Ok, using the standard function to get the mouse coords (the one that's posted everywhere) on the new 10.051 release, updates the active flag, but the X and Y coords are always zero for WebOS (Windows works fine). Incase I was going crazy I tested on 10.042 and it works fine in WebOS with my same code. Not sure if it still works on iOS as I'm informed, by the wife, that my dinner is ready :-[
Oh, and how do I go back a version in GLB? ie. so I can return to 10.042?
Cheers
You mean MOUSESTATE isn't working?
A quick test of a WIP app (that worked perfectly with the last update) confirms that there is indeed a problem with MOUSESTATE on webOS.
BUG!!!
Please put "SYSTEMPOINTER TRUE" at the start of your program. Update tonight.
I did use SYSTEMPOINTER TRUE :(
Cheers Gernot :)
99.9% sure I did too (am at work so cannot check)...
Anyone know how to go back a version? Would be nice if there was an easy was in GLB to have a history of updates that you can switch back to if you wish or somesuch. Or at least an undo for the latest install.
Cheers
I've never had to roll back, but you can obviously download the last update again. Someone recently mentioned unzipping the backup files. If Gernot is going to fix this update tonight (as stated above), then I'm not going to worry about rolling back.
systempointer did the trick for me. Display the mx, my position with a rectangle. With Systemprtr false it's moveing parallel to your finger, with "true" it should be fixed.
Not sure if this is related but I found that with a dual screen monitor windows program with a touch screen on the bottom that if SYSTEMPOINTER was false then the mouse coordinates were all over the place, with it True then it displayed correctly. This was the case in the v9 versions as well
yes. No systempointer means, the mouse "position" is only calculated from the velocity values.
Ok, I definitely have SYSTEMPOINTER TRUE. I've updated to the today's release 10.053 and it still doesn't work. As I said initially the old version (10.042) works fine on exactly the same code.
But, and here is the strange thing, I can get the coords of each finger to work (normally stays at 0x0) if you press all fingers down at the same time. Although they *all* show the same coords. I think the coords of the first finger.
Here is my main mouse update routine, which is a slightly modified version of the usual one:-
FUNCTION updateMice:
LOCAL imouse%,mx%, my%, b1%, b2%
fingers=0
FOR imouse% = 0 TO MIN(GETMOUSECOUNT()-1,4)
SETACTIVEMOUSE imouse%
MOUSESTATE mx%, my%, b1%, b2%
IF test=1 THEN PRINT imouse+":"+b1+"="+mx+"x"+my,0,200+imouse*10,10
mouse[imouse].x=mx/x
mouse[imouse].y=my/y
mouse[imouse].active=b1
IF b1=1 THEN INC fingers
NEXT
ENDFUNCTION
Note the only changes to the code are to print a debug string, to alter the coords (afterwards) for different size screens, and to count the number of fingers touching the screen. That's about it. And, again, the exact same code works with 10.042.
Oh, and the b1 flag (ie. if is active) works perfectly fine. One finger shows the first finger is active, two fingers shows two fingers are active etc. It's just the x and y coords all stay at zero until all fingers are on the screen.
Cheers
Yes, does seem to be a bit of problem there - multiple mice aren't being detected...
For me if all mice are active then they all return the same coords. If less are active then it says they are active but all coords for all mice are zero.
Cheers
I find that when using SETACTIVEMOUSE, everything returns 0 on a Pre. Presses are detected, but coordinates aren't...
[attachment deleted by admin]
Try pressing four fingers. I then get coords for all mice but they are all the same coords.
Definitely a new bug introduced in the last couple of days
Cheers
Thnaks for the detailed bug report. I'll have a look into that. Hopefully tonight.
OK, fixed now.
There's one tiny issue now. When you call MOUSESTATE or SHOWSCREEN, the pointer X,Y values are updated and thus, the acceleration values for this frame are set to 0. If you call MOUSEAXIS(0) (=acceleration X), you get the difference movement since the last MOUSESTATE or SHOWSCREEN. So, this works pretty good until you call MOUSEAXIS directly after MOUSESTATE (which is pretty redundant, anyway).
I cleared the code a lot up now and multitouch is working fine on all platforms now, I guess. Also switching systempointer true/false on the PC/Mac platforms works fine.
Update in a few minutes.
That's good news :)
iTouchMouse is working again :)
Excellent. I will try it out tonight. Ta!
Cheers
Can someone please test on iOS now? If that's it, I'll release.