W3C home > Mailing lists > Public > public-css-archive@w3.org > February 2019

[csswg-drafts] [css-syntax] tokenizer consume url (#3635)

From: Tab Atkins Jr. via GitHub <sysbot+gh@w3.org>
Date: Fri, 08 Feb 2019 00:02:47 +0000
To: public-css-archive@w3.org
Message-ID: <issues.opened-407952042-1549584166-sysbot+gh@w3.org>
tabatkins has just created a new issue for https://github.com/w3c/csswg-drafts:

== [css-syntax] tokenizer consume url ==
(migrated from mailing list)

**Johannes Koch said:**

> when trying to implement a tokenizer according to the CSS Syntax CR 20 
> February 2014, I encountered the following problem.
> 
> input: url("foo")
> 
>  From 4.3.1 we go to 4.3.3, consuming code points up to and including 
> '(', then to 4.3.5
> 
>    4. If the next input code point is a U+0022 QUOTATION MARK (") or
>       U+0027 APOSTROPHE (‘), then:
>      1. Consume a string token with the current input code point as the
>         ending code point.
> 
> Because the '"' has not yet been consumed, the current code point is 
> still '('. But 4.3.4 should be called with '"' as ending code point instead.
> 
> So I propose to add 'consume it' after
>    U+0022 QUOTATION MARK (") or U+0027 APOSTROPHE (‘)
> in 4.3.5 step 4.

---------------

**Tab Atkins said:**

> Good catch, looks like that's a leftover from a previous organization
> of that section, when I *did* consume the quote character.
> 
> I've fixed that, and slightly simplified the way you call "consume a
> string token" as well, so that if you don't specify an ending code
> point it uses the current input code point.

Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/3635 using your GitHub account
Received on Friday, 8 February 2019 00:02:49 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 06:41:43 UTC