blob: 1cb1fe6e35589ab5a0af01002d4db635ad5c7538 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
|
---
date: "2003-12-06T08:04:43Z"
title: Keith Packard and 32-bit Enlightenment
---
<p>
<a href='http://www.keithp.com/'>Keith Packard</a> explains how to make <a href='http://www.enlightenment.org/'>Enlightenment</a> work with 32-bit visuals. (linked from the <a href='http://lists.sourceforge.net/lists/listinfo/enlightenment-devel'>Enlightenment development mailing list</a>):
</p>
<blockquote cite='http://pdx.freedesktop.org/pipermail/xserver/2003-November/000302.html'>
<p>
Well, that's to be expected -- enlightenment happens to find the depth 32
visual and get excited about it. Unfortunately, that depth 32 visual has
alpha bits in the upper 8 bits that enlightenment doesn't understand.
Leaving those zero makes for some pretty transparent windows, hence you
see the background.
</p>
<p>
xcompmgr did have a bug where enlightenment set the _XSETROOT_ID property
to 0 which caused it to kinda lose when drawing the background. That was
easy enough to fix though.
</p>
<p>
I have a hacked up xlib which discards depth 32 visuals; using that makes
enlightenment work reasonably well. Of course, the right fix is to make
it understand those visuals and "do the right thing". It shouldn't take
very much work to make Eterm really be translucent, and to fix the
thumbnail window. Sample code for another thumbnail viewer can be found
in /cvs/xapps/uncover. That crashes the server when E is running
right now though. Time to go debug some more :-)
</p>
</blockquote>
<p>
<a href='http://pdx.freedesktop.org/pipermail/xserver/2003-November/000302.html'>Here's</a> the full message.
</p>
|