| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
<jamesd-> noam: Java GC is mostly about reclaiming memory; correct robust code can't rely on finalizers, and making correct finalizers even when they do run is relatively hard.
<jamesd-> So you have to use finally all over the place in client code in Java, as it doesn't even have C#'s using thing.
<jamesd-> Luckily most of the time all you care about is memory being reused, and GC is just fine for that.
<_m_> noam_: believe me, I know the advantages of GC pretty well. I still prefer C++ over Java.
<noam_> so do i
<jamesd-> Hurrah! We all understand GC! Let's sing a song and hold hands.
<noam_> what song will that be?
<jamesd-> Um... the British National Anthem? The Chicken Song? Jilted Kid John?
<jamesd-> Quite a garbage collection there.
<cynic> gc just makes me uneasy and gui apis keep on trying to emulate it as if SWI was a great hit
<cynic> I mean java-like gc
<binrapt> Wow, you put two spaces after question marks and exclamation marks!
<_m_> You apparently don't, binrapt!
<noam_> SWI?
<binrapt> I apparently do not, indeed!
<jamesd-> Old habits die hard.
<cynic> standard whacked interface
<noam_> that comparison table says that c++ has "modules"
<jamesd-> Walter's D vs C++ comparison table? Has he made it any less misleading recently?
<binrapt> He has not
<cynic> it was insane years ago, is it still the same?
<binrapt> I suppose so
<jamesd-> Ah well. Shame, D seems decent, I'd think some integrity in marketing it wouldn't hurt.
<cynic> trying to dis c++ for weaknesses that don't exist is bad marketing imo
<jamesd-> I might prefer it to Java or C#, even though both do now have some form of generics and automatic boxing/unboxing to save my eyes.
<jamesd-> D leans towards building more into the language, which avoids some of C++'s warts. C++ aims to make it possible to do almost everything in libraries, which is powerful but not perfect.
<noam_> so C++ doesn't "support all c types"
<jamesd-> noam: C++98 doesn't support all C99 types.
<jamesd-> If D does, that's a pleasant surprise.
<cynic> and macros
<jamesd-> Though why it would have _Bool is a mystery to me. Or _Complex. Ugly names.
<cynic> at least they follow a naming convention
<noam_> C has "direct access to C"
<_m_> cool, isn't it?
<jamesd-> C is *really* good at interoperability with C. So long as your compilers agree on an ABI.
<noam_> amazing, how they've implemented that
<noam_> java and c# use non-existent debuggers
<mauke> Perl is even better at interoperability with Perl; all Perl implementations use the same ABI!
<cynic> all perl implementations belong to
<_m_> Java is ever betetr at interoperability with Java. All Java implementations use compatible machines.
<mauke> even gcj?
<_m_> What's gcj?
<cynic> java interpreter for gnu compiler collection
<cynic> or is it compiler
<mauke> it can generate native code
<_m_> I didn't touch Java in this millennium.
<spsoup> Zeed: for what?
<Kniht> m: http://forum.crystalxp.net/index.php?act=findpost&pid=7448
<Zeed> c++ :)
<Kniht> m: http://sv2.3dbuzz.com/vbforum/showthread.php?s=&threadid=55409
<_m_> calc !tutorial
<CalcMe> !tutorial = C++ is not the sort of thing you learn (well) from a bunch of second-rate tutorials on the Web. The preferred method is to buy a good book (e.g. Koenig & Moo's "Accelerated C++"), get a good C++ compiler or development environment (e.g. Dev-C++), and practice a lot. You could even take a course at your local community college.
<pepsi_> I have a c++ exam tomorrow.. give me something tricky to read up about
<palomer> const_cast
<_m_> calc answer this
<CalcMe> answer this = Given struct A{};struct B:virtual A{};cl*** C:virtual B,A{};cl*** D:B,public C,virtual A{} a; What's the order of destruction of 'a'?
<noam_> er.
<cynic> I haven't seen the point in virtual inheritance yet, so no answer for that diamond inheritance
<noam_> diamond?
<cynic> look it up
<noam_> that's no diamond inheritance.
<palomer> diamong inheritance sounds _evil_
<_m_> This one is more like a piece of coal then like a diamond.
<cynic> oh
<_m_> a/then/than/
<jamesd-> cynic: seen IOStreams?
<cynic> jamesd-: yes, I have it right here
<jamesd-> cynic: iostream inheriting from istream and ostream, which both inherit from... blah de blah...
