| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8
Comments:
<0> /who #c++
<1> sk8: you could do worse than look at the Dragon book
<1> Amazon reviews from bitter undergraduates are hilarious
<2> the problem with that peterhu
<2> is that in future implementations
<2> it MAY change the state
<2> making an accessor necessary later
<2> and, forcing you to redo all the code that didn't use the accessor to begin with.
<3> huh? i just said that it doesn't make sense to have all "non-state changing" member functions as accessors
<4> so only make the methods that purely access & return private attributes accessors? (using const declaration)
<2> and, my argument is that it might make sense.
<2> for the sake of future maintenance.
<2> in case the way that cl*** functions changes.
<2> and, a non-state changing member becomes a state changing member somehow.
<2> Or, I guess my argument is more along the lines of, you should be careful with the decision.
<2> since I tend to hate accessor functions
<5> hi
<6> Properties > *
<5> what's the FP equivalent of itoa()?
<6> FP?
<6> Functional programming?
<5> floating point
<2> stringstream
<6> stringstream s; s << 123.56; string num = s.str();
<2> stringstream ss;
<6> Whee!
<2> float blah;
<2> ss << blah;
<2> yeah
<2> what he did.
<1> dont' forget to check state of the stream
<2> screw that!
<6> Or just use boost::lexical_cast
<3> lexical_cast > Solamente
<2> lexical_cast ****s
<1> Solamente: what you mean when you said Properties?
<3> noidea < Solamente
<1> is that C# or something?
<2> it just uses stringstream internally
<5> i see
<1> yea with allt he error checking done for you
<5> thanks
<2> pfft
<6> Cowmoo: .NET, really. It's not specific to C#.
<2> error checking is for losers
<2> what could possibly go wrong with his example?
<1> Solamente: oh ok
<6> I started writing a property template one time.
<6> I abandoned it for some reason... some kind of difficulty arose.
<6> I need to dig that back out.
<3> in c++/cli: ref cl*** Foo { property int Bar { int get() { return 3; } void set(int value) { /* whatever */ } } }; Foo^ f = gcnew Foo; f->Bar = 5;
<6> So, why haven't you added an article on that to the C++/CLI wiki?
<3> pfft
<3> i still have that irc bot to finish
<6> Lame excuse.
<1> irc bot
<1> wil it be for this channel?
<3> yes, it can join and ridicule jblitzen right now
<6> Dinnertime. Ravioli!
<3> but not much else
<1> I hope you're not writing it in C++
<3> c++/cli
<1> ok no comment.
<1> since I know nothing about that
<1> however, you are gaey.
<5> ok what do i do if i want to output a stringstream to the debug window?
<3> i'm glad moo doesn't have ops
<3> easier to mock him on wallops
<7> what debug window?
<3> ***uming windows, OutputDebugString
<5> i tried that and the compiler doesn't like the cast
<7> what cast??
<2> what's he helping with?
<2> I don't see him talking.
<3> wakjah, from what to what?
<3> the function takes a LPCTSTR, which is either const char* or const wchar_t* depending on UNICODE preprocessor define
<3> (in reality, the documentation takes that, OutputDebugString itself is a define)
<5> well it can't convert the parameter from a stringstream to an LPCTSTR so i tried a cast, which it didn't like either
<3> so if you're building UNICODE, you'd use a wstringstream, get the string, and then do c_str()
<3> if you're not building UNICODE, you'd use a stringstream as instructed above
<2> unicode is for losers anyway.
<1> we need Kniht and his calcs in here
<1> that deserter.
<2> his what?
<2> where did Kniht go?
<1> so somebody could've just calced convert2
<1> or convert
<1> alternatively, SOMEONE is taking his sweet time with the bot
<2> what?
<2> calced convert2?
<2> what are you converting?
<1> not for me fool
<1> for wakjah's original questino
<1> oh nevermind
<1> I'm off
<2> what was his question?
<7> wakjah and you don't cast
<4> cheers guys i'm off to read over my multimedia module (image processing in java - meh)
<7> you've been writing too much C
<5> sorry i was afk for a while there
<5> you don't cast? meaning just stringstream s; s << whatever; OutputDebugString(s); ?
<8> s.str() to recover the std::string, then c_str if you need a const char *
<8> or if this is still the "convert string of TYPE", boost::lexical_cast really is good
<8> everyone using C++ should have boost on their include path
<5> hum. is OutputDebugString(s.str().c_str()); right? if so then why is it saying there's a missing ; before OutputDebugStringA
<8> probably because you missed a ; before OutputDebugString
<5> i didn't, but then using a temp. const char * and setting that equal to s.str().c_str(), it doesn't complain
<5> however it doesn't work either
<6> Well, it's time to take the girls on their first golf game.
<6> I just hope they don't beat me.
<6> Elder daughter already drives around 175.
<9> will a declaration such as this one create 3 lists on the stack?
<9> list<ScatterVars*> l[3];
<7> bronaugh kinda
<9> kinda?
<7> well, what do you think a list is in this instance
<7> and a list of raw pointers????
<9> it could be a list of integers.
<7> that's not what you showed
<9> would it -matter- wrt my question?
<9> I don't have a bug. I'm asking now to make sure that the code snippet I pasted is going to behave as I understand it will, so that I won't have a bug later.
<7> well, WHAT exactlly do you expect to have "on the stack"
<10> A Queen of Spades and an Ace of Diamond
<7> c++ doesn't require a stack, you know
<9> well, I'd expect it to allocate 3 lists in that array and call the destructor for (and free the memory for) all three lists when they go out of scope.
<9> allocate and construct.
<7> well, it will do what you expect.... this isn't MS you know, not every thing is a special case
<8> stand warned on storing raw pointers in container though
<8> they become a pain to maintain
<7> and a list of raw pointers is kinda expensive..... two pointers overhead just to hold one pointer
<9> oh well. there's no more than 80 of them, tops.
<9> so it's not a huge problem.
<9> I suppose I should be dereferencing the pointers when I allocate them, then, and storing references to them?
<9> then taking the address of the object at the reference and free()ing that when needed?
<7> huh??????
<7> I have NO idea what you think you should be doing
<9> well... the items are dynamically allocated.
<9> items in the list.
<8> you might want to google for smart pointers
<9> nice...
<8> judicious use of smart pointers (i.e. choosing the right variety) gives very GCish behaviour to C++, except with a lot more flexibility
<9> so I see.
<8> (and more chances to shoot yourself in the foot, but there we go. don't treat them as a panacea)
<9> auto_ptr does look pretty nice for when you allocate memory within a function and just want it to clean up when it goes out of scope.
<8> indeed
<11> Smart pointers are the shiznit.
<11> They eliminate tons of potentially duplicated and therefore tedious and error-prone cleanup code.
<9> yeah...
Return to
#c++
or
Go to some related
logs:
#linuxhelp
#linuxhelp
undernet mangalia
#AllNiteCafe
Another synaptic is running ati
#php
#php
etho timed out debian
Binary was not build with debugging information msdn
wat does selam mean
|
|