exxos blog - random goings on

[exxos]

Some more final tests (still can't get it to work right).

This is with XaAES AND MULTITOS.

Capture.PNG (86.44 KiB) Viewed 2047 times

LEFT CLICK ON INFO WINDOW - XaAES - CLICK OK - CORRECT ID 4

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 2             1             3             4             20            3             0             4             14            40            173           31 

LEFT CLICK ON INFO WINDOW - MINT - - CLICK OK - CORRECT ID 4

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 2             1             3             4             20            0             0             4             10            10            620           275 

RIGHT CLICK ON INFO WINDOW - XaAES - - CLICK OK - CORRECT ID 4

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 2             2             3             4             20            3             0             4             14            40            173           31 

RIGHT CLICK ON INFO WINDOW - MINT - - CLICK OK - CORRECT ID 4

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 2             2             3             4             20            0             0             4             10            10            620           275 

LEFT CLICK ON GB7 MAIN WINDOW - XaAES (DOES NOT REGISTER THE CLICK ONLY A EVENT - HANDLE CORRECT 3)

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 16            0             3             4             21            3             0             3             0             0             0             0 

LEFT CLICK ON GB7 MAIN WINDOW - MiNT - (DOES NOT REGISTER THE CLICK ONLY A EVENT - HANDLE CORRECT 3 )

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 16            0             3             4             21            1             0             3             8             38            628           333 

RIGHT CLICK ON GB7 MAIN WINDOW - XaAES - ( CLICK OK BUT mess(3) is WRONG should be 3! )

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 2             2             3             4             20            3             0             4             142           40            125           47 

RIGHT CLICK ON GB7 MAIN WINDOW -MiNT - CLICK OK AND CORRECT WINDOW ID 3!

Code: Select all

e             buttona       aeshandle     aeshandle2    mess(0)       mess(1)       mess(2)       mess(3)       mess(4)       mess(5)       mess(6)       mess(7)
 2             2             3             4             21            1             0             3             8             38            628           333 

So as tested yesterday. XaAES is reporting the wrong window handle while MultiTOS doesn't.

But in both cases, left clicking the main GB7 does not work with the INFO WINDOW open. If it is closed, left clicking works fine in the main GB7 window.

So no idea what's going on. I think XaAES has a handle bug somewhere. No idea why the left click malfunctions with the info window open, when right clicking works. Same issue in TOS206 as well. So not sure if this is some bug in my code, or some long term bug in every AES, or what I am trying to do is just not possible.

As its stands I can't do any more. I am just spending all my time chasing bugs in HISOFT libs, bugs in the AES, etc than actually writing my own code and debugging. Kinda lost motivation for it all now. If I do pick up GB7 again in the future it will be TOS only.

[Badwolf]

So something is royally fu**ked up in that FREEMINT / XaAES setup of yours @Badwolf .Something in that setup is not reporting the correct AES handles. But it works perfectly well in good'ol MinT.

So if I continue with MiNT (EMUTOS) . The problem I actually have is why one of the mouse buttons is returning as a keyboard event only.

Are you mixing up terminology? Do you mean MultiTOS when you're saying MiNT here?

I'm totally lost as to what you're referring to.

@Badwolf Could you put evnt_multi in your 2 windows test program and just get it to output what window is clicked on ?

Code: Select all

       e=evnt_multi(mu_mesag+MU_BUTTON+MU_KEYBD,258,3,0,0,0,0,0,0,0,0,0,0,0,_
                       VARPTR(mess(0)),0,xa,ya,buttona,0,0,0) 

MU_BUTTON incorrectly returns "16" for left button clicks. I can't believe the AES is that badly messed up. TOS is doing the same. So its looking like the function is broken in hisoft. If it is, there is really no point in continuing with it all. The only work around would be to call it from assembly and pass the values back to GB7. But I need to prove the function is broken first

Oh crikey. It's been four years since I genuinely tried to work with GEM. I'll have to see what I can remember.

BW

[exxos]

Are you mixing up terminology? Do you mean MultiTOS when you're saying MiNT here?

Yeah I keep saying MiNT for some reason.

Oh crikey. It's been four years since I genuinely tried to work with GEM. I'll have to see what I can remember.

It was just to check clicking events on both windows. As mentioned in my previous post, the handles get messed up and half the time clicking the mouse ends up as a keyboard event! So either everyone AES version has similar bugs, or HISOFT libs are more broken than I thought. It's game over either way really.

