Implement ALC.getStringList()

[ACrazyTown]

Fixes #1986

There are certain OpenAL parameters, meant to be passed into ALC.getString(), that return a list of strings. For example, ALC_ALL_DEVICES_SPECIFIER to get the names of available playback audio devices or ALC_CAPTURE_DEVICE_SPECIFIERto get the names of available capture devices.

ALC.getString() assumes that the result will always be a standard C string with a single NULL character at the end, but this is not the case when passing the params mentioned above:

An alcGetString query of ALC_DEVICE_SPECIFIER or
ALC_CAPTURE_DEVICE_SPECIFIER with a NULL device passed in will return a list
of available devices. Each device name will be separated by a single NULL character and
the list will be terminated with two NULL characters.
(https://www.openal.org/documentation/openal-1.1-specification.pdf)

This PR introduces the ALC.getStringList(device, param) method which handles these lists properly and returns them as an Array<String>. If list handling is unnecessary for the param provided, the result will simply be [ALC.getString(device, param)]

[KarimAkra]

This looks good but a little hardcoded..?

I personally went with this:

And we just split the string at the comma on haxe

[ACrazyTown]

This looks good but a little hardcoded..?

Yeah, it's a bit ugly but I had to do that since lime_alc_get_string_list assumes it's dealing with a list and not a standard C-string. I figured it'd be fine since it's a private API meant to be called by Lime internally.

Maybe the check could be done on the C++ side but ngl my C(++) is a bit rusty.

And we just split the string at the comma on haxe

This could've be done as well but I opted in for a seperate method because I wanted to do as much as possible on the C++ side and other OpenAL bindings seem to do something similar.

[player-03]

Don't we need a free() call to go with this?

[dtwotwo]

Don't we need a free() call to go with this?

Maybe hl_alloc_bytes would be safer instead of malloc, since we can't call free() immediately?

[ACrazyTown]

I've added a free() call. I've tested it a bit and it seems to work fine, but I will say again that my C(++) is a bit rusty so please do tell if I've overlooked something!!

I will also note that there's other malloc() calls in the OpenAL bindings that do not have a corresponding free() call. Perhaps this should also be handled in a seperate issue?

lime/project/src/media/openal/OpenALBindings.cpp

Lines 2209 to 2217 in 36d7953

	HL_PRIM vbyte* HL_NAME(hl_al_get_string) (int param) {
	const char* result = alGetString (param);
	int length = strlen (result);
	char* _result = (char*)malloc (length + 1);
	strcpy (_result, result);
	return (vbyte*)_result;
	}

lime/project/src/media/openal/OpenALBindings.cpp

Lines 3389 to 3398 in 36d7953

	HL_PRIM vbyte* HL_NAME(hl_alc_get_string) (HL_CFFIPointer* device, int param) {
	ALCdevice* alcDevice = device ? (ALCdevice*)device->ptr : 0;
	const char* result = alcGetString (alcDevice, param);
	int length = strlen (result);
	char* _result = (char*)malloc (length + 1);
	strcpy (_result, result);
	return (vbyte*)_result;
	}

[Raltyro]

oh, i didnt notice youve made a pull request about this too until no
ive made my own implementation for funkincrew's pull request not long ago.
https://github.com/FunkinCrew/lime/pull/39/changes