| |
| |
| |
|
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 27
<lordpil> the hell
<lordpil> we have a JeffK now?
<cons_> how would i convert
<JeffK> ye, you got a JeffK now
<JeffK> waht does it mean to yuo?
<cons_> a struct called sockaddr_in to const sockaddr?
<noam_> cons_: la illaha illa la
<many> sa ftw
<noam_> etc
<Cowmoo> kthx
<cons_> la illaha illa la?
<cons_> wtf
<Cowmoo> is it amateur hour or something
<JeffK> i do proffesional c++ stuff, currently im working out how to do a for...next llop in cplusplus
<JeffK> really hard sutff
<JeffK> stuff i mean, sorry im drunk
<cons_> lol
<JeffK> for(int yuck=0;yuck<>100;yuck+=1) ****it waths wrongo?
<cons_> yuck ++?
<JeffK> dont you name your vars this way?
<ZorbaTHut> jeff, what's <>?
<JeffK> oh, got it, !=
<xdank--> != lol
<ZorbaTHut> there ya go. :)
<JeffK> hmm (searching for excuse)
<ZorbaTHut> there is no excuse
<ZorbaTHut> only Zuul
<JeffK> oh, eh, its the beer, all is the beers fault
<facehuggr> And the smell of newbs in the morning
<JeffK> hmm, how do newboos smell?
<JeffK> like i do? then they stink like my running shoes
<vamecum> I've got a on exit, crash bug, something gets deleted that causes things to breake. anyone got hints on what I should be looking for?
<binrapt> Which cl***es should offer a virtual destructor? All the ones that have virtual/pure virtual methods?
<JeffK> vamecon should look for a debugger
<JeffK> hmm, binrapt wat does virtual mean in context of native c++?
<many> gdb is ur friend
<JeffK> like overridable in .net?
<binrapt> Uhm something like that I suppose
<JeffK> virtual sounds so confusing to me
<JeffK> hmm, i guess yuo should have a virtual desctrucotr if you inherit a cl*** and add some custom child objects to that inherited cl*** that need to be ***kicked when the object is killed
<vamecum> JeffK:It proves a bit hard to debug. heisenbug-like
<JeffK> vamecum tell a butt more about your program
<JeffK> yuo guess its an object not properly freed or something in your messageloop, or ... what?
<upb> if you run it under linux, try valgrind
<upb> its ****ty but manybe it detects you problem
<vamecum> JeffK:well I blame a cl*** called BKMesh , it's basicly std::vector of things like Vector3d to store mesh data.
<JeffK> if os=windumb then i would use mickysofts visualstudio
<vamecum> JeffK:does the std::vector and likes clean themself up in their respective destructors?
<JeffK> vamecum, sorry thats outta my scope
<JeffK> i never used the std namespace objects
<binrapt> Yes, they clean everything up.
<many> vamecum yah they do
<binrapt> Except your mistakes
<JeffK> coz boss always sez, ***hole, tomorrow is deadline for project xy,i am forced to use .net to get a working program as quick as possible
<vamecum> binrapt:clever.
<JeffK> boss sez native c++ is evil, and if i say 'no', he sez: youre fired
<JeffK> anmd then he starts yelling i shall make a doom3 clone with c# that runs 10x fast as in native code
<binrapt> Likely
<james-_> virtual means choose the function to call based on the dynamic type of the object it's being called for. std::vector's destructor does the right thing, i.e., releases resources owned by the vector. Beer is good, wine is fantastic.
<[eloi]> james-_: beer is fantastic if you find the right beer :)
<james-_> eloi: fair point. I'm busy testing Belgian beers this year, mostly.
<JeffK> all the day?
<JeffK> ever tested german beer?
<james-_> Certainly.
<james-_> Had a bad German one recently. St Pauli Girl or something.
<JeffK> i guess german beer is the bbest
<JeffK> hehehe st pauli girl, funny
<james-_> Normally the Germans do a decent job. I'm biased to think British beer is better.
<JeffK> ***** beer from hamburg
<JeffK> the british ones are more bitter i think
<JeffK> i currently drink the worst beer ever
<JeffK> 3.5 euro per gallon
<james-_> sounds nasty.
<JeffK> rite, it is
<JeffK> but i helps me fallin asleep after a a hard day of doing for(;;) loops
<JeffK> boss sez, why isnt yuor for(;;){printf("ble");} loop ever end, and where is the counter variable, i say, its hidden in some code, that, eh, oh, the fone rings, i needto answaer this call...(i have no ****in clue why the lop never ends)
<[eloi]> james-_: fridge is loaded with belgian atm :0
<[eloi]> er, :)
<james-_> The fridge is in good health.
<khaladan> how do you guys usually name your accessor methods?
<james-_> khladan: sparingly. Member functions should normally tell objects to do things... "tell, don't ask".
