| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
<binrapt> b1999: byte = byte << 1;
<b1999> lol
<b1999> of course
<Kniht> b1999: if a += 1 is a = a + 1, then that is ...
<b1999> can you guys come to the exam with me?
<binrapt> Sure, to tell the teacher that you are cheating.
<NKD> lol
<b1999> you're saying you never cheated ? never got 'help' from soneone?
<Kniht> there's a difference
<roopjm> i cheated like a mother**** in 8th grade... teacher didn't teach, just put notes on the overhead, so we said **** her and blatently cheated
<binrapt> Does somebody use 'int const' instead of 'const int'?
<Soronel> roopjm: they were probably the same notes the teacher used the previous year too
<Kniht> binrapt: i've been going in that direction
<Soronel> binrapt: I've been trying to change my default style to use type const
<Soronel> though it conflicts with VS quite a bit
<binrapt> Oh, how does it conflict?
<Soronel> intellisense always shows it as const type, regardless of how you write prototypes/definitions etc
<Symmetry_> binrapt: sounds perverted to me
<binrapt> Symmetry_: Why?
<Kniht> Symmetry_: const int is a special case
<Symmetry_> I meant kinky. nothing wrong with the way our ancestors wrote "const int"
<Symmetry_> oh really... how?
<Soronel> Sym: the fact that const type works at all is a sortof artifact
<Kniht> calc const rule
<CalcMe> const rule = The keyword "const" modifies what it is after, unless there is nothing, in which case it modifies what it is before. Thus, the type "const int" is a special case and means the same as "int const". See also: calc method const
<Soronel> Sym: the general rule is that such modifiers apply to what is to the left of the modifier
<Symmetry_> ah, so it's not different after all
<Symmetry_> I see. good to know
<Soronel> calc test
<Symmetry_> so the canonical way would be int const?
<binrapt> Nod
<Soronel> hrmm
<Symmetry_> hah.
<Soronel> okay, brb
<binrapt> Hmm I love order, I think I will start using it
<oxix_m> I used to put comments above include guards, but Sutter says not to
<oxix_m> apparently compilers can detect include guards early or something and not putting anything before them (including comments) can help
<oxix_m> you guys do that?
<Genia4> http://bash.org/?105481 read and learn!
<binrapt> oxix_m: Yes
<YUY0x7> Genia4 LOL
<oxix_m> binrapt: put comments above include guards or avoid doing that?
<binrapt> oxix_m: Not putting comments above include guards
<oxix_m> ah
<Supram> Hey.. I gota few questions about c++.. I wanne try to start to learn the language.. But what do i need to start with it? What kind of programs?
<oxix_m> I like it now because it's nice and uniform
<Supram> Do i need a compile program etc?
<oxix_m> always at the top of any of my headers is an include guard
<binrapt> Supram: A book and a compiler
<Supram> Where do i get a compiler ?
<binrapt> Book: Accelerated C++, www.acceleratedcpp.com
<binrapt> Compiler: Dev-C++ for example, www.bloodshed.net, free
<Supram> thanks alot.
<oxix_m> return (a > b) - (a < b); <-- wazzat do?
<oxix_m> a > b and a < b can't both be true
<oxix_m> so it can be 1 or -1... or 0 if both aren't true
<oxix_m> is that just a clever way to write like a cmp function?
<wilx> Heh, interesting.
<wilx> Probably.
<oxix_m> ah - that makes sense in this function's context
<oxix_m> looking at C code
<oxix_m> that's a clever way to do it
<oxix_m> though C++ doesn't make use of any cmp functions like that generally, or does it? can't think of any STL parts that uses functions like that to order things
<oxix_m> strict weak ordering predicates for set and map
<Kniht> the STL uses a single comparison functor that is called twice (with arguments reversed) to determine equality
<oxix_m> ah
<Kniht> similar to the code you pasted, except use < instead of > and switch arguments
<Kniht> template< typename T, typename F > is_equal( T const& a, T const& b, F f ) { return !( f(a,b) || f(b,a) ); } // where f returns true if the first parameter is ordered before the second parameter
<oxix_m> very tight logic
<oxix_m> i like it
<Kniht> usually i use it like if ( f( a, b ) ) { /* handle a ordered before b */ } else if ( f( b, a ) ) { /* handle b ordered before a */ } else { /* handle equality */ }
<oxix_m> guess in C they want to jam more logic in there since they can't inline the function call using function pointers
<oxix_m> ah, no wasted comparisons that way
<kaber2> hey
<oxix_m> hi
<kaber2> if any of yall are on a fast connection. which are you gettin faster pull speeds at
<kaber2> http://www.fileget.net/test100M.dat or http://www.liquidweb.com/data/100mb.bin
<kaber2> on cable. liquidweb is about a 100kb faster
<skie> kaber2: 257 on the first, 389 on the second
<kaber2> k
<skie> I can try from work (university connection) on tues, but I can't RD into my machine right now
<kaber2> aite cool. appreciate that
<skie> trying to decide who to host with?
