On Thu, 13 Nov 2003 12:11:34 +0000 Gen Zhang <[EMAIL PROTECTED]>
babbled:

> > The problem i see is what Andrew E. wrote:
> > > 2) the background is not psudo transparent, it seems to pick a random
> > > bit of the background to paste behind it, then never refreshes. - either
> > > that or it is just getting random junk
> >
> > I made a simple pseudo transparent clock that uses the PT function from
> > evidence. (imlib2) http://tokyo.cored.org/code/eeclock.tar.gz
> >
> > You should look at bg.c
> The thing is, I _really_ want to keep the code clean from further deps on 
> other libraries, plus, to use imlib2 just for PT seems a little overkill. As 
> far as the updating thing goes, it's not supposed to. The idea was to set the 
> docker as always underneath everything else, and the grabs should only get 
> the desktop background, so that the "pseudo" part never becomes an issue. 
> Unfortunately, I still can't get that to work -- if anyone has ideas, help 
> would be appreciated. I used to be able to get the thing to write to the root 
> window, but that caused "interesting" problems with refresh.

well to do PT you will need to grab the root pixmap AND handle (in theory) every
depth known to man, and even ones not known. X could be in any depth (32, 16,
15, 8, 4, 2, 1bpp) and in 8bpp or less it could have pseudocolor & palettes.
it's your job to take screen pixels at screen depth and make yourself some ARGB
pixels. this is not a trivially small amount of code - ESPECIALLY if you want to
do it fast. basically it means writign a special case for every depth & RGB mask
combination that you want to speed up and then having a generic "handles all"
pair of functions - one for true/directcolor and one for
pseudocolor/staticcolor.

maybe it's worth taking some of this code out of imlib2 and putting it into a
library of its own that ONLY does "grabbing of pixels".

> Currently, I would still appreciate help on some way of capturing mouse move 
> events without having to create a "event capturing" object in evas. I think 
> that would help quite a bit with performance.

it doubt it would help at all with performance (help in anyway you could
measure). using the evas object isnt a performance hit and works well for many
things... and is portable to displaying in something other than x :)

> Cheers for the testing,
> Gen
> 
> -- 
> From 0 to "what seems to be the problem officer" in 8.3 seconds.
>               -- Ad for the new VW Corrado
> 
> genneth8691.wincoll.ac.uk
> Gentoo Linux, version 2.6.0-test8-mm1
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email sponsored by: ApacheCon 2003,
> 16-19 November in Las Vegas. Learn firsthand the latest
> developments in Apache, PHP, Perl, XML, Java, MySQL,
> WebDAV, and more! http://www.apachecon.com/
> _______________________________________________
> enlightenment-devel mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel


-- 
--------------- Codito, ergo sum - "I code, therefore I am" --------------------
The Rasterman (Carsten Haitzler)    [EMAIL PROTECTED]
熊耳 - 車君                         [EMAIL PROTECTED]
Mobile Phone: +61 (0)413 451 899    Home Phone: 02 9698 8615


-------------------------------------------------------
This SF.Net email sponsored by: ApacheCon 2003,
16-19 November in Las Vegas. Learn firsthand the latest
developments in Apache, PHP, Perl, XML, Java, MySQL,
WebDAV, and more! http://www.apachecon.com/
_______________________________________________
enlightenment-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel

Reply via email to