Using the right mouse button can cause a never-ending pointer about pointer-tracker HOT 4 CLOSED

Fixed in 472dacc and published as 2.5.2.

The library keeps an eye on move events, and if the left button is no longer down it runs the 'end' steps for that pointer.

from pointer-tracker.

Thanks for filing this. I've created a reduced example at https://static-misc-3.glitch.me/pointer-test/ that doesn't use this library.

I'm trying to figure out if I'm missing something, or if this is a browser bug.

from pointer-tracker.

Some additional findings (in Chrome):

• I was wrong about having to keep the cursor still during step 2, you can move it before pressing RMB and the bug will still occur.
• The bug doesn't occur if you wait a little while between pressing and releasing LMB during step 5, I mistook this for having to keep the cursor still, but you can still trigger the bug if you move and release quickly enough.
• With your reduced example I can now see that Safari doesn't trigger a pointerdown if you click another time after the context menu was closed, which is why the pointer state is successfully recovered in the painting demo.

It seems like the browser doesn't fire any pointer events while the context menu is disappearing.

So if the browser were to trigger a pointercancel for the first click when the context menu opened (no idea how this is specced), there would now be an unmatched pointerup event if you wait a second before releasing LMB in step 5, which would also be a little unexpected.

from pointer-tracker.

Filed w3c/pointerevents#407 and w3c/pointerevents#408

from pointer-tracker.