| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Comments:
<0> with that crazy C if substitute, it goes like so ...
<0> var ? /* if true */ : /* if false */;
<0> is that correct>
<0> s/>/?
<0> I'm looking at a bit of code that says ...
<0> arrayOfObjects[666] ? arrayOfObjects[666]->Method() : 0;
<0> isn't that backwards?
<1> return (planet == "earth") ? true : false;
<2> kiwi that's a little redundant
<2> oh, nevermind ;)
<0> presumably std::string planet("earth");
<1> just a example of '?
<2> yeah
<2> earthworm that looks like an ugly way of doing T* t = vec[666]; if( t ) return t->Method(); else return 0;
<0> well, that doesn't matter, but that matches what i said, so by that definition, the array example should bomb out all the time right?
<2> bomb out?
<0> crash, dereference NULL
<2> no
<0> right you are
<2> it should just be a little more verbose instead of trying to pack as much as possible into one line
<0> i think it is deliberate :)
<0> i would never right it like that, always use if, and always use == NULL
<0> that is my preference anyway
<2> I never use NULL
<2> it's just 0
<0> but yes, you are right, the code won't crash, it's just confusing
<2> and you don't need to include anything to know what 0 is
<0> what's wrong with NULL?
<0> i never do include anything, i thought it was a language feature
<2> it's got to be included from somewhere, or you need to define it
<2> no, it's not
<0> is this some msvc/g++ standard byp***ing?
<3> if (pointer)
<3> or if (!pointer)
<4> else..
<4> Nice.
<2> nod
<3> because, for one thing, they work with iterators, whereas == NULL doesn't
<2> or pointer = 0;
<3> no,
<3> never
<3> that's always false anyhow
<2> i don't mean as a conditional
<3> the IDIOM is if (something) {things are ok}
<3> like if(somestream)
<0> weird, i can do "main { char * wee = NULL; }" and run that through g++ and it will work, is g++ breaking the rules?
<3> or if(someiterator)
<3> likely, earthworm
<2> earthworm try it with -pedantic and -ansi or whatever those switches are
<0> does anyone know what header it is actually in?
<3> g++ breaks rules all the time
<3> what is in?
<0> NULL
<3> who cares
<3> don't use it
<0> that would mean killing an habit ...
<2> it's really not important enough to warrant ALL_CAPS, anyway
<3> good, kill the habit
<0> hehe
<0> i think msvc defines "null" as well, and maybe even colours it like a keyword, even worse (if memory serves)
<5> it accepts "main { " ? without the () and return type?
<0> oh come on
<2> no, but it does color nullptr, which is a keyword in C++/CLI
<0> YUY0x7 --set-pedantic="off"
<0> :)
<5> no, it does not color null, not when in native c++
<3> nope, neither null nor NULL
<0> okay, bad memory then
<5> earthworm, well i thought you were saying that it accepted that specific line of code
<5> which is exactly what you said
<0> nah, i thought people would get the drift
<1> never ***ume
<0> especially on IRC
<3> we all know how to parse *** / u / me
<0> well, i'll pack in using NULL then
<0> no major loss, 0 will do
<3> what do you want the 0 for?
<3> other than doing T* blah = 0;
<0> just so its crystal clear to read
<3> what's crystal clear to read?
<0> if(p_blah == 0) ...
<3> that's ****ed
<3> if (p_blah)
<3> or in your case
<3> if (!p_blah)
<3> AND it works with iterators etc
<0> hmm
<3> your ***umption that 0 is a meaningful value is so much hooey
<3> while (cin >> something)
<3> you don't so == 0 or != 0 there
<3> s/so/do
<0> fair enough mate, I'll try to kill that habit as well
<0> an thoughts on the if(69 == variable) {} style btw?
<0> somebody was on to me about that today, I always thought it read better the other way round
<0> the idea of course if to help you if you type if(69 = variable) {} by mistake
<3> well that way if you miss the 2nd = you get a compile error
<0> s/if/is
<3> instead of something that compiles but doesn't seem to make sense
<0> do you write your tests that way around?
<3> of course, have for years
<0> reckon much to the readability argument?
<3> what readability argument?
<0> excuse the barrage of questions here ... i just thought the statements read better the other way around
<3> it's likely the way you've used to them
<1> both ways are easy to read
<0> aye, probably is, somebody actually mandated the variable == constant way in our coding guidelines
<0> on the grounds of readability, but i'll stop boring the channel with this ****e now anyway, sorry folks :)
<6> what should I use to check if something is present or not in an input file? I.E. a number, inputted by the user.
<7> how big is "short" in bits ?
<5> thats platform dependent
<2> why does it matter, Cloud99 ?
<8> I think it depends on your os
<7> the os is windows 2000
<2> why does it matter, Cloud99 ?
<5> sizeof(short) * CHAR_BIT
<7> i am trying to p*** a memmory address to a function
<8> well he had actually platform
<2> Cloud99 - so?
<7> well the address is 2 bytes
<7> i am guessing its getting chopped off
<9> short is 16bits
<2> why are you p***ing an address as an integer?
<6> any ideas on my question?
<7> its a function exported by a dll
<7> to access parallel port
<2> and maybe you should p*** the type that the function accepts
<2> (I kinda doubt it accepts a short'
<2> unless it's an address into some other address space
<2> did you write this dll ?
<7> no
<7> i am just trying to use
<7> it
<2> so what does the function signature look like?
<7> one sec
<7> http://www.logix4u.net/inpout32.htm
<3> the address is TWO bytes? how do you know?
<7> all i know is that i need to write a byte into 0x39C to get a character to be sent over parallel cable
<3> that won't work in windows
<3> sorry
<7> well the .dll is supposed to solve that problem
<3> so, send the byte to the dll then. were's the problem?
<7> it doesn't work as its supposed to
<7> :)
<3> then I guess the dll didn't solve the problem
<6> so
<6> 8brownie bits in a brownie byte?
<10> hey - question - if i have a function hashing a bunch of bits using a CRC type function, would the fastest method be to open the file with filemapping? or use ReadFile with a normal buffer?
<10> i am not necessarily hashing the entire file
<3> hashing only part of the file seems almost pointless
Return to
#c++
or
Go to some related
logs:
#javascript
#MissKitten
#mysql
#linuxhelp
#skype
#MissKitten
#AllNiteCafe
#networking
#linux
#linux
|
|