Hi,

I have perhaps a relatively simple question about breaking out of a for loop.
​My current code does the following:

- loop through a std::vector of input contexts
​- if the requested button can be mapped as an action or state in a context, the for loop stops.
​So basically I map the button to an action, in the 1st context the loop runs into and then stop (a button can only lead to 1 action, in the 1st context it finds a mapping). The contexts are sorted on priority [0] being the highest priority.

The code:

void CInputManager::SetRawButtonState(const RAW_BUTTON pButton, const bool pState, const bool pPrevState)
{
	for(auto &context : mContexts)
	{
		if(context.IsActive())
		{
			if(context.MapButtonToAction(pButton)) break;
			else
			{
				if(context.MapButtonToState(pButton, pState)) break;
			}
		}
	}
}

Now, what I want to do is add another check:
​- I only want to try to map button to action in a certain condition (pState = true, pPrevState = false)

After trying to add that to the code, my mind gets confused, is it correct how I tried this, or is the theory on the 'break' no langer valid (because of if statement within if statement).

Back to the basics: I think that the break simply takes me out of the for loop. If so, all is good.
​Can someone confirm this or tell me where my mind confuses me? :)

The attempt:

void CInputManager::SetRawButtonState(const RAW_BUTTON pButton, const bool pState, const bool pPrevState)
{
	for(auto &context : mContexts)
	{
		if(context.IsActive())
		{
			bool actionDone = false;
			if(pState && !pPrevState)
			{
				if(context.MapButtonToAction(pButton)) actionDone = true;
			}
			if(actionDone) break;
			else
			{
				if(context.MapButtonToState(pButton, pState)) break;
			}
		}
	}
}

void CInputManager::SetRawButtonState(const RAW_BUTTON pButton, const bool pState, const bool pPrevState)
{
    for(auto &context : mContexts)
    {
        if(context.IsActive())
        {
            if(pState && !pPrevState)
            {
                if(context.MapButtonToAction(pButton)) break;
            }
            if(context.MapButtonToState(pButton, pState)) break;
        }
    }
}

FWIW I believe this should perform the same and I find it to be simpler to read.

edit: and if you want to compress it further:

void CInputManager::SetRawButtonState(const RAW_BUTTON pButton, const bool pState, const bool pPrevState)
{
    for(auto &context : mContexts)
    {
        if(context.IsActive())
        {
            if(pState && !pPrevState && context.MapButtonToAction(pButton)) break;

            if(context.MapButtonToState(pButton, pState)) break;
        }
    }
}

Thanks all.

@Ferrous: that's an option too, indeed. Basically breaking out of the for loop will result in the same as a return (in this specific case), because the function stops after the for loop.