If it's not simple then don't worry about it.

[Badwolf]

It was just to check clicking events on both windows. As mentioned in my previous post, the handles get messed up and half the time clicking the mouse ends up as a keyboard event! So either everyone AES version has similar bugs, or HISOFT libs are more broken than I thought. It's game over either way really.

If it's not simple then don't worry about it.

Not 100% sure what you want me to check, but I see what looks like normal behaviour to me.

I get a 1 if the left button is clicked on either window, a 2 if the right button is, a 0 if I resize (event on button up) and a 3 if I mash both buttons during the click.

AFAICT (not having done button clicks before) that's right.

Screenshot 2023-01-10 at 22.13.59.png (34.62 KiB) Viewed 2006 times

Here's the code.

& here's the relevant part

Code: Select all

void messageloop( int handle[] ) {
	short int msg_buf[8];
	short isfull;
	short int oldx, oldy, oldw, oldh;
	short xa, ya;
    short buttona;
    
    
	do {
//		evnt_mesag( msg_buf );
        evnt_multi(MU_MESAG | MU_BUTTON | MU_KEYBD, // flags
                    258,                            // max clicks
                    3,                              // button mask
                    0,                              // button state
                    0,                              // move flags
                    0,                              // move x
                    0,                              // move y
                    0,                              // move w
                    0,                              // move h
                    0,                              // mm2 flags
                    0,                              // mm2x
                    0,                              // mm2y
                    0,                              // mm2w
                    0,                              // mm2h
                    msg_buf,                        // message buffer
                    0,                              // timer low word
//                    0,                              // timer high word
                    &xa,                             // evnt_mouse x
                    &ya,                             // evnt_mouse y
                    &buttona,                        // evnt_button button state
                    0,                              // evnt_button special key state
                    0,                              // scancode
                    0 );                            // button click count
		switch( msg_buf[0] ) {
			/* msg_buf[0] is message type */
			
			case WM_MOVED:
				wind_set( msg_buf[3], WF_CURRXYWH, msg_buf[4],
					msg_buf[5], msg_buf[6], msg_buf[7] );
				break;
			case WM_FULLED:
				isfull = isfullwin( msg_buf[3] );
				if( !isfull ) {
					wind_get( msg_buf[3], WF_CURRXYWH,
						&oldx, &oldy, &oldw, &oldh );
					graf_growbox( oldx, oldy, oldw, oldh,
						fullx, fully, fullw, fullh );
					wind_set( msg_buf[3], WF_CURRXYWH,
						fullx, fully, fullw, fullh );
				}
				else {
					wind_get( msg_buf[3], WF_PREVXYWH,
						&oldx, &oldy, &oldw, &oldh );
					graf_shrinkbox( oldx, oldy, oldw, oldh,
						fullx, fully, fullw, fullh );
					wind_set( msg_buf[3], WF_CURRXYWH,
						oldx, oldy, oldw, oldh );				
				}
				break;
			case WM_SIZED:
				wind_set( msg_buf[3], WF_CURRXYWH,
					msg_buf[4], msg_buf[5], msg_buf[6], msg_buf[7] );
				break;
			case WM_REDRAW:
				redraw_window( msg_buf[3], (GRECT*) &msg_buf[4] );
				break;
			case WM_TOPPED:
				wind_set( msg_buf[3], WF_TOP, 0, 0, 0, 0 );
				break;
		}
        printf("ButtonA: %x\r\n",buttona);

	} while( msg_buf[0] != WM_CLOSED );

}

BW

[exxos]

Not 100% sure what you want me to check, but I see what looks like normal behaviour to me.

I get a 1 if the left button is clicked on either window, a 2 if the right button is, a 0 if I resize (event on button up) and a 3 if I mash both buttons during the click.

Thanks very much. So indeed 1=left click, 2-right click, 3=both down. I don't get that with the same event call

It only behaves on the main GB7 window. As soon as the second one opens, it starts going nuts and doesn't register left clicks on the main GB7 window anymore. It often triggers a message / keyboard event which makes no sense.

So I can only conclude HISOFT was never tested properly on later AES / CPU combos. It must have just been a total fluke the original GB4 worked as it only ever opened 1 window at a time.

[Badwolf]

So how come regardless of which window I actually click on, the AES returns window 3 ?!

