|
Muz
|
 |
« Reply #3540 on: September 05, 2013, 07:18:03 PM » |
|
I'M WEB/GAME NOW!  What game is that? I've been looking for a web roguelike I can procrastinate with.
|
|
|
|
|
Logged
|
|
|
|
|
Quarry
|
 |
« Reply #3541 on: September 05, 2013, 08:08:51 PM » |
|
It's a mini project I'd like to spend more time on but can't
|
|
|
|
|
Logged
|
|
|
|
|
Geti
|
 |
« Reply #3542 on: September 05, 2013, 09:03:40 PM » |
|
Me: guess what this is:  Friend: local file inclusion vulnerability exploit? Not too far from the truth haha, start of a completely "scripted" text editor looking at the first 250 characters of its source (non whitespace inclusive). static vars in each script are basically "script local" storage as long as each object's script is compiled separately. I'm keeping it data oriented so each script works on a bunch of data - in part to justify them to being separate compiled "programs" haha, for example the animation script integrates all the animations, not just the animation for just one object. Much faster, and much more self contained code. It's fun thinking in a very non-OO way about this kinda stuff. It's currently like building pipelines for the data to flow through - or crash  Don't think this kinda stuff is suitable for production sadly, too easy to crash with C. We'll see though.
|
|
|
|
|
Logged
|
|
|
|
|
Graham-
|
 |
« Reply #3543 on: September 06, 2013, 06:19:43 AM » |
|
I'm 2% of the way towards making emacs the powerful and useful editor it should be, for my own tasks.
considering I started out at about around 2% I think I'm making pretty good progress.
|
|
|
|
|
Logged
|
|
|
|
|
Kekskiller
Guest
|
 |
« Reply #3544 on: September 06, 2013, 03:00:23 PM » |
|
Finished a first version of a module to replace shell scripts. It's essentially a set generators and processors for text files only containing paths/filenames, each in a single line. So far I can read, write, sort, generate (recursively with lotsa path options) and filter stuff using wildcard patterns.
Only thing left is spawning tools, functions or threads per for listed line. I already got nice generic command line option parsers (complete with keyword checks, required and optional arguments etc) so it should pair nicely to make some tools possible on all compilers that provide some sort of POSIX implemention.
|
|
|
|
|
Logged
|
|
|
|
|
Average Software
|
 |
« Reply #3545 on: September 06, 2013, 06:39:25 PM » |
|
I had this idea for a nifty little generic, and I just got done implementing it. Before:  After:  Yeah, I'm happy.
|
|
|
|
|
Logged
|
|
|
|
|
Kekskiller
Guest
|
 |
« Reply #3546 on: September 07, 2013, 01:06:19 AM » |
|
Hey, that 404 is sort of an awesome for not-yet-implement but already desired parts of an hud.
May I shameless borrow it and become rich and wealthy?
|
|
|
|
|
Logged
|
|
|
|
|
Graham-
|
 |
« Reply #3547 on: September 07, 2013, 06:55:58 AM » |
|
466? Holy shit.
I'm using udk and there are two complaints: . slow . generating content (models) not straight-forward
|
|
|
|
|
Logged
|
|
|
|
|
Prinsessa
|
 |
« Reply #3548 on: September 08, 2013, 06:06:35 AM » |
|
I have today a semi-reasonable reason to overload the () operator in one of my C++ classes, and I will not hesitate. This is a great day.
|
|
|
|
|
Logged
|
|
|
|
|
Average Software
|
 |
« Reply #3549 on: September 08, 2013, 08:23:24 AM » |
|
I have today a semi-reasonable reason to overload the () operator in one of my C++ classes, and I will not hesitate. This is a great day.
You obviously don't write very many functors. I overload () all the time. The holy grail for operator overloading is , (comma). Still haven't found a good use for that one.
|
|
|
|
|
Logged
|
|
|
|
|
Prinsessa
|
 |
« Reply #3550 on: September 08, 2013, 09:56:10 AM » |
|
I have today a semi-reasonable reason to overload the () operator in one of my C++ classes, and I will not hesitate. This is a great day.
