Need help getting it to work

[onuralpbicer]

Hey, I'm a fellow computer science student, but I don't have too much experience with visual studio nor c++. I was trying to get this working on a MSI x470 Gaming Plus board but I can't seem to get it to work. Could you help if you're still around?

[SimoDax]

What issue are you facing?

[onuralpbicer]

So basically when I load your code into visual studio 2017 and try to compile, I get an error saying "Unable to start program '(path-to-folder)\x64\Debug\MSI-mystic-light-tool.exe'. The system can't find the file specified."

I have mystic light 3 installed if that would change anything.

Also in the editor, lots of things are underlined with red lines, starting with the #include "MysticLight_SDK.h"

As I said, I'm not familiar with visual studio. Maybe I made a mistake trying to import the project into the ide? How was I supposed to actually get the project into visual studio?

[SimoDax]

Theoretically you just need to clone the repo and open the .sln file in visual studio. Anyway, try doing it manually: go to files>new>project. On the left go to Visual C++>Windows desktop and create a new windows console application. Name it MSI-mystic-light-tool. Close visual studio and replace all the .cpp and .h files with the ones on this repo. Copy MysticLight_SDK.dll and MysticLight_SDK.h too. That should do it.
I noticed you're compiling for 64 bit, change it to x86 or it won't load the dll

[onuralpbicer]

Alright, it is compiling now at least. I'm getting "Function returned status code: Request is complete." and then nothing for a bit. Then I got a Debug Assertion Failed runtime library error in atlsafe.h line 273, expression m_psa != 0. If I don't do anything in that window for a while on the console I get "Function returned status code: MLAPI_Initialize is timeout." And then press any key to continue

I'm guessing that might be because I'm running ML3?

[SimoDax]

Try running visual studio as administrator, the program will inherit admin rights.
If it doesn't work with that I guess it's ML3 and you need to wait for an update to the sdk from msi. You could try installing mystic light 2 and see what happens

[onuralpbicer]

Admin rights did work, though the led count is 14966932 so obviously that's wrong, but that should be fine, I'm only trying to make a color cycle effect. I am actually going to try to use logitech's sdk alongside msi's to synchronize my keyboard and mouse with the motherboard.

Actually, as I was typing all this. After I press enter to continue after I see my device id, I get another assertion fail at atlsafe.h at line 199 expression (((HRESULT)(hRes)) >= 0)

[SimoDax]

That's odd, I don't know what it could be, try looking at the call stack to see what function caused that call. I wouldn't give it much weight tho, as it happens just before the program closes.

As for the led count, I had begun making a user interface with Qt, and this snippet should return the proper value (j):

int j = 1;
BSTR ledName;
CComSafeArray<BSTR> ledStyles;

status = ml.getLedInfo(devices.GetAt(i), (DWORD)j, &ledName, &(ledStyles.m_psa));

while(status == 0){     //try getting led info until there are no more and it fails

                //create a list of led names & styles if you need...

                status = ml.getLedInfo(devices.GetAt(i), (DWORD)j, &ledName, &(ledStyles.m_psa));
                j++;
}

[onuralpbicer]

I was just typing an edit, that didn't happen again after that time so I guess that's fine. I just tried a setColor just to test if I can use it though but it didn't work. I don't really need styles, I will write the color change myself with increasing/decreasing rgb values(I've already done that for logitech's sdk so I'll just put bunch of setColors).

ml.setColor(devices.GetAt(i), 0, 255, 0, 0);

That was the function call I used, most likely wrong because I have 0 as the second parameter but I'm assuming it should have something else. How would I call that function to set all LED's to the same color?

[SimoDax]

You can't do it directly, look at the reference:
int MLAPI_SetLedColor(BSTR type, DWORD index, DWORD R, DWORD G, DWORD B)

You need to pass both the device and the led index. which ranges from 1 to how many leds areas are in your device. So you need a loop to call the function for each index

edit:
jugglingcode linked a stub on how to do it days ago, see issue #1

[onuralpbicer]

Yeah, after looking at your sample getStyle code I think I understand how it works. I'm gonna give it a go later and see how it goes, thank you very much.

…

On Sat, 28 Jul. 2018, 20:09 SimoDax, ***@***.***> wrote: You can't do it directly, look at the reference: int MLAPI_SetLedColor(BSTR type, DWORD index, DWORD R, DWORD G, DWORD B) You need to pass both the device and the led index. which ranges from 1 to how many leds areas are in your device. So you need a loop to call the function for each index — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#3 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AjOdsRfRdeif4O2PUtVmXCLhgfTDfsSQks5uLDhggaJpZM4Vk5NW> .

[onuralpbicer]

So I've been trying to play around with it for a bit, but right now if I put any code after the print device type and led count in your sample loop(whether it is in the loop after that line or out of the loop) it just doesn't execute. I even threw in some prints to see if they worked but after your device type print, nothing else happens it just says press any key to continue.

I tired jugglingcode's example as well, I'm still getting the same behavior. I put another print after that print and it worked, but the mlapi function calls after that don't work, and neither do the printStatus functions.

"Exception thrown at 0x00000000 in MSI-mystic-light-tool.exe: 0xC0000005: Access violation executing location 0x00000000." is what I'm getting in the debugger when I try to call setColor after getting device info stuff.

[SimoDax]

I even threw in some prints to see if they worked but after your device type print, nothing else happens it just says press any key to continue

Seems like it runs the old code instead of compiling the new one. Have you tried cleaning and rebuilding the solution?

"Exception thrown at 0x00000000 in MSI-mystic-light-tool.exe: 0xC0000005: Access violation executing location 0x00000000." is what I'm getting in the debugger when I try to call setColor after getting device info stuff.

Put your code on gist or pastebin, I can't understand what you did

[onuralpbicer]

I'm uploading everything to git at https://github.com/onuralpbicer/mystic-light-sdk-test I haven't really done almost any changes to the code anyways, I just added a setColor.

Just to add, when I run with ctrl+f5, it doesn't give an error it just doesn't do anything after the initialization and getDevInfo. When I run it with the windows debugger button, it breaks with an exception at the first mlapi function I try to call(i've tried different functions too) after getDevInfo.

[SimoDax]

You forgot to initialize the aditional pointers to the mystic light functions. By default they have a NULL value, which is a C macro for 0, which is a forbidden memory address, resulting in an access violation.

Add the initialization in loadLib(), my stub only used init, errorMsg and getDevInfo so you'll want to add at least
ml.setColor = (LPMLAPI_SetLedColor)GetProcAddress(MLinstance, "MLAPI_SetLedColor");

[onuralpbicer]

Ohhhhh, that makes a lot of sense. I'll try that in a bit.