W3C home > Mailing lists > Public > www-style@w3.org > April 2013

Re: [selectors4] Non-editorial comments on the 2013-04-17 ED

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Wed, 24 Apr 2013 15:52:52 -0700
Message-ID: <CAAWBYDD3qw_UPRiCuBdsCGi0qne__yvfLNziMSTsO3vHvAmyNA@mail.gmail.com>
To: Simon Sapin <simon.sapin@exyr.org>
Cc: www-style list <www-style@w3.org>
On Sun, Apr 21, 2013 at 8:04 AM, Simon Sapin <simon.sapin@exyr.org> wrote:
> §6.7 "What makes attributes of type ID special is that no two such
> attributes can have the same value in a conformant document." What about
> non-conformant documents that reuse the same ID? Can ID selector ever match
> multiple elements, in practice?

Yes, it can.  Added a sentence to that effect.

> §7.3 "The fragment identifier of the document URI is stripped before
> matching against the link's URI; otherwise all portions of the URI are
> considered." Shouldn’t both fragment identifiers be stripped?

Yup, fixed.

> Same section: "Similarly if the document's URI is not a URL, the
> pseudo-class does not match anything." That’s only in the functional
> pseudo-class case, right? By the way, what’s the difference between URI and
> URL? (Definition needed, preferably by reference.) How are URIs parsed? Does
> http://w3.org:80/ match http://w3.org/ ? I suggest referring to the WHATWG
> URL standard for parsing, and use "has a relative scheme"[1] instead of "URI
> that is an URL".
> [1] http://url.spec.whatwg.org/#relative-scheme

"Relative scheme" is not what you want here; you want some notion of
"hierarchical".  This isn't defined in Anne's URL spec yet, but I've
gone ahead and just used the term.

> §8.1 "User agents not that do not support interactive media do not have to
> support this pseudo-class." What does it mean to not support a pseudo-class?
> Parse as invalid (and invalidate the whole selector list) or parse as a
> valid selector that doesn’t match anything? cssselect and WeasyPrint
> currently do the latter.

Fixed as part of previous edits to not use that wording, and instead
explicitly define that it's valid but matches nothing.

> §12.7 Should combinators inside :nth-match() and :nth-last-match() only be
> allowed in the Complete profile, as for :matches()?

They should be consistent, yes.  Fixed.

~TJ and fantasai
Received on Wednesday, 24 April 2013 22:53:43 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:28 UTC