Bill Currie
9a01e71f67
[win] Silence some debug
...
Missed that one.
2023-11-26 01:16:16 +09:00
Bill Currie
d2af58bfa2
[win] Implement mouse grabbing
...
This seems be behave quite nicely (at least for the standard win32
message pump).
Fixes #63
2023-11-25 22:33:57 +09:00
Bill Currie
7024e4efb1
[win] Hide the mouse in a system-friendly way
...
ShowCursor is system-wide and thus not suitable for general use. It
turns out hiding the cursor in windows is the same as x11: use a null
cursor.
2023-11-25 22:03:27 +09:00
Bill Currie
0774f70ca2
[win] Remove the close window confirmation dialog
...
It's annoying and unless there's data to save (not at this stage, at
least), rather unnecessary.
2023-11-25 22:01:03 +09:00
Bill Currie
c110119196
[win] Track mouse leaving the window
...
This fixes the incorrect mouse deltas experienced when looping the mouse
cursor through and around the window.
2023-11-25 21:05:37 +09:00
Bill Currie
30489e1812
[win] Set window style correctly
...
This fixes the incorrectly sized window (I think).
2023-11-25 21:03:51 +09:00
Bill Currie
ed3afeed39
[win] Convert extended scan codes to QF keys
...
This takes care of the separate arrow keys not working. Fixes #62 .
2023-11-25 18:33:38 +09:00
Bill Currie
9c3f0bc5d4
[build] Fix some release script issues
...
Set sharepath correctly and the menu is not gzipped any more.
2023-11-25 15:45:25 +09:00
Bill Currie
d49c6b0c27
[win] Improve mouse event handling
...
A lot is broken, especially direct input, but things are working. Better
yet, it seems the X11 and Windows key bindings are at least mostly
compatible.
2023-11-25 15:45:25 +09:00
Bill Currie
866b13b52c
[win] Use Windows char events for keys
...
Fixes #59
2023-11-25 15:44:55 +09:00
Bill Currie
78113919ea
[win] Avoid double destroy of window
...
I suspect it's from the recent event rework, but DestroyWindow seems to
be getting called twice, despite the null protection. Not good.
2023-11-25 10:57:25 +09:00
Bill Currie
43ce0eb3ec
[sys] Correct spelling of signals
...
Is a signlas a girl with a sign?
2023-11-25 09:51:33 +09:00
Bill Currie
5f45c8a0d2
[win] Handle move and size events
...
Hook up WM_MOVE and WM_SIZE with QF's internal events.
Fixes #61
2023-11-24 22:59:48 +09:00
Bill Currie
f58a14c2d3
[build] Disable 32-bit win32 in release build script
...
32-bit Windows isn't relevant enough any more. Anybody that wants 32-bit
windows QuakeForge can go through the pain.
2023-11-24 12:29:39 +09:00
Bill Currie
250cac6079
[vid] Remove direct draw interface
...
It has been dead for a long time.
2023-11-24 12:28:55 +09:00
Bill Currie
f721021fec
[win] Clean up event handling and a pile of dead code
...
The event handling changes take care of VagueLobster's segfaults on
startup for all renderers (vulkan will still be iffy depending on his
hardware: it dies on my GTX 965 M, probably due to memory and QF's
shadows). One nice side effect is it takes care of the broken CD audio
event handling (does anyone even care, though?).
2023-11-24 12:28:06 +09:00
Bill Currie
0a89c513b4
[win] Do not call IN_ClearStates when creating the window
...
The window is created before input is initialized. Fixes a segfault on
startup of nq-win.exe.
2023-11-23 21:13:47 +09:00
Bill Currie
f3a428fbdd
Fix up windows and the release scripts
...
I really need more feedback and the best way to get that is to have more
users, and for that, I need packages.
2023-11-23 20:28:19 +09:00
Bill Currie
97a9620cb9
Do the distcheck dance
2023-11-23 17:55:33 +09:00
Bill Currie
c9cfd9e8cf
Fix window build for recent config changes
...
And, more importantly get curl working in the mxe build.
2023-11-23 17:16:53 +09:00
Bill Currie
7a2a61d365
Merge branch 'wip-trails'
2023-11-23 14:00:11 +09:00
Bill Currie
a8ff58baa4
[glsl] Remove a debug print
2023-11-23 13:58:44 +09:00
Bill Currie
43cea64709
[renderer] Get trails looking vaguely acceptable
...
Still not great, but I like the falloff and the mixing is better. The
path offset seems to be unstable, so that needs work still.
2023-11-23 13:58:44 +09:00
Bill Currie
34daf9032d
[renderer] Get trails rendering again
...
They're not quite working (trail path offset is incorrect) but their
pixels are getting to the screen. Also, lifetimes are off for rocket
trails in that as soon as the entity dies, so does the trail.
2023-11-23 13:58:44 +09:00
Bill Currie
70d07d8026
[quakefs] Document that QFS_LoadFile closes the file
...
In the end, I decided that the behavior isn't so bad, but it does need to
be documented.
Closes #34
2023-11-19 16:40:24 +09:00
Bill Currie
e8b2822587
[nq] Add more checks for dedicated server in client
...
When the nq client is run in dedicated server mode, the client state is not
initialized thus none of the client code should be run. This became more
critical with the addition of scenes and an ECS as there are a lot more
pointers involved.
