[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
... is on http://www.sunsite.auc.dk/allegro/wip.html.
This consists primarily of George's patches to the Linux code. There are
only very minor tweaks to the DOS and Windows code. My apologies for not
doing anything particular useful myself over the last few weeks: I plead a
mixture of laziness, nice summer weather, and illness :-)
New things:
- lots of Linux patches from George Foot (proper fbcon mode setting,
different mouse types and setup program support for configuring them,
fixes to the shared library support, exedat no longer destroys the file
permissions, and probably some other stuff that I forgot about).
- fixed pckeys.c to leave sensible data in the key[] array when doing
numeral vs. arrow key numlock fakery.
- fixed a bunch of problems with the SWITCH_BACKGROUND mode (I'd stupidly
forgotten to disable a lot of graphics driver functions that could do bad
things if called when the Allegro console isn't active).
- added pivot_sprite() and pivot_scaled_sprite() functions (thanks to
Matt Jallo for the suggestion).
- added ustrerror() to the standard Unicode support functions.
- fixed the d_edit_proc() GUI object to handle shift+tab focus movement
properly.
George: are you getting mailing list traffic again now? I've stopped getting
bounces from you, but I'll cc: this to you anyway just in case. I made a few
changes to your patches:
- renamed your "mouse_type" config variable to "mouse", because that is
what the DOS drivers were already using.
- removed the mouse autodetect button from non-Linux versions of the
setup program.
- your change to the setup program popup routines broke the joystick test
function. I still don't understand why that code works at all, but I
modified it so that it does seem to be ok now :-)
- you had added the default fb mode timings data in allegro.cfg in front
of the keyboard and language entries, which left them in the wrong
section.
- the mouse probing code seems to mostly work, but every now and then it
thinks that my 3-button PS2 mouse is a Microsoft mouse (the PS2 probing
always works). I just swapped the order of detection for the PS2 and MS
drivers, which fixes it on my machine, but will cause problems if there
are any MS mice that the probe thinks could be a PS2 type.
- remove_mouse() was resetting the mouse_type global to MOUSE_AUTODETECT,
which overwrote whatever you had previously chosen in the setup program.
I took this out, so repeatedly removing/reinstalling the mouse will
always use the same driver it detected the first time, unless you reset
this variable in between calls (seems ok to me: there's no need to probe
for this more than once per run of a program).
--
Shawn Hargreaves - shawn@xxxxxxxxxx - http://www.talula.demon.co.uk/
"A binary is barely software: it's more like hardware on a floppy disk."