Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Josh @ Dreamland

Announcements / Re: Major forum upgrade
« on: December 04, 2009, 09:46:27 pm »
Anorexic... Font... used to... Ubuntu...

Announcements / Re: Major forum upgrade
« on: December 04, 2009, 03:46:57 pm »
The 'jump to' is broken.

I'm posting from the university library to tell you all:

*jumps out window*

Announcements / Re: Major forum upgrade
« on: December 02, 2009, 06:08:33 pm »

Proposals / Re: Just an appreciation
« on: December 02, 2009, 05:53:16 pm »
Retro-- Most of those are mutually exclusive
Score_-- It gets about 300 KB with -Os, I believe. I can probably do some custom optimization to help that when the time comes.
Anyway, essay'ing to do, ciao.

Announcements / Re: Trunkification (svn rearranged)
« on: December 01, 2009, 08:19:31 pm »
ENIGMA's going to be doing the same on my part for a bit here. I gave serp SVN commit privileges, so perhaps he will develop it. I'll be doing my three essays and presentation this week and Monday. After that, I'm free...

Off-Topic / Re: Code::Block Plugins
« on: November 29, 2009, 10:30:02 pm »

GML was easy to read because it had nothing. It had some 500 functions, all of which were highlighted. About 2,000 global things, also highlighted. That's it.
This C parser gives ENIGMA what it needs to highlight C++ functions, but honestly, I don't think that's such a good idea anyway.

Also, I hate when #define stops being highlighted after the space. That's annoying as hell. But if you like that, maybe someone else does and has written such a plugin.

Announcements / Re: Victory
« on: November 28, 2009, 08:01:30 pm »
Okay, I won't.

Announcements / Re: Victory
« on: November 28, 2009, 11:37:36 am »
Fede's going a little stir-crazy over the amount of time since I've made a release. Maybe someone will take the incentive to splice R3 and R4... hehehe. Right.

Announcements / Victory
« on: November 27, 2009, 03:24:50 pm »
Code: [Select]
No error.
Parse time: 380 milliseconds

Macros (518) [+]
Variables [+]
Define: acos
  acos:  Function with 1 parameters, returning double  Dereference path: (params = 1)</group>

After I had what I thought to be the basics down, all the Windows C headers parsed without error. The Linux ones, however... It was a constant uphill battle.
Math.h employs such faggotry as calling __CONCAT on itself, which only works if it's called internally as an alias. For example, the first code works, the second does not:

Code: [Select]
#define __CONCAT(a, b) a ## b
#define TRIPLE_CONCAT(a,b, c) ALIAS_CONCAT( __CONCAT(a,b),c)
#define ALIAS_CONCAT(a, b) __CONCAT(a,b)
Code: [Select]
#define __CONCAT(a, b) a ## b
#define TRIPLE_CONCAT(a,b, c) __CONCAT( __CONCAT(a,b),c)

I'm not entirely sure why, really, and I don't care. Both work for me, and I see no reason to change it just because GCC only likes the first.

Anyway, more work to do. (Along with two research papers and a "scientific" paper to write). I'm a bit swamped, as usual.

Issues Help Desk / Re: Compiling on Linux... Possible?
« on: November 25, 2009, 10:34:33 pm »
Sure. R3 had a mechanism for using GCC ad-hoc, if you liked. (Search compile.h or whatever it was for the specifics).

Not sure what you want to do. If you have GCC installed already and don't want to download the core, that's the reason it was divided into two segments. However, I later realized what a bad idea that was since most people just want an easy-to-install package.

If you intend on getting it to work for Linux, it already does for R4, and getting R3 to do it would be a trick (though feasible for those who can work SVN and C++ enough to download and install ENIGMA's Linux port from the repo).

Off-Topic / Re: What does ENIGMA stand for?
« on: November 22, 2009, 09:58:41 pm »
For open source, C++ DLLs, they can just be included, yes.

I was thinking about asking Ism to incorporate a DLL resource (I believe GM5 had something similar to that, but I was too young to use it at the time.) Basically, the resource would do implicitly what the code does, but at compile time to (potentially) save speed later. I'm unsure of the actual speed difference caused by conditionally pushing things onto the stack, as is necessary in making DLL calls of undefined argument counts of variable-sized objects. (Namely double (8 byte) and char* (4 byte) ). Considering the fact that the calls could be hard coded, I imagine you could see a speed difference of almost 2x, but that's just for call. Extensive functions you call would render the difference in call time insignificant.

Such a resource, therefore, could save on number of lines of code in the case that a DLL's source is unavailable, and can potentially save on speed. So you may see that come along in the not-so-distant future.

It means "Josh was too fackin' lazy to do the easy part first."

While the G-Java teams decided to tackle the project on a more manageable basis, starting with DND, I imagine that was because we were all younger then. They didn't know how to write a parser at the time, hadn't heard of Yacc or the like. Incidentally, I wrote their first parser, in GML. Haha. It was terrible, but it worked for the most part. Unlike my newer ones, it wasn't very structured. It only got the job done because GML is such a small language.

Anyway, enough reminiscing. Long story short, I did code first because I didn't feel like taking a shortcut and alienating the most useful resource. In R4, Ism converts DND tiles to their respective action_ codes before she sends it to me. Until R4 is out though, stick with code.

General ENIGMA / Re: C++ Socket Based Networking
« on: November 22, 2009, 09:49:39 pm »
My vision for MPlay was to take Ism's code and add a loose-fitting wrapper to it. By loose-fitting, I mean that half the functions won't do anything. Like session ID and the related functions. Honestly, I wrote an MPlay library for GM once, and the terminology was so vague and preposterous I never managed to give it my own naming convention.

Not to mention the fact that, as far as I can tell, GM's MPlay hardly works for anyone. Seems to me to be for a couple reasons. Number one, those who use MPlay do so because they don't understand 39Dll or don't have a registered GM. These are the people that don't understand that MPlay isn't designed to turn their routed network into an MMO server.

Everyone that does understand that concept moves onto 39Dll. Except me, I suppose. ...Okay, so some people have made games with it, but honestly, I've never seen an MPlay game that was truly worth porting anywhere. All the good ones are 39Dll (Do correct me if I'm wrong, I'd be happy to believe MPlay was a sound system)

So ideally, MPlay will "work," but it will work in the same sense that it is understood: Superficially.

Off-Topic / Re: What does ENIGMA stand for?
« on: November 22, 2009, 10:48:58 am »
As opposed to compiling it at, say, compile time. It'd probably be too much effort to re-reference that stuff at load time if he precompiled it. ...It'd still be pretty easy, though.

The GML frontend is a thin layer over C++. Not to mention the project is open source. Anyone can extend ENIGMA's functionality who knows even a little C++, and with the new parser, you can include C++ headers.

Indeed a play on Game Maker. ENIGMA doesn't have a bloaty format with a slow interpreter, it passes the code to a very capable compiler with a long-developed optimizer.

Game Maker-
That program that I'm considering disowning.

To augment is to change, specifically to add to, or intensify. ENIGMA adds C++'s syntax goodies to the mix, such as ++ and, well, typing, templates, you name it. Not to mention the possibility mentioned in E of including any C++ library.

Announcements / Re: stdio.h
« on: November 16, 2009, 10:31:43 pm »
It's a wrapper to ShellExecute. It's not implemented due to uselessness.

Because I ground all COW into BEEF.