<kaber2> i'm ssh'ng it from theplanet and i'm pulling from liquidweb about 1.2 MB
<kaber2> yea
<skie> hm. actually, I can try from OSU>
<skie> I forgot I had a shell acct
<skie> fileget hates OSU. 50kbps on fileget, 1.7M on liquidweb
<Ghostmod`> i get the following linker errors in VC++ 6.0, http://rafb.net/paste/results/J9jsFN77.html with the suggestion of using /NODEFAULTLIB:library. Where do I put this? I throw it in the box in the project->settings->link that has a list of all my linker options, but it doesn't change any of the errors.
<skie> (and liquidweb was still going up when it finished)
<kaber2> guess no contest. liquidweb is way faster then ev1server's
<skie> some of it may be DC location
<skie> & just how it gets routed
<Ghostmod`> kaber2: you looking for a webhost provider?
<Supram> Hmm.. i downloaded a project that i could edit. And i downloaded dev-c++ 4 too. But how do i compile the project? Its more that one file and i can only open one at once in the dev program?
<skie> Supram: makefiles.
<kaber2> aite, gonna go ahead and get it with liquidweb
<kaber2> Ghostmod`: dedicated server's
<Supram> skie, its my frist time will you please be a bit more precis?
<Ghostmod`> kaber2: well ev1 servers get good speeds in general, not just to a few random specific places
<Ghostmod`> the planet is good too
<kaber2> yea
<kaber2> I contacted them, but there all sold out
<Ghostmod`> my server is in a savvis data center
<Ghostmod`> the planet goes through savvis iirc
<skie> kaber2: try to get some Level3 bw :)
<Ghostmod`> just stay away from anyone that mentions 'cogent'
<skie> heh
<Soronel> that was annoying
<kaber2> hehe
<kaber2> but liquidweb is running CentOS 4. But i think that's fine, we just need to be on the new 2.6 kernel
<kaber2> yea. Centos4 has 2.6 installed
<Asmodee`> you could always do an upgrade as soon as your server is installed
<Asmodee`> i plan to start reselling space where my server is soon ;P
<kaber2> hehe
<skie> kaber2: what are you getting a ded box for?
<kaber2> for some server software needs 2.6 and own box
<domaro> i include #math.h yet i still get linker errors when using log() or other math functions...does anyone know what i need to do to fix this?
<skie> link with -lm
<jakl> hi, i declared an array of BallType which is a type from a library i downloaded. So i declared that array as a member of my cl***. However, i wanted to initialise that array in one of my function like this : BallType types[16] = {CUE, ONE, TWO, THREE, FOUR, FIVE, SIX, SEVEN, EIGHT, NINE, TEN, ELEVEN, TWELVE, THIRTEEN, FOURTEEN, FIFTEEN}; but for some reason it won't work unless i declare it global to the whole file...
<domaro> ok thanks skie
<Asmodee`> anyone know of other ways to force VC++ to not link standard libraries aside from /NODEFAULTLIB:library? it doesnt seem to have any effect when i throw it in to my link options
<YUY0x7> jakl, what does "it won't work" mean?
<jakl> i mean by that, that the whole array is instead initialised to 0
<jakl> ONE TWO THREE...etc are types defined to 1,2,3...
<agentsim> jakl: you cannot ***ign a value to an array using the {1,2,3} format, that is for initialization only
<Supram> what happen after i compile a file, ?
<jakl> so i will have to do types[0] = CUE, types[1] = ONE etc..?
<agentsim> jakl, yep
<YUY0x7> he is initializing there, so it should be fine
<jakl> well actually, in the declaration of my cl***, i have BallType types[16]; and then later on in one of my functions i wanted to do initialise types[]..
<Supram> what happen after i compile a file?
<Supram> Because seems that nothing happen here
<YUY0x7> jakl, oh well if it's a member variable you might not be able to initialize it like that
<kaber2> aite guys, i g2g. peace
<agentsim> you might be able to do it like that if you can initialize during construction
<YUY0x7> the only way i can think of would be in the constructor, mycl***() : types({CUE,BLA,BLY}) {} but iirc that doesn't work :/
<agentsim> hence the "might" in my suggestion :)
<YUY0x7> jakl, is this a static member?
<Indi_J> can't you make a hidden array that's already initialized, then just copy?
<piz> hey. i need to get advice, i want to start learning programming language, which one i should start with (it's better to be a "new" language and not an old on that doesn't support any more or a lot)
<Indi_J> or even better, wouldn't this work better as an enum?
<upb> php or vb
<piz> upb, php is a website programming language no?
<upb> yes
<piz> and c++ is for websites as well? (sorry for ignoarnce)
<YUY0x7> dont recommend vb :/
<YUY0x7> piz, no.. C++ is a general purpose language
<upb> just trying to avoid competition here :P
<Asmodee`> if you intend to do application development learn C++, if you want to do low level stuff, like embedded systems start with C
Return to c++
or
Go to some related
logs:
penntechnical.com
osdev
|
|