xref: Some page visibility entries not showing up

[marcoscaceres]

The following entry for "external now visible algorithm" is not showing up in xref for some reason:

https://github.com/w3c/webref/blob/cd238a8986d9c1355ff200e29db21f896c72d43f/tr/dfns/page-visibility-2.json#L228-L244

[sidvishnoi]

It's private according to line 236 there. Should export it from the spec.

[marcoscaceres]

Weird...
https://github.com/w3c/page-visibility/blob/gh-pages/index.html#L349

And:
https://github.com/w3c/page-visibility/blob/gh-pages/index.html#L365

Are both non-private. Could it be the dfn-type?

[marcoscaceres]

Ok, so, in any case, the bug is probably in WebRef as it's not picking them up as public.

[tidoust]

What is the exact definition you're looking for?

There are three distinct definitions:

1. now visible algorithm
2. now hidden algorithm
3. external now visible algorithm

Definitions 1 and 2 are the ones you linked to. They are defined with a data-export attribute and are flagged as public in webref. Definition 3 is the one you seem to be looking for. It is defined through a simple <dfn> tag and is correctly flagged as private in webref as far as I can tell.

[marcoscaceres]

@tidoust, yes, you are correct about "external now visible algorithm".

For whatever reason, the other two are not showing up in xref:

• https://respec.org/xref/?term=now+visible+algorithm
• https://respec.org/xref/?term=now+hidden+algorithm

But it's not a bug here... I'll move the issue over to Respec.

[sidvishnoi]

They both have type: "abstract-op", we only support dfn. From Bikeshed:

abstract-op (for "English-language algorithms")
dfn (for general terms and phrases, and a catch-all for anything else)

How do you suggest going forward? Rename abstract-op to dfn in ReSpec data, or add a new supported search type (with potentially new syntax?)?

[marcoscaceres]

How do you suggest going forward? Rename abstract-op to dfn in ReSpec data, or add a new supported search type (with potentially new syntax?)?

I don't think we need a new syntax. We should treat them as dfn and link to them with [=the whatever algo=] (same as BS).