[css-overflow-5] `:target-current` pseudo class should also imply `aria-current=true` on an active scroll marker

[danielsakhapov]

:target-current pseudo class is used to style an active scroll marker within each scroll marker group.
An example would be a table of content where the link to the chapter which is currently in view is highlighted via styling by a :target-current pseudo class.

In https://www.sarasoueidan.com/blog/css-scrollspy/ Sara makes a good point that in addition to just styling an active scroll marker we should set aria-current=true.

[css-meeting-bot]

The CSS Working Group just discussed [css-overflow-5] `:target-current` pseudo class should also imply `aria-current=true` on an active scroll marker.

The full IRC log of that discussion

<sakhapov> sakhapov
<emilio> sakhapov: We have a bunch of links and we can ask the browser to mark the link which has the link whose target is in view
<emilio> ... for a11y we should mark it as `aria-current=true`
<bkardell> q+
<emilio> ... we should add that to the spec
<flackr> q+
<emilio> q-
<Rossen9> ack bkardell
<lea> q?
<emilio> bkardell: you're not suggesting that we should actually set an attribute right? Just the a11y state?
<lea> q+
<emilio> sakhapov: yeah, I guess
<Rossen9> ack flackr
<emilio> flackr: that's what I was on the queue to talk about
<emilio> ... correct, that implies an aria-current state if there's no explicit one for the element
<emilio> bkardell: doesn't that belong to a11y mappings?
<emilio> sakhapov: you're probably right should I ask the a11y wg?
<lea> q?
<emilio> bkardell: I think it should go there
<emilio> q+
<emilio> bkardell: dunno if it goes in html accessibility mappings, not sure if we have a css one
<Rossen9> ack lea
<PaulG> q+
<emilio> lea: +1 to solving this but not quite sure what it's being proposed, so it's not a property
<emilio> ... :target-current is a property
<emilio> s/a property/a pseudo-class/
<emilio> ... so when is this applied, when somebody uses the pseudo-class?
<flackr> qq+
<emilio> ... it'd be easier to have a property
<emilio> flackr: this has nothing to do with whether or not :target-current pseudo-class
<Rossen9> ack flackr
<Zakim> flackr, you wanted to react to lea
<emilio> ... we've resolved that when you use scroll-target-group we select one link as :target-current
<emilio> ... so not affected by usage of the pseudo-class
<emilio> lea: maybe I got confused by the wording of the issue
<dholbert> emilio: it seems like this is the right thing to do. if it has to go into a CSS spec, we should resolve on it
<dholbert> emilio: lea, maybe a better way to frame it -- if you want to express in terms of css properties, you could think of UA stylesheet having :current { internal-property: state}
<Rossen9> ack emilio
<Rossen9> ack PaulG
<bkardell> https://w3c.github.io/css-aam/
<bkardell> https://w3c.github.io/html-aam/
<emilio> PaulG: I believe ??s comments are correct the APA have discussed this and we're going to talk about this at tpac but this should be in HTML probably
<bkardell> s/??s/Sara's
<emilio> sakhapov: so css triggers this behavior but then triggers this pseudo-class
<emilio> ... so it's mostly HTML accessibility
<emilio> sakhapov: so first of all I should find out where we should put this in
<emilio> ... if it's CSS we should resolve on adding aria-current for scroll-target-group
<emilio> ... the issue was also flagged for tpac
<emilio> Rossen9: do we have time scheduled with APA at tpac? We should probably discuss it here
<emilio> bkardell: does it make sense to open an issue in the html-aam repo?
<emilio> Rossen9: yes absolutely

[danielsakhapov]

Small clarification that it has nothing to do with pseudo-class styling.

scroll-target-group property combines a bunch of links into a group and select one active link from that group.

Such active link can be styled with :target-current pseudo-class.

But this issue is about also marking such active link as aria-current=true.