@# Quotes DB     useful, funny, interesting





Google
 
Web www.quotesdb.info
Undernet  |  EFnet  |  Quakenet  |  Freenode  |  Dalnet  |  Ircnet  |  Galaxynet
Page: 1 2 3 4 5 6 7 8 9 10 11



Comments:

<0> of course my code works
<0> i wouldn't paste it here without trying it first
<0> what do you mean "that isn't supposed to be done" ?
<0> it's well known that if you don't write an operator=, the compiler generates a default one that does a member-by-member ***ignment
<0> that's called convenience
<1> rdragon could you please take a quick look at my code to see what i'm doing wrong?
<0> paste it to the site
<1> errr. it's a little chunky
<1> like.. there's more than 1 file
<1> and i can't summarize sanely
<0> well, i'm not going to dig through mounds of your code
<0> if you separate the actual problem you're having out into a tiny program, you'll probably solve it yourself
<1> hmm. let me try
<0> i don't know what the problem even is
<1> alright.. i have 2 cl***es, Tensor and Transform. Tensor implements operator= in the public scope. Transform inherits Tensor, and within the constructor for Transform, i would like to write "*this = arg1", where arg1 is a static variable in the global scope
<0> what is arg1?



<0> and why ***ign something to the object from the constructor? that doesn't sound like a great design
<0> what type is arg1?
<0> Transform?
<1> its a static variable i declared.. ok that is not the actual name. it is IDENT_MATRIX, and it is a Tensor
<0> are you sure you don't want to use a copy constructor to do that?
<1> i am building a transformation matrix out of an identity matrix.. so it initializes the values of the transform to look like the identity
<0> so Transform represents a matrix?
<0> what does Tensor represent?
<1> yeah, i am sure i do not want a copy constructor, because i am building a rotation matrix
<1> Tensor = matrix
<0> what about
<0> Matrix m;
<0> BuildRotationMatrix( m, some_rotation_value );
<0> or just a function
<0> Matrix BuildRotationMatrix( float some_rotation_value );
<1> i prefer to have a separate cl*** represent rotation matrices :)
<0> why?
<0> a rotation matrix is no special kind of matrix
<0> it's just a matrix
<0> it's just the values you put inside that happen to make it rotate another matrix when you apply one to the other
<0> all you need is a convenience function that sets up a matrix to have the effect of rotating by some value when it's applied
<1> that may be true, but it doesn't answer the question as to why operator= doesn't work for the Transform cl***
<0> because you're ***igning a base cl*** type to a derived type
<1> in your code example, u used cl*** B for both objects..
<0> exactly
<1> here, i see i am ***igning the base cl***
<0> you're ***igning an object of A to an object of B
<0> that doesn't work unless you write the code that makes that happen
<1> however, shouldn't the base definition for operator= recognize exactly the base ***ignment RHS
<1> ?
<0> no, because that's not inherited
<0> the signature for operator= for B is B& operator=( B& )
<2> Duck and cover would actually help, khan
<0> an A isn't a B
<0> so it doesn't fit
<0> a B is an A, so you could do the reverse, and ***ign a B to an A
<1> I SEE!
<1> i think i see the light
<0> but I disagree that you should subcl*** Matrix, still
<1> do u want to know something even sicker?
<1> i derived cl*** Vector from Matrix
<0> well that's just stupid
<1> haha
<1> i knew u would say that
<0> I hope you're just joking
<1> but i'm recognizing the fact that matrix and vector are both tensors
<1> and they share the same basic principles when it comes to the dot product
<0> you can implement everything just fine if vector and matrix are completely separate types
<1> and transposing
<0> why are you writing all of this yourself, anyway"?
<1> hmm. because it gives me more control
<0> control of what?
<0> it's not like you're implementing them better
<0> are you using this stuff for graphics?
<1> so i can inherit vector from tensor, just the way i like
<1> cuz no one else does it this way :P
<0> probably because it doesn't have any benefit
<1> yeah, i do graphics
<0> what graphics library are you using to render this stuff, and why aren't you using the matrix/vector cl***es in the library?
<1> but this time i'm trying to rewrite my old library from the ground up
<1> i'm rolling my own