<james-_> khladan: but I avoid silly prefixes like "get".
<oxix> same here
<binrapt> German beer is ok
<JeffK> istn get_Ble() common for accessors?
<JeffK> or getBle()
<JeffK> or waht?
<oxix> i prefer just ble()
<james-_> ble().
<upb> i heard g_E_t_B_l_e() is
<binrapt> Czech beer!
<james-_> getFoo() often indicates that someone is thinking in terms of blobs of data, not objects.
<khaladan> so you call the data something other than blee?
<JeffK> hmm, sounds commmon to the discussion of nameing private member variables
<JeffK> m_Ble, _ble ...
<james-_> Czech beer is good. Budweiser and Pilsner Urquell are faves.
<oxix> I like ble_
<JeffK> hmm, budweiser is good
<james-_> m_ble or ble_ are reasonable, IMO. _ble is less so. Too close to being invalid C++.
<oxix> not big on the m_ble and really don't like _ble
<james-_> Not the US stuff :)
<khaladan> i need added behavior for getting and setting, so, that's why i have accessors
<oxix> i like ble_ since I read from left to right and the pollution is at the end
<binrapt> Budweiser Budvar
<JeffK> i recently read some microsoft paper where they said m_Ble would be obsolete, we shall use _ble, hmm, maybe its a brainwash action thing, im confused
<james-_> khaladan: are you really "setting" then? Are you not telling the object to do something? In which case, wouldn't a function name that says what you're telling it to do be better?
<binrapt> Czechvar
<JeffK> if i start using _ble, im going to buy ms-booze, ms-diapers and billy-gl***es n stuff
<upb> the best is to make a structure named members in each cl*** and put the member variables inside it
<oxix> could do: ble_m
<oxix> that's new
<oxix> and it's stylish
<james-_> Czechvar, branded to look like Budvar.
<khaladan> james, so, should i call my function something like "set foo and then notify all subscribers that foo changed"? setFooAndNotify?
<oxix> and it supports my left-to-right pollution at the end preference
<khaladan> i'd really rather not
<james-_> khaladan: updateFoo might be better than setFoo there.
<oxix> ble_m; ble2_m; foo_m;
<JeffK> but the best is i got a coworker that mixes german and english language to build the names of member variables
<oxix> int size() {return size_m; }
<oxix> err int size() const
<JeffK> oh, sorry, ex coworker, i told boss to fire him
<oxix> actually I think I can warm up to that
<oxix> _m suffix
<oxix> I find it more attractive than m_
<oxix> unsigned size() const {return m_size; } vs unsigned size() const {return size_m; }
<JeffK> uhm, correction, i told boss to fire my second coworker coz he's too slow, goes to the doctor ever tuesday 3pm and does make these annoying yawn noises allday, wich i dont like
<khaladan> and i prefer getters since something might be a calcuated value
<ZorbaTHut> jeffk, your namesake is funnier than you.
<JeffK> and last but not least: he drinks his orange juice VERY LOUD
<oxix> I think if I glance at one of these for 10 milliseconds, I would more likely recognize the 'size' part in size_m over m_size
<JeffK> ZorbaTHut, im not the JeffK you think, im Jeff Klotzki from germany
<oxix> and that's the important part, I think.. for the logical meaning
<binrapt> ZorbaTHut, ditto
<binrapt> i meen dito lol!!1
<JeffK> yoyo, i got it, IM NOT FUNNY
<JeffK> but that not my intention
<oxix_m> m_size.. you gotta read past the m_ first to even get to the size part
<JeffK> i just wanto release some pressure
<oxix_m> and if you see a whole row of variables declared with m_ - that m_ part just pops out a whole lot
<JeffK> maybe yuo know similar ****: i did approx. 60 hours per week the last 8 weeks to get our stuff ready for the 'systems 05', my ****n coworker did exactly 40 hours a week, despite of knowing his project had to be finished last monday, now i have to fix his **** until friday so boss can show it on the mess, coz my collegue is too dumb
<JeffK> im so ****n pissed about that
<oxix_m> you just want to keep looking at it like a naked lady and then you forget what the variables are even supposed to stand for since you are just thinking m_ m_ m_
<james-_> So get off IRC, get it done, and then take a break.
<JeffK> nex week he can drink his damn orangejuice at home again
<oxix_m> like with oxix_m.. the oxix is the part that pops out - the _m is just an afterthought
<binrapt> Hmm I do not really see the point of all those member prefix/suffix stuff
<binrapt> s/those/that/
<james-_> binrapt: I don't always use them, but storing to a member variable is a bigger deal than writing to a local -- it persists after the function ends, obviously. It can just make things easier to read.
Return to c++
or
Go to some related
logs:
undernet ipv6"
"licota" +bilde
show derss
xorg faster rendering
chatzone
|
|