You obviously don't write very many functors. I overload () all the time. This is my first or second time or something implementing a component-based engine powered by scripting rather than an inheritance-based one completely made in pure, compiled programming, so things are turning out very differently than they normally do. This is such a nice way to work, anyhow. I'm also using a lot of very unorthodox templates and nobody shall stop me from doing it. The holy grail for operator overloading is , (comma). Still haven't found a good use for that one.
I think I had some funny idea for that some time, but I no longer remember it.
|
|
|
|
« Last Edit: September 08, 2013, 10:07:54 AM by Skomakar'n »
|
Logged
|
|
|
|
|
_Tommo_
|
 |
« Reply #3551 on: September 08, 2013, 12:42:31 PM » |
|
I think I had some funny idea for that some time, but I no longer remember it.
This might be an use: Vector v = 0,1,2;
Also, if you do that on code I work on I will punch you in the face 
|
|
|
|
|
Logged
|
|
|
|
|
BleakProspects
|
 |
« Reply #3552 on: September 08, 2013, 01:04:28 PM » |
|
You could also use it to generate a sequence in a behavior tree: Seq = B1, B2, B3;
|
|
|
|
|
Logged
|
|
|
|
|
Prinsessa
|
 |
« Reply #3553 on: September 08, 2013, 02:38:30 PM » |
|
I like both of those suggestions, actually. I made a vector class today. I'm so trying that. It's genius. Well, wait, does the comma operator not just return the vector again or something? How is it supposed to know which coördinate is which? I don't think it will work... 
|
|
|
|
|
Logged
|
|
|
|
|
Prinsessa
|
 |
« Reply #3554 on: September 08, 2013, 05:36:02 PM » |
|
|
|
|
|
« Last Edit: September 08, 2013, 10:48:06 PM by Skomakar'n »
|
Logged
|
|
|
|
|
_Tommo_
|
 |
« Reply #3555 on: September 09, 2013, 01:54:31 PM » |
|
Well, wait, does the comma operator not just return the vector again or something? How is it supposed to know which coördinate is which? I don't think it will work...  I guess you could make a constructor that takes 1 float and sets x to it and y and z to NaN; then comma sets either y or z depending on which one is NaN?
|
|
|
|
|
Logged
|
|
|
|
|
Average Software
|
 |
« Reply #3556 on: September 09, 2013, 05:12:05 PM » |
|
Well, wait, does the comma operator not just return the vector again or something? How is it supposed to know which coördinate is which? I don't think it will work...  You return a bridge object that holds a reference to the vector and an indicator of which position to assign next, and you overload comma for that too. Something like so: CommaBridge operator , (Vector &lhs, int rhs) { lhs[1] = rhs; // Position 0 was already assigned by Vector::operator = (int).
return CommaBridge(lhs, 2); // Position 2 is next. }
CommaBridge &operator , (CommaBridge &lhs, int rhs) { lhs.vector[lhs.position++] = rhs; Assign the next position, then increment.
return lhs; }
|
|
|
|
|
Logged
|
|
|
|
|
Geti
|
 |
« Reply #3557 on: September 09, 2013, 05:40:41 PM » |
|
...Is it just me that doesn't have any problem with Vec3f v(1,2,3); then?
|
|
|
|
|
Logged
|
|
|
|
|
eigenbom
|
 |
« Reply #3558 on: September 09, 2013, 05:52:06 PM » |
|
...Is it just me that doesn't have any problem with Vec3f v(1,2,3); then? You mean the normal way? Used in glsl, sfml, box2d, etc. I don't think you're alone..
|
|
|
|
|
Logged
|
|
|
|
|
Average Software
|
 |
« Reply #3559 on: September 09, 2013, 06:10:50 PM » |
|
...Is it just me that doesn't have any problem with Vec3f v(1,2,3); then? You mean the normal way? Used in glsl, sfml, box2d, etc. I don't think you're alone.. That's fine for a vector of fixed size. This is assuming a vector of unknown size. Of course in C++11, you can just use a std::initializer_list
|
|
|
|
|
Logged
|
|
|
|
|