-
Notifications
You must be signed in to change notification settings - Fork 522
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[IntersectionObserver] #295 V2: visibility detection #523
Conversation
index.bs
Outdated
: <dfn>isVisible</dfn> | ||
:: | ||
True if {{IntersectionObserver/trackVisibility}} is <code>true</code> | ||
and the <a>visibility</a> algorithm, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This implies the value false
if trackVisibility
is false
, while later on the algorithm is specified to immediately returns true
in that case. Of the two, false
has a safer failure mode if someone thinks they're tracking visibility but they screwed up somewhere. Either way, the two parts of the spec need to agree.
Even better would be if isVisible
were completely undefined if you had not enabled trackVisibility
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For the purpose of feature detection, how about setting isVisible
to null
if trackVisibility
is false? I think that's also more consistent with other IDL I have seen.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
you can't set a boolean to null, can you? Honest question -- I'm not sure what the WebIDL can express, nor what would be most consistent with similar APIs. This feels kind of off, but a permanent false
when tracking isn't enabled wouldn't be terrible. The true
specified in the algorithm below will lead to trouble.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not an IDL expert, but for the record chromium's IDL code generator allows null-able booleans. I wrote a chromium patch which would implement the behavior I proposed, and it works.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You can have nullable booleans (the IDL of this PR does not do that btw), but generally it would be better to use an enum.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not sure an enum will work here, because we need two different values that will evaluate to false
when cast to boolean.
If there are no objections, I'll just make it a permanent false
when tracking isn't enabled. I think that's probably what we want from a developer ergonomics no-surprises point of view. Feature detection can always be done via IntersectionObserverEntry.prototype.hasOwnProperty('isVisible')
.
Update the PR based on comments so far, PTAL. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for making these changes—looks good
SHA: 027e2dd Reason: push, by szager-chromium Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
`intersection-observer-v2`: spec merged into the Intersection Observer spec w3c/browser-specs#1210 w3c/IntersectionObserver#523 Co-authored-by: Daniel D. Beck <[email protected]>
The V2 set of features adds visibility reporting, with detection of occlusion and visual effects (explainer, issue).
Preview | Diff