Silly question, as I only ever dabbled in GEM and have never gone further, but are you sure the AES is *meant* to send you the window handle on a button event?

I can't find any reference to getting the window from an MU_BUTTON event, only from a MU_MESAG event.

I'm just wondering if you have to keep track of where your windows are and work out which window has been clicked on yourself?

BW

[mfro]

Silly question, as I only ever dabbled in GEM and have never gone further, but are you sure the AES is *meant* to send you the window handle on a button event?

Not so silly. If you receive a button event from an evnt_multi() call, the message buffer contents has no meaning (i.e. is invalid).

[exxos]

I think left clicks are sending the ON TOP event when left clicked. It works fine in some other window is ontop (IE desktop window) . I can only assume this is just how the AES works. Even though right clicking still works when not ON TOP.

[Badwolf]

I think left clicks are sending the ON TOP event when left clicked. It works fine in some other window is ontop (IE desktop window) . I can only assume this is just how the AES works. Even though right clicking still works when not ON TOP.

I think what mfro was saying and what I'm observing is that BUTTON events are not window-based so contain no window handle information. You get a window handle from MESSAGE events -- one of those is the TOPPED event (going from not topped to topped) but once topped you don't get another one at the next click.

But if you're trying to read out the window handle from a BUTTON event, you're probably just getting whatever old value was in the variable you're using at the time of the previous MESSAGE event.

BW

[exxos]

It has been a rough week hence me being quiet. But frankly had enough of all things computer related for the time being.

Aside from the usual life drama which I shall spare everyone with for the time being. My girlfriend have problems with her computer because chrome updated it seems and would not work any more on her computer. It would just come up as a blank page. Literally nothing else other than the window.

Did the usual pointless Internet searching and tried to find a previous version of chrome to download but they basically do not exist. The one I did find just seem to be a folder dump which just made things worse. Something about " side-by-side configuration not correct". I know I have had that before and just could not be bothered again. So restored the backup of her C: drive from the day before.

Chrome must have already downloaded the update because shortly after it updated again and stopped working So them restored the backup again and more Internet-foo to disable it from running or installing the updates. Which did not work. So reinstall the backup for a third time and eventually found the update folder and deleted that. So it managed to behave then.

It seems her PC had version 108 of Chrome and the last version is 109. I don't know why it totally broke on her PC. Her windows laptop was actually fine. My laptop and already updated to 109 and my PC. No problems. So I have no idea why 109 just completely screwed up.

On my own PC I was going to check the version of chrome only I have had a JAVA icon in my shortcut docker Which I tried to delete. I then ended up deleting control panel from the thing and things just rapidly went downhill after that. The backup I had for the settings for my taskbar and shortcut bar were from 2018 I restored its own program files folder from my backup and it still was screwed up. I resorted to restoring the registry and that just totally screwed up my computer. No great surprise there really.

The last backup I had was Tuesday teatime 5pm backup on my NAS drive. Only After much hair pulling, turns out Paragon Cannot restore from the NAS drive for some crazy reason. It just kept saying the volume needs exclusive access which I could just not work around.

But "thankfully" I had the utilities CD, so thought I would run the software from that instead of Windows. Which then turns out it cannot see my NAS drive, presumably because it has no network drivers in the thing. At this point I just thought solder and I'll use my dinnertime backup from a few hours previously as it was on a SSD in my PC anyway. only to find my C: drive does not appear in paragon CD utilities either. Presumably again due to lack of a driver for my new motherboard and the M.2 SSD drive.

So booted back up into Windows which ran incredibly slow. It literally took 10 minutes before the desktop was even functional! so then managed to restore the dinnertime backup from Windows. And then restored my email folder from the teatime backup but I don't think there was much difference between the two backups anyway.

Needless to say I have backed up my start menu and docker settings. Also now I am reorganising my backups which do 2 backups per day ( morning and night) onto my PCs backup SSD drive and then The teatime backup gets dumped onto my NAS drive.

Then after hours of fighting with all that I still have that icon which I cannot bloody delete

Capture.PNG (285.9 KiB) Viewed 1848 times

So after fighting with the GEM AES, Microsoft stupid support, chrome screwing up, and me screwing up. I just had enough of all things computer and technology related for the time being. Oh and I also spent most of yesterday trying to get the printer to print the yellow out properly as well.

So if I am quiet for a while and no progress updates then people know why. I'm going to spend my time more constructively by not getting out of bed etc!