W3C home > Mailing lists > Public > public-cdf@w3.org > April 2007

Re: 2nd Call: is there a test suite and harness?

From: Chris Lilley <chris@w3.org>
Date: Wed, 11 Apr 2007 20:57:10 +0200
Message-ID: <1875976086.20070411205710@w3.org>
To: "~:'' ありがとうございました。" <j.chetwynd@btinternet.com>
Cc: public-cdf@w3.org

On Wednesday, April 11, 2007, 6:07:29 PM, ~:'' wrote:

~?> Chris,

~?>> Could you point to the section or sections of the CDF  
~?> specifications that this tests?

~?> this test relates to CSS pseudo-class events.
~?> I'm having difficulty finding specific reference to CSS within CDF  
~?> specifications. Is this by design?

No, but it does indicate that your test might be for a different specification.

~?> Please find attached a selection of possibly suitable sections.

In general, tests are designed to exercise the testable assertions of a specific specification.

~?> Please let me know which you consider the attached test or a  
~?> variation, suitable for.

The working group, as with most others, generally proceeds by analyzing the spec and then producing tests for specific parts of the spec; rather than producing tests and wondering what they test.

I'm encouraged by your desire to contribute to the CDF tests, and suggest that you adopt a similar methodology.

~?> regarding RDF I was disturbed to find no suitable reference to RDF.

OK, why?
For HTML, the relevant spec says that RDF causes a syntax error and makes the document invalid. Hence things like GRDDL to smuggle RDF into HTML. Or comments (!). While regrettable, CDF does not change that for WP1. For WP2, ie Compound Documents by Inclusion, I hope it does change things.

For SVG, you just put your RDF into a metadata element. CDF doesn't change that.

For CSS, you can't put RDF in CSS.

What else were you expecting to see, for RDF in CDF Compound Documents by reference?

~?> Jonathan Chetwynd

~?> Compound Document by Reference Framework 1.0

~?> 2.12 Events

~?> note no reference to pseudo-class events is made. is this intentional?

Does the CSS 2.1 spec say that pseudo-classes are event driven, or define a list of the events that trigger them?

~?> Compound Document Use Cases and Requirements Version 2.0

~?> 1.3.6 CSS

~?> makes no comments on pseudo-class events
It makes no reference to the properties, values, and syntax, either. it just imports them unchanged.

Specific tests that depend on CDF would be appropriate. For plain HTML+CSS, see the CSS test suite. For plain SVG+CSS, see the SVG test suite.

~?> 2.6.1  Synchronization of HTML content with audio visual content:

~?> ". starting and stopping playback of an embedded video (or audio) in  
~?> response to interaction with another HTML element"

~?> if the test provided is not quite right, another reduced test case  
~?> could be provided.

~?> http://www.peepo.co.uk is an example of this type of event handling  
~?> without use of script.
~?> the user mouses over an icon, which raises a further symbol, that  
~?> whilst moused over plays a sound.

Could be the basis for a test. The context here though is more pausing and resuming an audio or video. 


~?> Compound Document by Reference Use Cases and Requirements Version 1.0

(Those are mainly requirements on other specifications, not directly testable in content).




-- 
 Chris Lilley                    mailto:chris@w3.org
 Interaction Domain Leader
 Co-Chair, W3C SVG Working Group
 W3C Graphics Activity Lead
 Co-Chair, W3C Hypertext CG
Received on Wednesday, 11 April 2007 18:57:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:10:41 GMT