Re: [Gnash-commit] gnash ChangeLog server/edit_text_character.cpp ...

[strk]

On Fri, Jul 20, 2007 at 06:47:23PM +0200, Udo Giacomozzi wrote:
> Hello Udo,
> 
> Friday, July 20, 2007, 6:42:36 PM, I wrote:
> UG> +  cxform cf = character::get_world_cxform();
> 
> Is this the correct way? Guess not. Anyway, the ancestor should be
> called...

This is fine, I just don't like the whole thing :)
Trying to emulate the limitation by introducing such hacks in
primitive functions has to be discouraged IMHO.
I haven't seen a test showing that the r/g/b transformations should
work, btw.

> UG> +  if ( 0 /* if using a device font (PP compatibility, TODO) */ ) {
> 
> Sandro, once you add that "is device font" flag to the font definition
> (you said that in IRC), then change this so that the flag is checked
> here.

We have other problems it seems.
See http://foo.keybit.net/~strk/tmp/TextFieldColor.swf.
The color transform flattens the colors to a grayscale.


--strk;