Re: GtkTree*

[Tim Janik <timj gtk org>]

On 17 Aug 2001, Jonathan Blandford wrote:

> Tim Janik <timj gtk org> writes:

> > > A better idea is to add incremental reflow.  *Sigh*.  That should go in
> > > bugzilla somewhere.
> > 
> > hm, i'm not sure what you mean with incremental reflow, am i right
> > in assuming that you mean any signal from the model should just
> > "queue" an update in the tree view via idle handler and that
> > way multiple updates get coalesced?
> > that still can have subtle issues with thing like the selection
> > API, e.g. you'd basically need something like:
> 
> Only the size gets updated in the idle handler.  The idea is that
> generating the tree (and walking the tree) is relatively quick, but
> measuring text can be slow.  So nodes are added with a height of 0 (so
> they appear to not be there momentarily).  However, functions like
> find_node, selection, etc will still work.

ok, sounds good. except that this, compared to freeze/thaw, still
takes the hit of lots of signal emissions from the model.
but you can argue that i should simply take that as another motivation
point to finally speed up signal emissions ;)

> 
> Thanks,
> -Jonathan
> 

---
ciaoTJ