- From: Brad Kemper <brad.kemper@gmail.com>
- Date: Thu, 24 Jan 2013 12:08:41 -0800
- To: "Tab Atkins Jr." <jackalmage@gmail.com>
- Cc: Lea Verou <lea@w3.org>, Sylvain Galineau <sylvaing@microsoft.com>, Mounir Lamouri <mounir@lamouri.fr>, www-style list <www-style@w3.org>
On Jan 24, 2013, at 10:00 AM, "Tab Atkins Jr." <jackalmage@gmail.com> wrote: > On Wed, Jan 23, 2013 at 9:40 PM, Brad Kemper <brad.kemper@gmail.com> wrote: >> On Jan 23, 2013, at 5:12 PM, Lea Verou <lea@w3.org> wrote: >> >>> The main issue with that is that if authors don't style it but they do style input, the UA stylesheet can't override that with input:not(:completed), similarly to how a UA¡¯s a:hover won't override the author¡¯s a. But a :placeholder pseudo-class has the same issue. I wonder how browsers implementing placeholders with a pseudo-class have solved this. Are they violating the cascade or am I missing something? >> >> A UA stylesheet with an 'a:hover' rule defined should override what the author has for the same property in an 'a' rule, because 'a:hover' has more specificity. > > No, it's in a weaker origin, so it loses to the author rule before > specificity is even checked. OK, I am mistaken then. I didn't realize the matching was at the element level only. I had assumed somehow that step 1 of the cascading order was to find all declarations that matched the element **in its current state** and the property in question, but I see now that this isn't the case. Sorry for the noise.
Received on Thursday, 24 January 2013 20:09:14 UTC