solved GUI pointer observables functionality is different in latest version?

[royibernthal]

Was there any change lately that could affect the functionality of GUI pointer observables?

I attached the log of the event bubbling in Desktop and Mobile (device mode) at the pointer down event of the same button.

In desktop it starts from the last child, while in device mode it seems to skip 2 children, there are other weird behaviors in different cases, e.g. many callbacks called for different controls for the same pointer down event of the same button - I imagine they're controls nested in the pressed control or the other way around, or possibly they're just close enough to the pointer down area to be notified as well.

Is this possibly related to the newly added pointerId? Why is the behavior different on mobile?

No PG repro for this yet as I thought you might know what's going on without.

 

EDIT:

After more testing I'm under the impression that the pointer area detection is not accurate now in mobile, possibly when combined with using idealWidth.

For instance clicking some place can trigger a button somewhere else at its original positioning before idealWidth is applied to it. Let me know if I got this right.

[brianzinn]

Did you try on different browsers?  Also, can you check your viewport - some people reported needing special meta tag? ie:

<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1, user-scalable=1">

 

[royibernthal]

It used to work properly a few bjs versions ago so the problem shouldn't be in the HTML, no?

[brianzinn]

Some people were reporting the new version of Chrome had that problem, but otherwise you are onto something else.

[royibernthal]

Just checked, the old version is working perfectly on the same up-to-date chrome browser, meaning this is most likely a bjs bug.

[brianzinn]

Can you repro in a PG?

[royibernthal]

I should really start testing more thoroughly before speaking. The meta tag fixed it exactly as you said, thanks

[brianzinn]

No worries. This does feel like something BJS should take care of, though.  It's caught a few people already and I feel it should work without the meta tags.  Otherwise, maybe we do need a Tips and Tricks page for stuff like this

[royibernthal]

I agree, I think adding in bjs such a meta tag to the head if it doesn't exist should not be a problem, sort of like an HTML polyfill.

[brianzinn]

I know - it's the point of using a library that cross-browser/platform quirks are handled transparently.  Did you find an underlying reason - like clientWidth/clientHeight being calculated incorrectly - or not reporting devicePixelRatio correctly?  I searched when this was first discovered through Chrome bugs, but did not find anything.

[royibernthal]

Sorry for the very late reply. Yup. I didn't look deep enough to give you a detailed explanation about that quirk or how to reproduce it, but I was dealing a lot with resizing and different device sizes.