- From: Joe Pea via GitHub <sysbot+gh@w3.org>
- Date: Tue, 12 Jan 2021 23:26:42 +0000
- To: public-pointer-events@w3.org
Seems that:
- if `pointer-events` is configured such that one element has `pointer-events: auto` (all others are disabled because `html {pointer-events: none}` is in place),
- If the element has no children, there should be zero hit testing.
- If the element has any children, then only those children are hit tested (makes sense).
So, for an element with class `.container` that has no children, then
```css
html { pointer-events: none; }
.container { pointer-events: auto; }
```
would be the same performance as the capture that we described above, and for an element with children, we can selectively include which things should be hit tested. Either,
```css
html { pointer-events: none; }
.container { pointer-events: auto; }
```
to hit-test all children and grand-children inside of `.container`, or
```css
html { pointer-events: none; }
.container .specific-elements { pointer-events: auto; }
```
to hit-test only particular elements that are (grand)children of `.container`.
It would be more intuitive for this to be the case, with `pointerover` working by default, without the whole capture idea (that despite the length of time I've been a professional web developer have never had to worry about).
In fact, that's how I always intuitively expected `pointer-events` to work. The key is that it is intuitive.
--
GitHub Notification of comment by trusktr
Please view or discuss this issue at https://github.com/w3c/pointerevents/issues/346#issuecomment-759101363 using your GitHub account
--
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Tuesday, 12 January 2021 23:26:43 UTC