next »

[Hideous]

I downloaded and installed/compiled AllegroGL. I had no idea it lets me use normal Allegro drawing commands, except with hardware acceleration o.o

*squeee*

[increpare]

I've found myself using unsigned ints a lot recently.  I derive a definite pleasure from doing so.

[dustin]

I got bitmaps working in flash and I think my idea will be very easy to implement!

[Zaknafein]

I'm half grumpy, half happy about the Code Contracts features of .NET 4.
I loved the Spec# syntax for defining contracts, preconditions, etc., and knowing that they're using the ideas in real C# is great, but I think I would've preferred reserved keywords to a library.
The "!" for non-nullable types was especially sexy.

[Mikademus]

I take objection to this thread. Programmers aren't supposed to be happy, they're supposed to hate MS APIs, any code not made by them (and old code made by them), and quaff Mountain Dew and Doritos in a futile attempt to hide from the miseries of life.

There, I have burst your false bubble of illusory happiness, now please return to the grind.

[Hideous]

irrlicht comes with a codeblocks project file.

easy compilation of libraries make me very happy (unlike that freeglut thing) <3

[Cymon]

I am happy now that I got my dude walking across the screen, pushing boxes, and stepping up on them. Freaking yeah, I got a game now.

[Will Vale]

I've found myself using unsigned ints a lot recently.  I derive a definite pleasure from doing so.

Good stuff! Far too many 'counts of things' are stored in ints, even though you can't have a negative number of whatever thing is being counted.

[bateleur]

Good stuff! Far too many 'counts of things' are stored in ints, even though you can't have a negative number of whatever thing is being counted.

Equally, one could argue that storing a count of something as an unsigned int is crazy if you can't have more than 2147483647 of them!

[increpare]

Equally, one could argue that storing a count of something as an unsigned int is crazy if you can't have more than 2147483647 of them!

No, that's precisely the best time to have counts stored as unsigned ints 

[Ivan]

When i was using pure C, I paid attention to the size of my variables. I would go so far as to only use single bytes for counts that would never be larger than 256. Now that I use only C++, i don't think about it too much, especially since most of the math that happens uses floating point math and i find myself storing things as floats so i dont have to bother with casting things left and right.

[mirosurabu]

I downloaded and installed/compiled AllegroGL. I had no idea it lets me use normal Allegro drawing commands, except with hardware acceleration o.o

*squeee*

Try OpenLayer. It's basically AllegroGL but comes with some neat things.

Regarding topic,

I haven't touched C++ since december last year. That makes me happy programmer.

[Gold Cray]

I tried to compile a game with VC++ and then Code::Blocks in an attempt to use the C++ version of fmodex. After failing in both cases (in retrospect, I could have succeeded with code::blocks if I had realized what was going on), I finally installed Dev-C++ and converted my game to the C version of fmodex. It worked the third time (silly mistakes on my part involving included files). Dev-C++ has always been there for me.

[increpare]

devc++ stopped working for me recently (even when I try reinstalling stuff).  managed to get allegro compiling nicely with mingw myself today though.

I've abandoned cygwin for the moment; it had some nice parts to it, but overall I find mingw (well; I think I'm using something that's not strictly mingw, some sort of newer build of it so I can compile the latest boost versions) more amenable.

[Greg]

unsigned is good, but signed-unsigned comparisons can be a gotcha.

Code:

unsigned int i = 1;
int j = -1;
if ( i < j )  
    printf( "what the?\n" );

[increpare]

-Wall

[David Pittman]

-Wall

This. It will suck at first and you'll want to turn it off, but it feels way, way better to fix everything and maintain 0 warnings.

[Cthulhu32]

I rewrote the Winsock2 hook code at my new company for our main app, apparently nobody except the original Russian coder understood what it did. I also found an error in the assembly, because it works a little like this

[call function] 5 bytes
[nop space to fill old winsock2] 8 bytes
[jmp function] 5 bytes

Original article here: http://www.internals.com/articles/apispy/apispy.htm

the call function calls our new function,
the nop space is filled in by the previous first 5 or 6 bytes of the winsock function,
the jmp function is overwritten by an actual jmp offset between the old winsock2 call and our DLL inject.

Everything was fine except when send() was overwritten, the nop space was overwritten WITH the new offset jmp, leaving 2-3 bytes behind. Which means if any of those bytes is a 0x10, thats a system interrupt, kablooey to whatever is in eax possibly. (including system restarts)

I also hooked WSASend after I rewrote the code, and I have impressed the entire office 

[Lynx]

Most impressive!  That's Arcane with a capital A.

[increpare]

got guile examples up and interpreted with close to 0 difficulty, and just about to try getting ecl up and working as well.  (I've been feeling a little guilty at how much I've been using lua recently; need to broaden my horizons just a little (not that I've really used lua that much or anything, more that a change of scene would be nice   ).

It constitutes a sort of bedtime treat: I've not had too productive a day today coding-wise.