- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Tue, 4 Mar 2014 14:36:38 -0800
- To: Johannes Koch <koch@w3development.de>
- Cc: www-style list <www-style@w3.org>
On Tue, Mar 4, 2014 at 1:47 PM, Johannes Koch <koch@w3development.de> wrote: > Hi, > > 5.4.1. Consume a list of rules > > ... > <at-keyword-token> > Reconsume the current input token. Consume an at-rule. If anything > is returned, append it to the list of rules. > > [Then 5.4.2. Consume an at-rule] > > Create a new at-rule with its name set to the value of the current input > token, its prelude initially set to an empty list, and its value initially > set to nothing. > > Repeatedly consume the next input token: > > [Because of 'Reconsume the current input token' in 5.4.1. the > <at-keyword-token> is still the current input token.] > > ... > anything else > Reconsume the current input token. Consume a component value. > Append the returned value to the at-rule’s prelude. > > [The component value consumed is the <at-keyword-token> and is added to the > prelude.] > > The same applies to 5.4.1 > anything else > > also reconsuming the current input token before consuming a qualified rule. > > I think the 'Reconsume the current input token' has to be removed. Fixed, but in a different way. > Additionally, I don't understand the case > <CDO-token> > <CDC-token> > ... > Otherwise, ... > > input: <!--foo > creates a prelude with 3 entries: > 1. CDO > 2. CDO > 3. Ident foo I don't understand why this would happen. In any case, it certainly doesn't happen in the current spec, now that I've fixed the algos. ~TJ
Received on Tuesday, 4 March 2014 22:37:24 UTC