<0> oh, you've written your own; I should have guessed
<0> i suppose you're writing an MMORPG too
<1> thanks for the help tho, i don't have good c++ foundations
<0> then you shouldn't be writing a graphics engine
<0> that's a job for a pro
<1> haha
<1> no, i'm writing a thesis on robot control
<1> and i need to have more control over my code
<0> then why are you wasting your time writing a graphics engine?
<1> it's not for a graphics engine
<0> a math library, then
<0> boost has a linear algebra library in it, by the way
<0> and I bet there are a bunch out there that would work perfectly for what you need
<0> the boost library is called uBLAS
<1> that's true, but i just wanted to take this opportunity to give my own code a lil facelift anyhow
<0> http://www.boost.org/libs/numeric/ublas/doc/index.htm
<1> i'm using this lib more for the math, but i don't want to use matlab because i want to implement a standalone program
<0> look at that library
<0> all of the math is done
<1> oh wow. they've got stuff for sparse matrices
<1> i might need that
<1> do u use this library?
<1> i used to write my own linked lists.. long ago. i never understood why.
<1> it felt stupid once i discovered stl
<1> but right now my compiler doesn't like BOOST.. so i guess it rules out that option
<1> i am on vc6, and reluctant to upgrade. i still use a voodoo card too :)
<3> Can I declare two variables of different types within a for() loop ?
<2> voodoo card, haha
<4> bacl
<5> threat not sure what you're asking
<5> dextre what compiler are you using?
<2> 6.0
<0> yeah too bad he left
<5> which one? the guy w/ the ancient compiler or the other one?
<0> dextre
<0> my pizza got delivered so i stepped away heh
<0> i found apartments that I can walk to work from
<5> that's not real useful as you're moving in a week
<0> I mean where I'm moving to
<5> ah
<0> will have to call them tomorrow, hopefully they've got some stuff available
<6> <@JBlitzen> Duck and cover would actually help, khan
<6> yeah, you'd be the one to believe that heh
<6> bedtime w00t
<2> It's not just me
<2> Any idiot knows it would help
<2> Where are you, Baltimore?
<2> http://meyerweb.com/eric/tools/gmap/hydesim.html?ll=-96.7940,32.7838&yd=1000
<2> If you're standing in Arlington cemetary during that sort of detonation, you're in for 2-5 psi
<2> ***uming you survive the thermal effects, which is likely, exactly how will ducking and going for cover not help avoid debris?
<2> I mean, that's just common sense
<2> Look at that map and it's quite clear that, sure, people within a couple miles of ground zero are in for a bad day, but there are a whole lot of people in areas that will be affected but not vaporized
<2> Duck and cover was never for the people at ground zero, it was for everyone else
<2> But idiots like Jon Stewart latch on to it, are totally oblivious to the realities, and convince all the other idiots to laugh about it.
<6> wow...1 megaton
<6> we **** more than that these days
<6> and in the 50's, when this program was popular, we were testing hydrogen bombs
<6> 10 times as destructive
<6> god knows what we can do know
<6> now
<7> lol
<6> also, the footage john steward was particularly ridiculing was the footage of school children in the streets ducking and covering when they saw the flash
<6> good chance you're ****ed
<2> No, the average Russian warhead is a 450 kT device
<2> That's for a ground burst, too, ala terrorist attack
<2> We can do about 25 mt now, but why bother
<0> happy april fools day
<2> Hi rd
<0> hey
<2> The average nuclear device in any inventory is under a megaton
<2> I don't recall what our's are
<8> this is really wierd my purly abstract cl*** (just an interface) is not being enforced. I inherit this abstract interface as public yet its purley virt. methods are not enforced in the child cl***es
<0> Enferex are you sure they're pure virtual? and they're not defined?
<8> Not defined


Name:

Comments:

Please enter the result of the sum 63 + 46 (to avoid spam):






Return to #c++
or
Go to some related logs:

gentoo xkeyboard
#MissKitten
#london
remix lebnani
#linux
#php
jermas xxx
patit B
serial telebubbies
#teens



Home  |  disclaimer  |  contact  |  submit quotes