Commit graph

5318 commits

Author SHA1 Message Date
Bill Currie
53c9732303 add PF_checkextension for later use. The pr_checkextension cvar will come
when I've had some sleep.
2001-03-07 10:22:51 +00:00
Bill Currie
e370eb3ce6 this should fix the abyss protocol compatability problems 2001-03-06 16:41:17 +00:00
Bill Currie
771648a5b4 forgot to remove the old prstring stuff 2001-03-06 16:10:28 +00:00
Bill Currie
97bf8d2c51 garbage collector implemented but untested. Also unsused because the issue is
WHEN to call it. I don't imagine it will be cheap :/
2001-03-06 05:29:46 +00:00
Bill Currie
4489319298 first step towards garbage collected dynamic strings in qc. All strings
external to the progs file are now malloced and then freed at progs reload.
All that needs to be implementd for gc to work is the scanner and deallocator.
2001-03-06 04:22:33 +00:00
Bill Currie
c4616346a1 dx_hack_hash stolen from Daniel Phillips <phillips@innominate.de> in the lkml.
This might or might not give hash tables a better spread. We'll see :)
2001-03-05 16:19:10 +00:00
Joseph Carter
4c3676b9ac Forgot to commit this removal. Mr. Menu, get the hell outta my tree! 2001-03-05 10:06:39 +00:00
Bill Currie
f40b361fb3 hash.[ch]:
slight api change: the getkey and free functions now take a user data
	parameter (which is an aditional parameter to Hash_New.
cmd.c, cvar.c, quakefs.c:
	clean up the resulting errors.
pr_edict.c:
	use hash tables for lookups of function, global and field definitions.
	should speed things up a bit, ESPECIALLY when type checking is enabled.
2001-03-05 05:11:26 +00:00
Bill Currie
aad21e3c00 put the pointer comparison back in so dynamic strings don't cause blowouts 2001-03-05 04:15:59 +00:00
Bill Currie
234dcb4cc9 better(?) string sharing. e1m3 died a sad death in nq due to running out of
strings, but that was because only the address of the string was checked for
similarity. Now strequal is used.
2001-03-05 04:08:08 +00:00
Bill Currie
ff2828d82d thought something was wrong with the previous checkin. I keep forgetting .S
files don't get dependencies in automake :(
2001-03-05 02:18:03 +00:00
Bill Currie
09b58f7e5f merge math code into utils 2001-03-05 02:06:05 +00:00
Joseph Carter
d407f5328f Since nobody cares the structure of the menu after it's rewritten, I've
replaced the menu structure with just the binds listing so whoever does
that doesn't forget a binding we never use (run key?  My forwardspeed is
600, wtf do I need a run key for?)
2001-03-04 13:55:05 +00:00
Joseph Carter
fb9f7a1ba4 Forest's COM_FileBase rewrite. Doubt this is faster, but it's sure a lot
easier on the eyes.
2001-03-04 10:09:59 +00:00
Joseph Carter
3ab3941c81 NQ's menu is gone too, menu.c is laying around till I add more info to the
TODO file, then it dies too.
2001-03-04 07:43:27 +00:00
Bill Currie
cf3629b3b3 merge in my progs work off the `taniwha' branch. This prigs us a few nice
thinks:
  o Full progs modularity
  o CSQC should now be just a matter of creating the builtin functions and
    loading the code.
  o total independence from progs globals, functions and entity field layouts
    on the conditoin that their definitions have not been stripped from the
    progs file.
  o optional (though currently forced on) type checking on access to progs
    entity fields from C
  o the progs engine is fully shared between nq and qw.
2001-03-04 06:30:30 +00:00
Bill Currie
6bb1f172d0 THOU SHALT NOT SCREW UP THY DOT-PRODUCT MACRO!!! :P 2001-03-04 06:08:28 +00:00
Bill Currie
021c1347ed remove the reliance on MAX_EDICTS 2001-03-04 05:42:26 +00:00
Bill Currie
4137fbabb5 nq seems to like the new progs setup (got to e2m2 in medium without problems) 2001-03-04 05:27:29 +00:00
Joseph Carter
b507853b18 A number of keys should bring up the menu in demo playback, they now bring
up the console.
2001-03-04 01:55:12 +00:00
Joseph Carter
e4e80437df Escape is still unconditional, but rather than being unconditionally menu
(or now unconditionally nothing since I killed the menu), it works as the
console.  In the future, when the new menu is implemented, it will fall
back to console if there's a problem with the menu (like there isn't one.)
