Trying to understand darkplaces source code

[vibok]

An svc_signonnum has been received, perform a client side setup.

[vibok]

BSDmakefile -- this is for BSD OS, don't need that.

[vibok]

I can just upload the builded with doxygen documentation somewhere, if anyone here needs it.

[vibok]

Both makefile and BSDmakefile include makefile.inc with the use of "include makefile.inc".

[vibok]

Probably crypto is needed for online play, haven't tested.

[Baker]

Choose Your Own Adventure:

You are standing at the edge of swimming pool.

1. If you want to stand at the side of the pool ... there is a video series that should entertain you. https://www.youtube.com/watch?v=yPHJgfSnTbI

2. If you plan to jump in the pool, there are QuakeC tutorials here. InsideQC

I hope you enjoy watching the videos

[vibok]

@Baker
First video series is, probably worth watching. I should've linked it by now. As a warning though, it's a bit slowpaced, it takes him way too much time to get in the quake related stuff (first interesting video is the Handmade Quake 4.3 - Loading Quake's Pak Files, way towards the end). Pretty good if you want to learn winapi, I guess. And he uses software rendering, which you have no reason to even know how to do nowadays. (Not sure if he even got that far.)

I'll read the second link, but it's probably more for modders. May be useful, who knows.

You enjoy being a dick, huh. My sense of humour circuits fried a long time ago, while paranoia seems quite active.

[vibok]

Just watched "Handmade Quake 5.1 - Intro To Memory Management", and I'm actually glad he went for software renderer. He shown how memory is organized, and where the z-buffer is stored. Dos times memory management is hilarious, it's a lost art nowadays.

Darkplaces does things much differently, I don't understand what's going on in zone.c and zone.h. Maybe I'll get it one day.

[Baker]

You enjoy being a dick, huh.

Will I call out a make-believe modder? I sure will.

You'll be able to watch those videos for weeks, collecting more and more information while not applying it.

You'll be the most knowledgable guy that hasn't opened an IDE, hasn't opened a map editor and hasn't done a QuakeC tutorial ever!

[vibok]

What's a make-believe?

[vibok]

Any idea why this doesn't work?

static void _vector_max(const vec3_t * a, const vec3_t * v) {
    if(v[0]>a[0]) {a[0]=v[0];}
}

error: assignment to expression with array type

This is C, not QuakeC.

[vibok]

annoying warning in this function, let's throw it here
i guess i'll just disable it, don't wanna touch it

or maybe I'll redo it as a macro or something, it blends together as it is.

sv_phys.c:758
In function 'RotateBBox':
warning: this 'if' clause does not guard... [-Wmisleading-indentation]

static void RotateBBox(const vec3_t mins, const vec3_t maxs, const vec3_t angles, vec3_t rotatedmins, vec3_t rotatedmaxs)
{
	vec3_t v, u;
	matrix4x4_t m;
	Matrix4x4_CreateFromQuakeEntity(&m, 0, 0, 0, angles[PITCH], angles[YAW], angles[ROLL], 1.0);