Fixes #46
2023-11-19 15:03:13 +09:00
Bill Currie
49667b439d
[build] Batch-report missing packages
...
Very debian-centric and the reported packages rely on package dependencies,
but it's better than getting part-way through a build only to find
something is missing and having to restart..
2023-11-18 12:16:35 +09:00
Bill Currie
1545480ecb
[debian] Clear out a large pile of bitrot
...
This gets debian packages building again, though it was mainly a test for
some improvements to configure and missing package reporting.
2023-11-18 11:53:23 +09:00
Bill Currie
053e9f2fb4
Add more library packages for building
...
freetype, fontconfig, harfbuzz, xinput and xfixes. Also, create a shell
script that installs all the debian packages needed to build QF.
2023-11-17 20:51:12 +09:00
Bill Currie
c322fb68d3
[cexpr] Remove unnecessary config.h include
2023-11-17 20:51:12 +09:00
Bill Currie
3a38560ceb
[libs] Fix some c23 build errors for mxe
2023-10-15 18:22:06 +09:00
Bill Currie
c467d2f9b8
[vulkan] "Use" most params for QFV_duSetObjectName
...
When disabled, its params go unused and thus results in gcc unused
locals warnings.
2023-10-15 18:20:55 +09:00
Bill Currie
ef6b10afb1
[renderer] Remove dead r_part_comp.c
2023-10-03 21:27:04 +09:00
Bill Currie
a905dcd4f9
[renderer] Update particle scale
...
It seems I forgot this and didn't notice until now. The glsl renderer's
particles (style 1) look a lot better now.
2023-10-03 18:37:59 +09:00
Bill Currie
4554ea47b0
[glsl] Apply an old debug stash for trails
...
The code is mostly disabled, but this should prevent too much bitrot.
2023-10-03 15:05:13 +09:00
Bill Currie
5311f9906a
[nq,qw] Call Mod_ClearAll in CL_Shutdown
...
Although the model subsystem does this too, it does it too late relative
to the video shutdown, resulting in segfaults for glsl due to the
drivers having been unloaded.
2023-10-03 14:28:32 +09:00
Bill Currie
7f3fc6d832
[nq,qw] Call Mod_ClearAll in CL_Shutdown
...
Although the model subsystem does this too, it does it too late relative
to the video shutdown, resulting in segfaults for glsl due to the
drivers having been unloaded.
2023-10-03 14:26:54 +09:00
Bill Currie
1c6986f0ab
Merge branch 'master' into wip-trails
2023-10-03 14:11:22 +09:00
Bill Currie
c5731374cd
[glsl] Separate out the trails implementation
...
This gets things *compiling* again, though it's still non-functional and
definitely wrong (don't want trail in renderer_t), but I need to think
about the design for getting trails as components. Also need to think
about integrating trails into the client effects system so trails can be
shared between renderers.
2023-10-03 13:49:25 +09:00
Bill Currie
ba35ce71b3
[qfcc] Support 32-bit and 64-bit handle types
...
32-bit is nice because it's small, but 64-bit is handy for special
handle encodings.
2023-10-02 23:33:37 +09:00
Bill Currie
360dfc0038
[qfcc] Respect commutativity when optimizing adds
...
This fixes the really odd bug of certain string values getting swapped
in vkgen when DEBUG_QF_MEMORY was defined in expr.c. It will also
prevent a lot of fun with floats in the future, I imagine.
2023-10-02 22:33:29 +09:00
Bill Currie
0974983a39
[qfcc] Set expression associativity field
...
I forgot this when adding the option.
2023-10-02 22:32:23 +09:00
Bill Currie
7f42677a34
[qfcc] Clean up the last general uses of expr's next
...
It's now meant only for ALLOC. Interestingly, when DEBUG_QF_MEMORY is
defined in expr.c, something breaks badly with vkgen (no sniffles out of
valgrind, though), but everything is fine with it not defined. It seems
there may be some unpleasant UB going on somewhere.
2023-10-02 21:38:16 +09:00
Bill Currie
ef807f411e
[qfcc] Check alignment when looking for immediates
...
I'm not sure why this showed up now (I guess just not enough large
immediate values), but this fixes a segfault in the algtypes test (the
mystery is why it showed up this late).
2023-10-02 21:14:14 +09:00
Bill Currie
4952d298ac
[qfcc] Fix some memory errors
...
Found thanks to running valgrind over qfcc through the entire build
whith -DDEBUG_QF_MEMORY.
2023-10-02 21:14:14 +09:00
Bill Currie
bde35749f5
[util] Define ALLOC_FREE_BLOCKS for DEBUG_QF_MEMORY
...
More bitrot.
2023-10-02 20:40:21 +09:00
Bill Currie
471f5488ff
[qfcc] Catch some error expressions
...
Not really enough in general as an internal error still occurs, but the
triggering error is caused by some broken changes.
2023-10-02 14:46:42 +09:00
Bill Currie
261ea0c4de
[qfcc] Be more const-correct with expressions
...
Diagnostics that return an expression now return const, and fixes error
not returning an error expression.
2023-10-02 14:46:39 +09:00
Bill Currie
f3ca0c9222
[qfcc] Dump conflicting type expressions in debug
2023-10-02 12:06:14 +09:00
Bill Currie
b89e243c47
[qfcc] Improve handling of different types in ?:
2023-10-02 09:07:37 +09:00