W3C home > Mailing lists > Public > public-wai-ert-tsdtf@w3.org > September 2006

Re: TCDL - EARL (Was: TCDL Draft E)

From: Christophe Strobbe <christophe.strobbe@esat.kuleuven.be>
Date: Tue, 19 Sep 2006 15:49:37 +0200
Message-Id: <6.0.0.22.2.20060919154238.02f54b68@mailserv.esat.kuleuven.be>
To: TSDTF <public-wai-ert-tsdtf@w3.org>

Hi Shadi, Vangelis,


At 11:00 19/09/2006, Shadi Abou-Zahra wrote:

>Hi,
>
>Evangelos Vlachogiannis wrote:
>>Shadi Abou-Zahra wrote:
>>>We should also consider adding EARL reports in the metadata section of 
>>>the test samples repository or in a new branch of the hierarchy.
>>I was thinking (if I am not missing something) that the "expected EARL 
>>reports" of the test samples (that would be for instance be compared with 
>>those produced by tools) would be generated automatically (XSLT with 
>>parameters) from TCDL (i.e. rules element  do we have all required info?).
>
>For EARL reports we need the following pieces:
>
>* Assertor - constant value (for example "TSD TF")

OK.

>* Subject - the test sample from the TCDL "files"

OK.

>* Test - the technique from the TCDL "rules" (?)

OK. 'techniques' is currently optional; it can contain
multiple 'technique' elements. Can you have an EARL
statement that does not 'Test'?

>* Result - has the following parts:
>+ Validity - pass/fail from the TCDL "rules" (?)

rules/rule/locations/@expectedResult

>+ Instances - pointers from the TCDL "locations"

OK.

>+ Message - human readable error message
>
>Ideally the error message is also somewhere in the TCDL but I am not sure 
>where. IMO, all the required info should be available in the TCDL and 
>generating EARL reports should be as simple as an XSLT file or similar 
>mechanism for translation.

We don't have real error messages in TCDL, but 'description' combined with 
'purpose' would be very informative. However, 'purpose' is not necessarily 
written in the kind of language we would expect in an error message.



>>This maybe raises an issue of either developing a mapping document of 
>>TCDL-EARL either a kind of integration. Am I with you (??)
>
>Actually, I am not sure if I am with you ;) I hope a mapping document will 
>not be necessary...

Well, XSLT (mentioned above) is a kind of mapping, isn't it? ;-)

Best regards,

Christophe


-- 
Christophe Strobbe
K.U.Leuven - Departement of Electrical Engineering - Research Group on 
Document Architectures
Kasteelpark Arenberg 10 - 3001 Leuven-Heverlee - BELGIUM
tel: +32 16 32 85 51
http://www.docarch.be/ 


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
Received on Tuesday, 19 September 2006 13:50:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:18:33 GMT