2001-03-03 12:30:48 +00:00
Joseph Carter
d28e15d171 Software netgraph fixed. Probably not the same as it was before, someone
else is welcome to change it to their liking, optimize it, remove the box,
whatever..
2001-03-03 12:16:24 +00:00
Joseph Carter
9c39a00a54 Fixed the netgraph 2001-03-03 11:05:20 +00:00
Joseph Carter
fe26a1c565 This should not be here yet, reverting 2001-03-03 10:01:49 +00:00
Joseph Carter
c057c27d9a Menu has been killed completely from qw. It's dead. I doubt anybody will
even notice, given that QW's menu was crippled to the point that it was
completely useless anyway.
2001-03-03 09:42:35 +00:00
Bill Currie
8e52f7dc39 more nq fixes, not there yet 2001-03-03 09:09:39 +00:00
Bill Currie
baa5ce01a1 more independence cleanups and nq is closer to running 2001-03-03 08:59:55 +00:00
Bill Currie
6d6c34320f nq compiles with the new progs setup, but doesn't have a hope of running yet.
None of the globals etc structs are filled in and there will be a lot of type
errors in SVFIELD usage.
2001-03-03 08:31:58 +00:00
Joseph Carter
82522a6fd1 Added menu structure to TODO file, it's not in typical TODO format since
it's either done or it's not.  Very soone the menu code will be axed and
this will aid in replacing it.
2001-03-03 07:23:22 +00:00
Joseph Carter
168d47402a Mostly whitespace, these files are now identical and COULD be merged, if
they weren't about to get a rewrite first.
2001-03-03 06:51:18 +00:00
Chris Ison
8ad00b04c3 added sharepath search when loading server list 2001-03-03 06:01:57 +00:00
Bill Currie
26e2d545e4 macro parameter usage must have () around them :P
This seems to fix the aim bug
2001-03-03 05:34:50 +00:00
Chris Ison
050b49c479 fixing missing oss dev as shown by taniwha 2001-03-03 00:54:46 +00:00
Bill Currie
51ca0b6a41 missed this in the previous checkin 2001-03-02 23:41:50 +00:00
Bill Currie
3c17a888be more cleanups. rocket/shotgun bug not found yet 2001-03-02 23:40:34 +00:00
Ragnvald Maartmann-Moe IV
6962f954ed gl_colorlights gone, replaced by gl_lightmap_components. Valid values are 1, 3, 4. 4 (RGBA) default. Significant speedup going from RGB to RGBA on some boards.
Thanks to LordHavoc for the idea, and much of the design.
2001-03-02 23:38:40 +00:00
Ragnvald Maartmann-Moe IV
09f7689a09 Tweaked lava splash. 2001-03-02 23:32:24 +00:00
Bill Currie
26c387408c fix a silly edict print bug 2001-03-02 23:07:21 +00:00
Bill Currie
6c9de8061b fix the silly OP_STATE bug. also erradicate pr_offs.c and its contents (the
source of the bug)
2001-03-02 19:18:04 +00:00
Jeff Teunissen
ade87c72b3 Woo, one more file builds cleanly...very few actual changes needed. 2001-03-02 09:36:51 +00:00
Jeff Teunissen
14772123d3 Forge.m now builds cleanly. It probably doesn't work, I had to comment out
a lot of PostScript-related code.
2001-03-02 09:13:35 +00:00
Jeff Teunissen
1bd5d98c4d Compile fixes, we're getting a bit closer. 2001-03-02 07:01:51 +00:00
Bill Currie
ffc5838599 more debug changes. not there yet 2001-03-02 00:32:10 +00:00
Joseph Carter
c7d74c6743 Minor cleanup to download draw function. More to do, this is just a good
checkpoint before I make changes that may not work.
2001-03-02 00:29:35 +00:00
Bill Currie
be8f1cef7c just some temporary debug info 2001-03-01 23:30:55 +00:00
Bill Currie
6f590ff9d0 forgot to change ev_function to ev_func in pr_edict and pr_exec was using the
wrong time. weaponframe still not fixed, but static entities were fixed in the
previous checkin
2001-03-01 23:03:40 +00:00
Bill Currie
10e6b7af84 edict field access type checking :) Currently forced on, but it will become a
configure option in the future
2001-03-01 21:11:35 +00:00
Bill Currie
68220d1fb0 fix up packet logging for the new msg stuff 2001-03-01 20:41:20 +00:00
Bill Currie
4fcf7704d3 clean out unused fields 2001-03-01 19:05:40 +00:00
Bill Currie
ef38fd2e79 progsdef.h is now used only for the PROGHEADER_CRC def. Much of the protocol
still seems to be broken :(
2001-03-01 18:44:19 +00:00