| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Comments:
<0> Oh dear
<0> Your_Dadd: Get the book "Accelerated C++" by Koenig & Moo
<1> I recommend the books at http://rudbek.com/books.html
<2> No grammar skills whatsoever, either.
<1> the tutorials online are not very good
<3> Some things have to be hidden deep, deep, down inside some function... like this. Edge edge(*this, callee); this->edges_out().push_back(edge.get_iter()); const_cast<Function&>(*callee.get_iter()).edges_in().push_back(edge.get_iter()); // Does you momma know you're looking at this?
<4> ya ok thxs
<3> The bad part of that code is that you can't see HOW bad it is.
<3> lol
<3> Hmm, I was so surprised that everything compiled - it took me two minutes to realize I now had to start the program.
<5> maybe don't run it because it might break your computer
<3> It didn't work :/
<3> Parsing function declarations........................................................................................................................................................................................................................................................................................ done.
<3> That worked yesterday
<3> Found 192 different cl***- or namespaces.
<3> Found 0 cl***es with constructor.
<3> Zero?!
<5> is that a C++ parser or something?
<3> yeah, something like that.
<6> heh
<3> A C++ analyzing cockroach ... can't work.
<5> what's it going to do?
<3> Drawing graphs, visualizing large C++ projects.
<5> that is fricking sweet... I seen something like that, but for MIPS ***embly in like some wirless card this guy was reverse engineering
<6> yeah but Runs doesnt work... yet :)
<5> heh
<5> I wonder what the f' is up with my ISP... I never get disconnected, but now it happens all the time. wtf
<7> can anyone take a look at this ***ignment operator implementation and tells me if it's correct (for doubly linked list)
<7> http://www.noidea128.org/sourcefiles/15869.html
<5> for templated things, the function has to come inside of the cl*** outline... having it in a separate file will result in problems down the road
<0> a.) What's wrong with the standard library? b.) At least mimic standard iterators, if you won't use the standard library. c.) What happens to the existing data on you're replacing in the ***ignment operator?
<0> Anyone know of any decent and inexpensive web cache/proxy software?
<0> I'm having to share a rather slow fractional T1 in a school computer lab.
<7> I have it inside the cl***
<7> I'm just posting one function
<7> the ***ignment operator
<7> is it a proper way to do it?
<5> it doesn't look like it's in the cl***
<7> oy
<7> it's in the .h file :S
<3> I fixed it :)
<0> isaacmk: It's better to say "inline" or "in the header"
<3> Found 154 cl***es with de- or constructor.
<5> is it defaultly inline if it
<5> 's in the header?
<3> no :)
<5> what about the cl*** that's in the header?
<0> From a syntactical point of view, yes. Of course, the optimizer will still do what it wishes.
<7> the function definition is in the header
<3> no way - you can put anything in a header, but the compiler won't even know.
<3> Only the prepocessor knows what is in a header, and that doesn't understand C++
<0> template <typename T> cl*** Foo { public: operator=(Foo<T> const& rhs) { /* ... */ } }; // I prefer to see this
<1> Run I sorta vaguely remember a "simple" way to make a collection of "keywords" in Spirit
<7> but I mean in the implementation body not header is it a good way to do an ***ignment operator?
<0> template <typename T> cl*** Foo { public: operator=(Foo<T> const& rhs) }; template<typename T> inline Foo<T>::operator=(Foo<T> const& rhs) { /* ... */ } // this will do
<1> that might solve your "I found int at the betinning of integrate"
<3> vawjrwrk: distinct_parser<> keyword_p("A-Za-z0-9_");
<3> But that is five times slower as what I use now.
<0> Ton-iK: Aside from where it is, see my previous three comments.
<0> "[11:28] <@Solamente> a.) What's wrong with the standard library? b.) At least mimic standard iterators, if you won't use the standard library. c.) What happens to the existing data on you're replacing in the ***ignment operator?"
<1> I was suggesting a list of keywords, since there arent that many
<3> I now use:
<0> Probably the most annoying thing about that ***ignment operator, to me, is the letter 'C' at the beginning of the cl***. That's a complete waste of a character.
<3> contiguous<sequence<strlit<char const*>, difference<epsilon_parser, alternative<alpha_parser, chlit<char> > > > > keyword_p(char const* str) { return lexeme_d[str >> (eps_p - (alpha_p | ch_p('_')))]; }
<3> My original problem was that I had forgotten the lexeme_d
<3> Hmm, I need my parser to make a list of every type it sees...
