quake-c-users

[Quake-C] Compile errors in FTE

[Cobalt <co...@te...>]

One of my side projects is getitng the Runequake src to compile with FTE, 
and its been a long hard road as it was originally using qxxc to do the 
fancy compiling stuff thats all over this mod. Fte does have a -Fqccx mode 
that I am trying as per Spike, but its already started flagging stuff in 
defs qc as follows:

string timeset = "5\{0}":"10":"15":"20":"25":"30";
string timeset_gold = "\5\{0}":"\1\0":"\1\5":"\2\0":"\2\5":"\3\0";

defs.qc:1636: warning: Missing semicolon at end of definition
defs.qc:1636: error: ":" is not a type
defs.qc:1637: warning: Missing semicolon at end of definition
defs.qc:1637: error: ":" is not a type

************ ERROR ************
Errors have occured


Last year when I got around this I changed it to:

string timeset = "5\{0}:10:15:20:25:30";
string timeset_gold = "\5\{0}:\1\0:\1\5:\2\0:\2\5:\3\0";

But Im not sure that is a legit fix?


Runequake uses a makefile / .mk4 preprocessor system, seems like an old Unix 
language, but cant make heads or tails of it except it seems to let you 
compile the src for
either Proquake (poq IN_POQ) or Quake World (qw / IN_QW). Would be great if 
we could figure out a DP mode , IE: IN_DP, and I guess thats my job . .  ..

qccx1.0 can handle those, but only if you run from the already supplied poq 
or qw directories that come when you get its src.

==
Cobalt

Re: [Quake-C] Compile errors in FTE

[<qu...@ca...>]

Wow.

Just from fixing the timeset compile issue, I find a bunch of compiler 
stuff that fte does different!


On 2016-10-11 16:06, Cobalt wrote:
> One of my side projects is getitng the Runequake src to compile with 
> FTE,
> and its been a long hard road as it was originally using qxxc to do the
> fancy compiling stuff thats all over this mod. Fte does have a -Fqccx 
> mode

Really?  I thought I was running the latest version:

] fteqcc -Fqccx
warning: Unrecognised flag parameter (-Fqccx)


> string timeset = "5\{0}":"10":"15":"20":"25":"30";
> string timeset_gold = "\5\{0}":"\1\0":"\1\5":"\2\0":"\2\5":"\3\0";
> 
> defs.qc:1636: warning: Missing semicolon at end of definition
> defs.qc:1636: error: ":" is not a type
> defs.qc:1637: warning: Missing semicolon at end of definition
> defs.qc:1637: error: ":" is not a type

> But Im not sure that is a legit fix?

This depends entirely on how the compiler was setup to handle array / 
string.
That is what it looks like anyway, as I used it before.

I was using frikqcc and that is the string array format.
I dumped that crap when I went to fte as it wasnt supported the same 
way.


I make no promises...but try

defs.qc:

string timeset[5];

world.qc:

// in worldspawn

timeset[0] = "05";
timeset[1] = "10";
timeset[2] = "15";
timeset[3] = "20";
timeset[4] = "25";
timeset[5] = "30";

This compiles and may even work as intended.  YMWV (note the W = will)

However... They use %{n} as an array reference - I believe fte does not.


> Runequake uses a makefile / .mk4 preprocessor system, seems like an old 
> Unix
> language, but cant make heads or tails of it except it seems to let you
> compile the src for
> either Proquake (poq IN_POQ) or Quake World (qw / IN_QW). Would be 
> great if
> we could figure out a DP mode , IE: IN_DP, and I guess thats my job . . 
>  ..

Yuk :)~

Once you get married to a compiler, unless you built the code with the 
express purpose
of using other compilers, qc coder life can be most challenging.




As for head.m4 - these are some variety of define macros...

This is a lot like a function, but the compiler spits out code to 
compile.

m4_define([-IN_POQ-], [-m4_ifdef([-POQ-], $@)-])

Some places in code I saw:

create_te_explosion (self.origin, IN_POQ(1, 0));

So, if POQ is set:

create_te_explosion (self.origin, 1);

and if POQ is not set:

create_te_explosion (self.origin, 0);



If I understand this correctly, this was done to merge the qw and poq 
source trees, then define
a value for the one you want to compile.

If you really intend to rip out all the IN_* stuff, you will have to 
leave the correct code behind.

You will still face all the other issues.


Maybe pick either the qw or poq source and avoid the IN_ marcros?


Good luck!


Stop by and see my new game design: https://www.patreon.com/six_gaming