	v[0] = mins[0]; v[1] = mins[1]; v[2] = mins[2]; Matrix4x4_Transform(&m, v, u);
		VectorCopy(u, rotatedmins); VectorCopy(u, rotatedmaxs);
	v[0] = maxs[0]; v[1] = mins[1]; v[2] = mins[2]; Matrix4x4_Transform(&m, v, u);
		if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0]; if(rotatedmins[1] > u[1]) rotatedmins[1] = u[1]; if(rotatedmins[2] > u[2]) rotatedmins[2] = u[2];
		if(rotatedmaxs[0] < u[0]) rotatedmaxs[0] = u[0]; if(rotatedmaxs[1] < u[1]) rotatedmaxs[1] = u[1]; if(rotatedmaxs[2] < u[2]) rotatedmaxs[2] = u[2];
	v[0] = mins[0]; v[1] = maxs[1]; v[2] = mins[2]; Matrix4x4_Transform(&m, v, u);
		if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0]; if(rotatedmins[1] > u[1]) rotatedmins[1] = u[1]; if(rotatedmins[2] > u[2]) rotatedmins[2] = u[2];
		if(rotatedmaxs[0] < u[0]) rotatedmaxs[0] = u[0]; if(rotatedmaxs[1] < u[1]) rotatedmaxs[1] = u[1]; if(rotatedmaxs[2] < u[2]) rotatedmaxs[2] = u[2];
	v[0] = maxs[0]; v[1] = maxs[1]; v[2] = mins[2]; Matrix4x4_Transform(&m, v, u);
		if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0]; if(rotatedmins[1] > u[1]) rotatedmins[1] = u[1]; if(rotatedmins[2] > u[2]) rotatedmins[2] = u[2];
		if(rotatedmaxs[0] < u[0]) rotatedmaxs[0] = u[0]; if(rotatedmaxs[1] < u[1]) rotatedmaxs[1] = u[1]; if(rotatedmaxs[2] < u[2]) rotatedmaxs[2] = u[2];
	v[0] = mins[0]; v[1] = mins[1]; v[2] = maxs[2]; Matrix4x4_Transform(&m, v, u);
		if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0]; if(rotatedmins[1] > u[1]) rotatedmins[1] = u[1]; if(rotatedmins[2] > u[2]) rotatedmins[2] = u[2];
		if(rotatedmaxs[0] < u[0]) rotatedmaxs[0] = u[0]; if(rotatedmaxs[1] < u[1]) rotatedmaxs[1] = u[1]; if(rotatedmaxs[2] < u[2]) rotatedmaxs[2] = u[2];
	v[0] = maxs[0]; v[1] = mins[1]; v[2] = maxs[2]; Matrix4x4_Transform(&m, v, u);
		if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0]; if(rotatedmins[1] > u[1]) rotatedmins[1] = u[1]; if(rotatedmins[2] > u[2]) rotatedmins[2] = u[2];
		if(rotatedmaxs[0] < u[0]) rotatedmaxs[0] = u[0]; if(rotatedmaxs[1] < u[1]) rotatedmaxs[1] = u[1]; if(rotatedmaxs[2] < u[2]) rotatedmaxs[2] = u[2];
	v[0] = mins[0]; v[1] = maxs[1]; v[2] = maxs[2]; Matrix4x4_Transform(&m, v, u);
		if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0]; if(rotatedmins[1] > u[1]) rotatedmins[1] = u[1]; if(rotatedmins[2] > u[2]) rotatedmins[2] = u[2];
		if(rotatedmaxs[0] < u[0]) rotatedmaxs[0] = u[0]; if(rotatedmaxs[1] < u[1]) rotatedmaxs[1] = u[1]; if(rotatedmaxs[2] < u[2]) rotatedmaxs[2] = u[2];
	v[0] = maxs[0]; v[1] = maxs[1]; v[2] = maxs[2]; Matrix4x4_Transform(&m, v, u);
		if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0]; if(rotatedmins[1] > u[1]) rotatedmins[1] = u[1]; if(rotatedmins[2] > u[2]) rotatedmins[2] = u[2];
		if(rotatedmaxs[0] < u[0]) rotatedmaxs[0] = u[0]; if(rotatedmaxs[1] < u[1]) rotatedmaxs[1] = u[1]; if(rotatedmaxs[2] < u[2]) rotatedmaxs[2] = u[2];
}

[MadGypsy]

Well, all those ifs have misleading indentations. They aren't a child of v but they are indented as if they are. It would be really sad if this was the actual problem. This would mean your compiler treats tab as if it has meaning. I would seriously hope something else is the problem.

[vibok]

Oddly enough, replacing all
if(rotatedmins[0] > u[0]) rotatedmins[0] = u[0];
with
if(rotatedmins[0] > u[0]) {rotatedmins[0] = u[0];}
makes my compiler shut up. Even if I keep identation the same.

And if I fix identation, but skip {}, it will complain.

Haven't pasted the whole error, my bad. Here it is:

warning: this 'if' clause does not guard... [-Wmisleading-indentation]
note: ...this statement, but the latter is misleadingly indented as if it is guarded by the 'if'

[MadGypsy]

I thought about that as well but argued myself against mentioning it cause ifs should only need brackets if it regards more than one line of code. The environment is too strict. It shouldn't be worried about how the user format's their code... only that every thing between semi-colons is legit. Regarding the users whitespace is silly.

You know the first thing every text parser I have ever written does? It sequesters all strings, replaces them with tokens and removes all whitespace. This is because whitespace has no meaning unless it is in a string so, first order is to get rid of all of it.

My point being that your compiler should be compiling code that is technically all on one line due to a quick preprocessing of the script.