<1> maybe I should dig out what I used
<3> Because when a coder didn't define any constructor or destructor (so the compiler is using the default ones), then I won't see any constructor definition.
<3> I DO see, for example: A::f()
<3> So, if elsewhere I see 'A' used as type, then I know it's a cl*** and not a namespace.
<1> what are you writing, btw?
<3> I guess I can use everything for that... the return type of the functions, the parameter lists of the functions and all template parameters that I run into.
<3> Something that should allow one to get an overview of a (large) C++ project.
<3> It automatically finds all 'groups' that exist in the project.
<3> files, directories, projects, objects, cl***es, namespaces, functions that call eachother etc etc.
<3> Once I get all that info in STL containers, I'll make something to show it, or a part of it, with graphviz
<3> My drive is that I want to take out code from libcwd and replace it with libdwarf, but I have no idea anymore where to cut the code in two. I write this in a general way, but it should be able to tell me what the interface/API is that I need to provide to cut out some piece of code.
<1> ah
<3> ie, I tell it: I want this in and that out - and then it finds a minimal 'border'
<3> Hmm
<3> I guess I better replace "<unnamed>" of anonymous namespaces with "<unnamed:filename.cc>" ... the question is, which filename to use when the <unnamed> appears in a header file?
<1> it's unclear what namespace {int a;} namespace {int a;} in a file will do
<3> I suppose there are two possibilities: 1) No coder in his right mind would use an unnamed namespace in a header file, or 2) if they do, then nobody will access variables inside it in some .cpp file, so I might as well use the name of the headerfile to distinguish the namespace from other anonymous namespaces.
<1> or what it really "means" w/ respect to the standard
<3> I thought that meant the same 'a', I doubt it even compiles.
<3> I'll issue a warning and then use the name of the headerfile I think.
<3> dang, that means I need to know what are headerfiles and what not :/
<3> I guess I won't give a warning then ;)
<1> lol, good plan
<8> Does that mean you kicked him for no reason?
<9> The reason is obvious.
<8> I was being sarcastic.
<3> Painfully obvious.
<9> He already did that 50 minutes ago.
<10> hmmm
<3> Someone say something and have this scroll... it's burning into my monitor.
<9> something
<10> what?
<10> who said that?
<3> three more
<3> two
<8> _m_ did
<3> one actually
<10> yay its gone :)
<3> here too :)
<9> You've got a small window.
<8> It's not gone here yet
<3> you 19" lamer
<8> I have to go back to work anyways
<9> ~45 to go.
<8> I've got about 10 more
<10> i've about 5 lines window open
<3> yeah really, 45 ?!
<9> This window is 75 lines.
<3> I only see 16 lines, more than enough.
<9> problem solved :)
<8> one more
<8> gone :)
<10> irc isnt that interesting for a full screen window
<11> if the background is white and the fonts too no line actually :)
<3> He has only a large irc window to hide the nakked gurl on his desktop, from his boss.
<3> You could use three little irc windows...
<11> white is tiring for the eyes, anyway.
<3> China is attempting to deliver a spam email here, once per second.
<3> It's time to stop even logging them :/
<10> all of the chinese?
<3> most are from 221.218.*
<10> that'd be 6 billion :)
<3> 221.218.116.224
<11> :)
<3> That IP apparently doesn't get it that I refuse their connection.
<2> Hm.. 221.218.116.224 resolves to the CIA
<1> it's a shame the U.S. consideres it terrorism to DDoS them
<3> TCA!
<3> TCIA ?
<2> Hi Run. :)
<2> Uh.. yeah. The I is silent
<3> Maybe they want to warn me, or hire me... all the same, it remains spam in those quantities.
<2> The Central <whisper> Agency
<3> lol
<3> Ok, how to have a program embarress itself:
<3> Found 84 different cl***- or namespaces.
<3> Found 83 cl***es with de- or constructor.
<3> That leaves 1, likely, namespaces.
Return to
#c++
or
Go to some related
logs:
no job is done till the paperwork is done
hardfucking
RopeAssaultProtection
#java
#chatzone
javascript format=###,###,##0.00
#chatzone
12What mythyical Scottish town appears for one day every 100 years
#windows
#MissKitten
|
|