<cynic> I know there's a dual inheritance there
<jamesd-> And that it's virtual, as there's common data in the base cl***...
<palomer> shift left corresponding to multiplication means we usually read bit strings from right to left, right?
<jamesd-> palomer: read them how you like. It just means that we consider the right-most bit to be the least significant.
<palomer> gotcha
<ismell> hrmmm question, i have a node template cl***, how could i make a node of node pointers ? node<node*> *cities; doesnt work
<_m_> calc q
<CalcMe> q = We are able to recognize a properly asked question even without the word "question" introducing it.
<mauke> ismell: node isn't a type
<_m_> node<node<whatevertype>*>
<mauke> node* doesn't make sense
<ismell> well i want to make pointers to the nodes
<ismell> hrmm im going to have to rethink this
<cynic> std::deque<node
<cynic> *>
<ismell> deque ?
<_m_> cynic: doesn't work. node is not a cl***.
<cynic> cl*** node; // =P
<ismell> yeah im going to have to rethink this
<ismell> template <cl***name Item> cl*** node ;)
<cynic> ah you didn't instantiate it
<jamesd-> ismell: a node<node<foo>*> might make sense.
<ismell> il just point to the data inside the Item
<ismell> node i mean
<ismell> node<void*> *cities; ? that work
<ismell> then i would have to cast it anytime i use it
<ismell> meh nm
<Mirza> cout << line << ' ' << ul << endl; ... line is a string, ul is an int.. result: " 6sconsin" ... what's going on here? :S
<Mirza> if i follow the string by and endl it looks okey..
<cynic> Mirza: the output order is probably unspecified until you flush
<mauke> no
<mauke> line contains \r
<cynic> but I have seen it do that without a \r
<_m_> hehe, someone opened a text file in binary mode.
<jamesd-> endl does flush. A control character in line is likely, like character code 13 at the end.
<Mirza> crlf stuff? will opening it in text mode automagicfix all?
<_m_> If you're on a crlf system, yes.
<jamesd-> Mirza: in text mode, a newline (however your system represents it) will be read in as '\n'.
<Mirza> i'm on a mac, the file is from a linux i think
<jamesd-> No magic. Just translating between the OS representation and the C++ representation of newline.
<jamesd-> Linux doesn't use CR (character 13) in text files.
<_m_> text mode makes the iostream implementation automatically convert between the internal representation of end-of-line (\n) and the external representation (system defined).
<jamesd-> The Mac can normally cope with it.
<Mirza> alright, text mode baby
<Mirza> what a weird output tho.. what does /r mean that makes it like that?
<jamesd-> Mirza: carriage return means "go to the left margin".
<_m_> \r is for CR, carriage return. Think of an old typewriter.
<Mirza> oh, so it does that without skipping to the next line :D
<_m_> yup
<Mirza> neat...... :X
<_m_> And so you lost the Wi of Wisconsin.
<palomer> omg
<Mirza> alright, thanks, makes sense
<palomer> vector<> doesn't have push_front!
<_m_> Didn't I tell you yesterday to sort the vector backwards?
<palomer> I wasn't here yesterday
<_m_> (maybe, I told someone else)
<cynic> maybe he's not that sort of a person
<_m_> palomer: you might want to consider std::deque instead of std::vector.
<Mirza> hm so how do i specify text mode when using ifstream.open? there's just ios:binary listed..
<cynic> you remove that
<Mirza> i just used open(name) before.. it puts binary as default i guess? so just open(name, ) ?
<Mirza> ios:in
<Mirza> ?
<cynic> you remove the comma as well, else you get a syntax error
<jamesd-> pcalc guess
<jamesd-> Hmmmmn.
<palomer> gah, when has it been so hard to move between ints and vector<bool> ?
<jamesd-> gotta get work account fixed up with CalcMe.
<jamesd-> palomer: why on earth are you using the-evil-vector-bool?
<ismell> std::cout << "oCity Now: " << oCity->data << " At:" &oCity->data << std::endl << std::endl; if oCity is my node, and data is my City, couldent i just add the & to get the address of data ?
<ismell> &(oCity->data) doesnt work either
<_m_> Mirza: if you don't use ios::binary then you'll have text mode.
<cynic> ismell: you are missing another operator<<
<Mirza> i didn't use that, and it gave me my little output problem :/
<ismell> oh wtf
<ismell> hahah
Return to c++
or
Go to some related
logs:
beginner
Kurt Cobain
football
|
|