100, Andre Poenitz wrote: > As almost every lfun has the potential of changing something (even cursor > movement can do) I'd simply mark eveything as dirty after each dispatch. > Pretty robust approach ;-) I'm talking about the buffer's dirtiness not redraws.

Re: Handling markDirty

2003-03-18 Thread Andre Poenitz
On Tue, Mar 18, 2003 at 06:18:09PM +, John Levon wrote: > Can't we just add a flag in LyXAction to each dirtying lfun, and handle > this in dispatch ? As almost every lfun has the potential of changing something (even cursor movement can do) I'd simply mark eveything as dirty after each dispat

Handling markDirty

2003-03-18 Thread John Levon
Can't we just add a flag in LyXAction to each dirtying lfun, and handle this in dispatch ? We don't handle the dirty state cleverly in all sorts of circumstances anyway already. If we end up /wanting/ to do that for some cases, it would be relatively easy to do (limited_stack<2> for the dirty, an