W3C home > Mailing lists > Public > public-mobileok-checker@w3.org > June 2007

RE: ACTION 515 -CSS stuff partially done

From: Abel Rionda <abel.rionda@fundacionctic.org>
Date: Fri, 29 Jun 2007 14:15:55 +0200
Message-ID: <09700B613C4DD84FA9F2FEA5218828197A1963@ayalga.fundacionctic.org>
To: "Sean Owen" <srowen@google.com>
Cc: <public-mobileok-checker@w3.org>


Hi Sean,

First of all,sorry because our delay in writing a response
on this,but this week has been very hard for us.
Regarding with your comments, 

> I also believe we need to keep test-specific results out of the
> intermediate doc - this "styleSheetSupport" element needs to be part
> of the results instead of intermediate document I think.

Our first approach on this was making the XSLT CSS tests only based 
on intermediate doc. But yes ,this can not be done in an elegant way
without having CSS serialized inside moki doc,so we will eliminate 
the test-specific info from moki and will use java functions from XSLT
(similar approach to CachingTest).

 > First would it be more desirable to re-use the <position> tag syntax
 > that we conceived for the result document? this presents a similar,
 > but different tag called <location>. We also used <info> rather than
 > <description> in the results document.

Yes, sure.We will update this format both in CSS and XHTML validation
stuff.

As soon as we can we will update this changes in CVS.

Regards,

Abel.

-----Mensaje original-----
De: public-mobileok-checker-request@w3.org en nombre de Sean Owen
Enviado el: jue 28/06/2007 22:01
Para: public-mobileok-checker@w3.org
Asunto: Re: ACTION 515 -CSS stuff partially done
 

Also I see that the code that validates the XHTML uses a
<location>line, col</location> syntax. Mind if I update this too? I
think we need to get these details straightened out before we start
writing tests in earnest.

Sean

On 6/28/07, Sean Owen <srowen@google.com> wrote:
> Thoughts on this? I'd like to modify the <location> element naming to
> be consistent with the result document naming.
>
> I also believe we need to keep test-specific results out of the
> intermediate doc - this "styleSheetSupport" element needs to be part
> of the results instead of intermediate document I think.
>
> On 6/26/07, Sean Owen <srowen@google.com> wrote:
> > This is good stuff -- I have two comments on the result format.
> >
> > First would it be more desirable to re-use the <position> tag syntax
> > that we conceived for the result document? this presents a similar,
> > but different tag called <location>. We also used <info> rather than
> > <description> in the results document.
> >
> > I remain a little concerned that the line between the "preprocessing"
> > and "tests" is becoming blurred and the result will be difficult to
> > comprehend. It makes sense to produce an intermediate document that
> > records the result of accessing a CSS resource and even parsing it.
> >
> > Here I think we've gone a step beyond that and put mobileOK
> > Basic-specific information into this document -- that is, which
> > properties in the stylesheet aren't allowed by mobileOK Basic.
> >
> > I think this logic should be implemented only in the test and
> > presented in the results document. Thoughts?
> >
> > On 6/26/07, Abel Rionda <abel.rionda@fundacionctic.org> wrote:
> > >
> > >
> > >
> > >
> > > Hi everyone,
> > >
> > >
> > >
> > > Although code in CVS is being updating with the new test format, we have
> > > committed
> > >
> > > some changes related to our CSS action.
> > >
> > >
> > >
> > >
> > >
> > > *We have introduced stylesheet block. So far this stylesheet tag can be
> > > built from
> > >
> > >  linked  or embedded CSS Resources. In both cases W3CValidator will process
> > > any
> > >
> > > @import rule found (But we lose the retrieval information of imported CSS).
> > > So there will be
> > >
> > > an stylesheet block per top level CSS resource)
> > >
> > >
> > >
> > > *CSS validity messages from W3CValidator tool are allocated inside a
> > > CSSValidity block with
> > >
> > > the same structure  used for grammar
> > > validation.(error-location-description).
> > >
> > >
> > >
> > > *For each CSS test we have an special block with the information needed to
> > > pass the test
> > >
> > > via XSLT. Currently we only have the information for Style Sheet Support
> > > Test.
> > >
> > >
> > >
> > > To illustrate these changes see the following extract from moki document.
> > >
> > >
> > >
> > >
> > >
> > > <stylesheets>
> > >
> > >       <stylesheet type="embedded">
> > >
> > >
> > > <URI>http://idi.fundacionctic.org/bk/google.xhtml</URI>
> > >
> > >          <CSSValidity valid="false">
> > >
> > >             <error code="-1">
> > >
> > >                <location type="LineAndColumn">1, 0</location>
> > >
> > >                <description>Property colo doesn't exist</description>
> > >
> > >             </error>
> > >
> > >          </CSSValidity>
> > >
> > >          <stylesheetSupportTest>
> > >
> > >             <error code="-1">
> > >
> > >                <location type="LineAndColumn">1, 0</location>
> > >
> > >                <description>float:left</description>
> > >
> > >             </error>
> > >
> > >          </stylesheetSupportTest>
> > >
> > >       </stylesheet>
> > >
> > >    </stylesheets>
> > >
> > >
> > >
> > >
> >
>
Received on Friday, 29 June 2007 12:19:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:13:03 GMT