remove includes of qdefs.h and compat.h
pr_comp.h:
merge pr_comp.h from quake and qfcc, removing the copy in qfcc
cmdlib.[ch]:
nuke the endian code.
qendian.c:
initialise the LittleLong etc pointers at compile time rather than run
time
com.c (both nq and qw):
nuke the LittleLong etc init code
everything else:
fix up after the qtypes.h cleanup
however I've removed the stupid limits that I'd accidentally left behind
in my panic to comment out the messed up code in r_part.c so software
clients now can use a setting of zero. Particles in sw clients will not
default to 2048 if you use a number less than 1 - it will now use 0.
Otherwise, I made comments to myself for when I actually fix this and
cleared up some silliness in comments I'd made. Nothing special.
Special note: To use cl_max_particles *right now* you can either set it
while playing and then restart the client (I think this should work,
it's archived. May not however) Or do this which I absolutely am
*positive* works:
clientname othercommands +set cl_max_particles number othercommands
So, again, while changing in game does not work, it at least is still
useful somewhat.
Misty-chan
I goofed up my code. cl_max_particles will not dynamically update with
this change, but ATM, this is safer until I can grok what I need to know
to fix the code I wrote.
Sorry guys,
Misty-chan
new cvar: cl_max_particles. This cvar is archived, has no lower or
upper limits (well, less than 1 is not allowed) and can be changed in
game at any time.
BUGS:
Only one so far. I can't figure out why it's doing this, but in software
clients, (well, at least X11) if you set it to 1 particle, it acts like
you set it far higher. 2 acts like you set it to zero... Or maybe
it's showing 2 and I just can't see it on my 320x200 window. In any
case, the vagary must be something in the software particles code,
because I basically used the same code from the GL particles code for
this as I used for the software renderer.
If nobody can find fault with my code, I'll just make a special note in
the console help.
In any case, let me know of any problems.
Misty-chan
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.
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.
