class Ship { public: static const int TEXTURE_WIDTH = 32; static const int TEXTURE_HEIGHT = 32; static const int VELOCITY = 640; //Constructor and deconstructor Ship(); //Player event handle void handleEvent(SDL_Event& event); //ship core void move(float timeStep); void render(); bool hit(std::vector<Ship>& ship, int index); //Get Ship info float getX(); float getY(); bool getGrace(); void setGrace(); void unsetGrace(); int getEnergy(); SDL_Rect getBB(); private: float mShipX, mShipY; float mSpeedX, mSpeedY; bool mGrace; int mEnergy; SDL_Rect mCollider; };

class Asteroid { public: static const int TEXTURE_WIDTH = 32; static const int TEXTURE_HEIGHT = 32; static const int VELOCITY = 640; //Constructor and deconstructor Asteroid(); //Asteroid core void move(float timeStep); void render(); void respawn(); void hit(std::vector<Asteroid>& astrd, int index); //Get asteroid info float getX(); float getY(); ASTEROID_STATUS getSize(); SDL_Rect getBB(); private: float mAstX, mAstY; float mSpeedX, mSpeedY; ASTEROID_STATUS mSize; SDL_Rect mCollider; };

I have one vector made of 1 item and another vector made of 20 items, vector.erase(vector.begin() + index) works fine on the vector made of 20 items but not on the vector made of 1 item

Are you sure the index is correct in the second case?

this works, the other doesn't

Bro, if u can use debugger and try to see what is failing

Are you sure the index is correct in the second case?

yes, I've tried manually setting it to 0 and even removing it. as I said the vector is made of one item of index 0 and .begin() should delete it

Bro, if u can use debugger and try to see what is failing

debugger doesn't seem to show anything particular, it executes the method and does nothing else

Look Index variable, and its value

I'll try to see what happens to the asteroid element when it gets deleted

I guess it's sloppy coding error. Your mEnergy must be greater than zero all the time.

I guess it's sloppy coding error. Your mEnergy must be greater than zero all the time.

well actually

the method .erase() gets called

but doesn't do anything

the method .erase() gets called

is your index within 0..size-1 range?

is your index within 0..size-1 range?

index is hardcoded to 0 for the ship

which meanse erase has ship.begin() for argument

index is hardcoded to 0 for the ship

And your vector isn't empty?

no, the vector only has one item

actually

So weird mmmm

the vector after the first call becomes made up of 1000 items

the vector after the first call becomes made up of 1000 items

whut?)

first erase: the vector becomes empty but I can still control the ship; second erase: vector fills 1000 items with garbage data (and the first item still works)

first erase: the vector becomes empty but I can still control the ship; second erase: vector fills 1000 items with garbage data (and the first item still works)

So you have a pointer to it doncha?

Also calling erase on an empty one is ub

So you have a pointer to it doncha?

I'm not really using pointers, I'm using vector with operator[]

like this

I'm not really using pointers, I'm using vector with operator[]

So you are using operator[] on an empty vector?

asteroids just disappear, the ship doesn't though

🙈

So you are using operator[] on an empty vector?

I have a bad news for you

Did you overaloading the operator [ ] ?

I have a bad news for you

what

what

It's UB

maybe I got what happens, but I still don't understand why it doesn't disappear like the asteroids

well maybe i am stupid

It's UB

What is UB btw?

and i'm rendering it regardless lol

Looks like shooting yourself in the foot

What is UB btw?

undefined behaviour

Looks like shooting yourself in the foot

why?

anyways it looks like I should be more careful with vectors

maybe I got what happens, but I still don't understand why it doesn't disappear like the asteroids

Cuz if you freed the memory it doesn't mean that there are not leftovers on that pointer.

It's use after free basically

why?

Because you're trying to erase the same object on which you're calling hit

anyways it looks like I should be more careful with vectors

Just don't access empty one via operator[] lmao

for some reason I was thinking it wrong, like the program was making disappear. in reality I had a for cycle at the rendering section that wouldn't render if the vector was empty, likewise all other operation with the asteroid vectors had this check, while the ship didn't have one and I didn't realize.

Because you're trying to erase the same object on which you're calling hit

Even if it would not be so - it looks just as a erroneous design.

What I would do is to a have a boolean whether the object is empty

And if it is, than I won't update it/render it

What I would do is to a have a boolean whether the object is empty

yes that's what I am implementing right now

for some reason I was thinking it wrong, like the program was making disappear. in reality I had a for cycle at the rendering section that wouldn't render if the vector was empty, likewise all other operation with the asteroid vectors had this check, while the ship didn't have one and I didn't realize.

Also do you have like a separate size variable for the vector?

Also do you have like a separate size variable for the vector?

yes

Instead of using for each loop

yes

consider this: for(auto& a : vec) { // do stuff a.render(); }

consider this: for(auto& a : vec) { // do stuff a.render(); }

what does this do?

what does this do?

It's for each loop

with a being an element of array

And all the bounds are checked

struct N { constexpr N() {} N(N const&) = delete; }; template <typename T> consteval void assert_copyable() { T t; T t2 = t; } using check = decltype(assert_copyable<N>());

I think it was illegal at some point, to increase consistency they decided to align consteval more with constexpr and not evaluate in this context. P1073R3: Immediate functions P1937R0: Fixing inconsistencies between constexpr and consteval functions

How to install tarbo c in laptop

Thank you

How to install tarbo c in laptop

#rules you shall not and instead use something modern,

Hello friends

Hello friends

Hi

Finee

I think it was illegal at some point, to increase consistency they decided to align consteval more with constexpr and not evaluate in this context. P1073R3: Immediate functions P1937R0: Fixing inconsistencies between constexpr and consteval functions

Yeeees My quiz was outdated When I was writing it I thought that it should be a compile-time error But while researching the problem I found this proposal and now it's legal

Any good and easy to make project Ideas ??

A kernel?

>easy to make >Kernel Hmmm

A simple kernel is not that hard to program.

Well it depends on how deep you want to go with your kernel, but still

@OxFFFFFFFF hey there is a person with username "$" you have restricted him for writing here , could you please check that , and unrestrict him

@OxFFFFFFFF hey there is a person with username "$" you have restricted him for writing here , could you please check that , and unrestrict him

Thnx 😊

A simple kernel is not that hard to program.

++

Any site where I can check the posix thread scheduling attribute related program practice? Got only 2 program on IBM site in google

In c programming

gdb

gdb

Not so.

gdb

Bt little bit experience