#===================================================
#
#This script converts the M1-3 keys to FN keys, rather than toggle keys. I have
#named this MShift, as it emulates the functionality of GShift on Logitech's mice.
#It also converts the current M-key to an extra key, and adds back togglability via
#G-keys in any M-layer, or even while MShifted for the all 30 possible buttons in a
#layer to be re-mapped without worrying about losing access to M-layers.
#
#This script also adds the toggling M-layers to keys, and can be customized to allow
#for a single key to be mapped to enable MShift, and then from within MShift, to be
#toggled-on. For example purposes, the following G-keys are configured as such,
#where the format is Button[M-State]:
	#G24[M1, M2, M3] mapped to M2Shift
	#G9[M2] mapped to M2 Toggle
	#Current M-key mapped to LCtrl
		#(i.e.: M1[M1], M2[M2], M3[M3])
	#And all M-keys that are not the current M-state are now MShifts instead of
	#toggles.
		#(i.e.: M2[M1], M3[M1], M1[M2], M3[M2], M1[M3], M2[M3])
#To use this functionality (and allow the maximum number of buttons available in M1
#by only having MShift mapped, and no "MToggle"), if - in M1 (or M3) - M2 is held,
#and then G9 is pressed, when M2 is released, M2 will stay toggled on.
#This also allows for binding the current M-State's button to an additional key,
#essentially making it G30 (see next comment section for details)
#
#===================================================
#
#This is all that needs to be changed for assigning a keypress to the current
#M-state's M-key;
#e.g.: in M1, if M1 is pressed - an extra re-mappable button!
#To map this, follow the syntax as defined in the Appendix A of
#https://douile.github.io/logitech-toggle-keys/APIDocs.pdf ; i.e.: keep the quotes
#for the keyname, or remove them for the scancode. e.g.:
#MMap = 0x1d
#
MMap = "lctrl"
#
#To map the G-key to toggle between different M-states, you will have to edit the
#code after the following block of comments.
#
#===================================================
#
#G1-22 are labelled, G23 is the left thumb button, and G24 is the bottom thumb button.
#When the joystick is not assigned as a joystick, the click is G25, then starting
#from Up, going clockwise, is G26-G29. Visually, this looks like:
#
#    Click       G25
#      Up        G26
#          Right G27
#     Down       G28
#Left            G29
#
#But I find it more useful in a top-down view of the joystick:
#
#      Up        G26
#Left            G29
#    Click       G25
#          Right G27
#     Down       G28
#
#===================================================
#
#Side note: if you have a Logitech mouse in addition to a device with M-layers (like
#the G13 and G15), you can map functions that are only available from either device
#onto the other, for example: mapping GShift onto the G22 key will allow you to hold
#down G22 with your left hand, and use a single finger on your right hand to press a
#key that is mapped in GShift.
#
#More specifically to my use case, with the mouse emulation the G13 offers, it
#enables horizontal scrolling to be mapped onto the G600 in Logitech Gaming
#Software. (G HUB does allow horizontal scrolling to be mapped natively, but G HUB
#doesn't support the G13.)
#
#===================================================
#
#Add toggling M-States back; MShift disables this functionality by default
#Note: M-states act as if they are constantly held down, until a new M-state is
#assigned. This is why a separate variable, "MShift", in addition to "MKeyState",
#must be assigned. Without this additional variable to compare to, MShift would only
#work in one M-state. Additionally, this enables the possiblility of toggling M-states.
if event == "G_PRESSED" then
	MShiftRevert = 1
	#OutputLogMessage("G_PRESSED!\n")
	MShift = GetMKeyState("lhc")
	#M1 Toggle
	if MShift == 1 then
		#G24
		if arg == 24 then
			#Toggle to M2
			MKeyState = 2
			SetMKeyState(MKeyState, "lhc")
		end
	#M2 Toggle
	elseif MShift == 2 then
		#G24
		if arg == 24 then
			#Toggle to M1
			MKeyState = 1
			SetMKeyState(MKeyState, "lhc")
		#Stay on M2
		elseif arg == 9 then
			#Ends MShift on this layer
			MKeyState = 2
		end
	#M3 Toggle
	elseif MShift == 3 then
		#G24
		if arg == 24 then
			#Toggle to M1
			MKeyState = 1
			SetMKeyState(MKeyState, "lhc")
		end
	end
end

#===================================================
#The following code need not be edited, and is what converts M-keys into MShifts.
#===================================================
#More detailed code explanation
#===================================================
#When MShift is released, the M-key for the current locked state is quickly toggled.
#As a result, the MShiftRevert variable keeps track of when this happens, to ensure
#that whatever is mapped to current M-State is not triggered erroneously.
#===================================================

#Set variables on launch so everything works
if event == "PROFILE_ACTIVATED" then
	#Store MKeyState in variable "MKeyState"
	MKeyState = GetMKeyState("lhc")
	#Don't ignore next M-Key activation
	MShiftRevert = 0
end

#Add Current-mode M-key as additional key
if event == "M_PRESSED" then
	if arg == MKeyState then
		if MShiftRevert == 2 then
			MShiftRevert = 0
		end
		if MShiftRevert == 0 then
			#OutputLogMessage("M-Remap_PRESSED\n")
			PressKey(MMap)
		end
	end
end

#MShift; on release, revert to MKeyState [ignore current state]
if event == "M_RELEASED" then
	#Ensures the binding to the current M-state M-key is not pressed when it is
	#toggled or MShifted.
	if MShiftRevert == 2 then
		MShiftRevert = 0
	elseif MShiftRevert == 1 then
		MShiftRevert = 2
	end
	#M2-Shift
	if arg == MKeyState then
	#Add Current-mode M-key as additional key
	#Current M-key bind release
		if MShiftRevert == 0 then
			#OutputLogMessage("M-Remap_RELEASED\n")
			ReleaseKey(MMap)
		end
	else
		#Ignore next M-Key activation
		MShiftRevert = 1
		SetMKeyState(MKeyState, "lhc")
	end
end