W3C home > Mailing lists > Public > public-css-testsuite@w3.org > February 2009

RE: Explaination request: MS test (IE8, RC1) -- H4\at-rule-001.htm

From: Robert Stam <robert@tallcomponents.com>
Date: Fri, 27 Feb 2009 02:34:02 -0600
Message-ID: <40BA97F694BF944EBB093147C342379005EE406F@34093-EVS2C2.exchange.rackspace.com>
To: "Arron Eicholz" <Arron.Eicholz@microsoft.com>
Cc: <public-css-testsuite@w3.org>
Hi Arron,

 

Thanks for the mail, I didn’t realize I was actually looking for the wrong part of the spec.

Now it clear. Maybe such a hint can be appended to the test-file as well (for other people making the same mistake).

 

Best Regards,

Robert Stam

 

TallComponents

 

From: Arron Eicholz [mailto:Arron.Eicholz@microsoft.com] 
Sent: Sunday, February 22, 2009 8:31 PM
To: Robert Stam; public-css-testsuite@w3.org
Subject: RE: Explaination request: MS test (IE8, RC1) -- H4\at-rule-001.htm

 

I think you are getting your error handling mixed up with @keywords. This is not going into @keyword error handling because of the space right after the @ symbol. Why you ask?

 

The @ symbol is not followed by an IDENT as required by all @ keywords. And specifically @charset, @import, @page or @media, thus the @ symbol in this case is just an IDENT under the simple_selector rules in Appendix G and it’s an invalid IDENT for that matter. In this case the parsing ends up like this:

 

@ = invalid IDENT for an element_name

import = valid IDENT, for an element_name

“../../../Support/at-rule-red.css”; = is also attempting to be an IDENT for an element_name which it is also invalid.

Next the div = which is a valid IDENT for an element_name

Then we move onto parsing the { } of the div.

 

Even assuming that the @ and “…“; were valid element_names you couldn’t match an element on the page with that selector.

 

This is one of those real tricky cases that tricks your brain into thinking that this is a case that should follow the Invalid at-keyword rules in section 4.2 but it really is a 4.1.7 test case.

 

So the case is a correct case just a little complicated to determine where the rule is supposed to be handled from a spec perspective.

 

--
Thanks,
Arron Eicholz

 

From: public-css-testsuite-request@w3.org [mailto:public-css-testsuite-request@w3.org] On Behalf Of Robert Stam
Sent: Thursday, January 29, 2009 2:15 AM
To: public-css-testsuite@w3.org
Subject: Explaination request: MS test (IE8, RC1) -- H4\at-rule-001.htm

 

Hi,

 

Find below the test document.

Please notice that the @ and import identifier are separated by a space.

 

I expected that the invalid at-keyword would end at the ; after at-rule-red.css";

 

The specification says: “User agents must ignore <http://www.w3.org/TR/CSS21/syndata.html#ignore>  an invalid at-keyword together with everything following it, up to and including the next semicolon (;) or block ({...}), whichever comes first.”

 

This results in a red “Filler Text”, can somebody please give me an explanation why the div rule should be ignored as well?

 

Thanks in advance.

 

Best Regards,

Robert Stam

 

TallComponents

 

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

<html xmlns="http://www.w3.org/1999/xhtml">

    <head>

        <title>CSS Test: Parser error handling with invalid at-rules: @ invalid selector</title>

        <link rel="author" title="Microsoft" href="http://www.microsoft.com/" />

        <link rel="help" href="http://www.w3.org/TR/CSS21/syndata.html#parsing-errors" />

        <meta name="flags" content="invalid" />

        <meta name="assert" content="At-Rules start with an at-keyword (@ followed immediately by an identifier)." />

        <style type="text/css">

            @ import "../../../Support/at-rule-red.css";

            div

            {

                color: red;

            }

            *

            {

                color: green;

            }

        </style>

    </head>

    <body>

        <p>Test passes if the "Filler Text" below is green.</p>

        <div>Filler Text</div>

    </body>

</html>

 

Received on Friday, 27 February 2009 08:49:35 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 20 September 2010 17:51:57 GMT