Re: [g-a-devel] other question about the gnome-mag design

[Bill Haneman <Bill Haneman Sun COM>]

Carlos Eduardo Rodrigues Diogenes wrote:

> Hi Bill,
>
> After tinking a little with the gnome-mag code other question about 
> it's design come in my mind.
>
> The question is: It isn't better to consider all the screen like 
> source_bounds? 

I think that what you mean is (please correct me if I am wrong), 
"consider the entire source display to constitute the source bounds".  
This does make sense when we use COMPOSITE, but it doesn't work today, 
without COMPOSITE, when source and target are the same display.  If 
source and target are different displays, gnome-mag already does what 
you suggest.

The reason we do not consider the entire source display to be the source 
bounds is because of two things: one, when we did this before, users 
complained that the magnifier was "magnifying itself".  Secondly, the 
source bounds need to be returned to clients, like gnopernicus, as the 
"usable source bounds", so that "proportional mode" tracking and other 
client-side algorithms can be correctly implemented.

One possibility which I think would be useful for split-screen 
magnification, although it would not solve the 'proportional mode' issue 
above, would be to allow gnome-mag to move its ROI into the current 
target area, but suppress 'self-magnification' by replacing those pixels 
explicitly with gray areas, rather than clipping as we do now.  This 
would allow the source bounds to be 'non rectangular', which would be 
useful for magnifying the bottom panel, for instance, if the magnifier 
window is on the right-hand-side of the screen but does not extend all 
the way to the bottom, and would be useful for "quarter screen" 
magnification. 

However, doing this will require a sizeable source code patch, and its 
priority has not been high enough to justify this work up until now.  I 
would certainly be happy to review a patch which did the above (used 
fill-with-blank-pixels instead of clipping, to avoid self-magnification.

best regards,

Bill

> This way we can track the pointer and only magnify the regions of the 
> screen that aren't the magnifier window. Other thing that I think in 
> this way is that we can verify if the user have the COMPOSITE 
> extensions and magnify also when the pointer is over the magnifier 
> screen, and if the COMPOSITE extension is not present we show an gray 
> color like is done today when we run only the magnifier and the 
> pointer are over the magnifier screen and in the corners of the screen.
>
> Thanks,
> Carlos.
> _______________________________________________
> Gnome-accessibility-devel mailing list
> Gnome-accessibility-devel gnome org
> http://mail.gnome.org/mailman/listinfo/gnome-accessibility-devel