Re: Nu-HTML checker - <details> as child of <figure>?

--IKC1RK7+nRZ8SZaO
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

For the figure element, the HTML defines the following content model:

https://html.spec.whatwg.org/multipage/#the-figure-element:concept-element-=
content-model

> Either: one figcaption element followed by flow content.
> Or: flow content followed by one figcaption element.
> Or: flow content.

So an example of a non-conforming case would be something like this:

<figure>
  <img src=3Dfoo alt=3Dbar>
  <figcaption>...</figcaption>
  <details><summary>...</summary></details>
</figure>

That example would cause the checker to report an "Element =E2=80=9Cdetails=
=E2=80=9D not
allowed as child of element =E2=80=9Cfigure=E2=80=9D in this context" error.

Specifically, the spec requirements don=E2=80=99t allow content both before=
 and
after the figcaption element =E2=80=94 either the figcaption must be the fi=
rst
child of the figure, or it must be the last child.

But both of the following would be conforming, and would not cause the
checker to report any error.

<figure>
  <figcaption>...</figcaption>
  <img src=3Dfoo alt=3Dbar>
  <details><summary>...</summary></details>
</figure>

<figure>
  <img src=3Dfoo alt=3Dbar>
  <details><summary>...</summary></details>
  <figcaption>...</figcaption>
</figure>

Joseph Polizzotto MA <[email protected]>, 2020-06-01 16:34 -0700:
> Archived-At: <https://www.w3.org/mid/CAGQNLEUAvA_hYSGV9A8mCMhY9wb-9b_n_n3=
[email protected]>
>=20
> I would like to use <details> as a child of <figure> but get an error
> message when running the Nu-HTML checker.
>=20
> error: Element =E2=80=9Cdetails=E2=80=9D not allowed as child of element =
=E2=80=9Cfigure=E2=80=9D in this
> context. (Suppressing further errors from this subtree.)
>=20
> I believe my use of <details> as a child of <figure> is flagged because I
> am also using <figcaption> within the same <figure>. When <details> is a
> child of <figcaption> within the <figure>, the error goes away.
>=20
> I also noticed that when I use <details> as a child of <figure> that does
> not have <figcaption>, there is no error.
>=20
> I am wondering why the Nu_HTML validator allows <details>1) as a child of
> <figcaption> and 2) as a child of <figure> when there is no <figcaption>,
> but not 3) as a child of <figure>, which also has <figcaption>.
>=20
> Note: I have tested <details> as a child of <figure> with a <figcaption>
> with Firefox, Chrome, and Edge and noticed that these browsers do support
> this combination.
>=20
> Thanks for your help,
>=20
> Joseph
>=20

--=20
Michael[tm] Smith https://people.w3.org/mike

--IKC1RK7+nRZ8SZaO
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=h3Q4
-----END PGP SIGNATURE-----

--IKC1RK7+nRZ8SZaO--