RE: [ebxml-iic] comments on TestFmwk, response.

[Jacques Durand <JDurand@fsw.fujitsu.com>]

Title: comments on TestFmwk, response.

Mike:

totally agree. I think you got the last word...

Can you update and send to the list? (keep the diff visible)

 

Given these are editorial changes, I propose that this be the

actual 1.0 version of teh spec. However, as this is nonetheless a change,

I would understand if if anyone objects to these updates:

in which case we would revote the updated spec. So let me know this week....

 

Beside this update, we still need some cosmetic brush-up,

that I will check with Jamie Clark from OASIS, to make sure

we got right details such as IP statement, etc.

 

Note that our site is in turmoil right now, as it is in reorg

process, so we should share docs by mail list.

 

regards,

 

Jacques

-----Original Message-----
From: Michael Kass [mailto:michael.kass@nist.gov]
Sent: Monday, March 24, 2003 3:52 PM
To: Jacques Durand; ebxml-iic@lists.oasis-open.org
Subject: Re: [ebxml-iic] comments on TestFmwk, response.

Jacques,

 

   My one suggestion is:

 

Instead of  using

to the Mute action of the Test Service associated with the Response URL.  ...which implies interop testing only (service mode), and not also conformance testing (connection mode)

 

we use

 

with a Mute action element,  to the test component (Test Service or Driver)  associated with the Response URL.

Comments?

 

Mike

 

 

----- Original Message -----

From: Jacques Durand

To: 'ebxml-iic@lists.oasis-open.org'

Sent: Sunday, March 23, 2003 4:04 PM

Subject: [ebxml-iic] comments on TestFmwk, response.

FYI, comments from Serm (NIST), on some editorial ambiguities.
I have tried to adress these in the following attached update of Test Framework  ("rev1", see the diff in red), please review.

NOTE: the approved version of Test Framework (updated here) has not been posted yet, will be very soon.
(Monica: that is the version you asked for).

regards,

Jacques
<<TestFmk0.1-rev1.zip>>

---------------------------------------------
I couldn't post email to the iic list so I forward this comment to you.
thank you,
serm
----- Original Message ----- From: Boonserm (Serm) Kulvatunyou <mailto:serm@nist.gov>
To: ebxml-iic@lists.oasis-open.org <mailto:ebxml-iic@lists.oasis-open.org>
Sent: Friday, March 21, 2003 6:22 PM
Subject: possible confusion in the test framework
In section 3.2.4.2 - Test Service Actions, the term 'requestor' and 'requestor Test Service' has been used in several places (look in the response destination) which I am sure if applies in all cases. To my understanding it should be 'response URL'.

<Jacques> agree "requestor" is ambiguous (a remnant of some previous sections we removed on "requestor/responder"). Response messages are always sent to response URL, as defined in 3.2.3., so we should make that more explicit.

I think the simplest reason is the requestor could be either Test Driver or Test Service. If the requester is Test Driver and it is conformance testing, then response destination is Test Driver. If the requester is Test Driver and it is interop testing, then response destination is another Test Service. So response URL should always be used.

On line 703-704, "The action responds with a notification message to the requestor ....". This confused me. First, "Notification" was used for reporting (to the test driver). Then it has the word 'requestor'. Should the sentence be "The action responds to the ResponseURL about ...."?

<Jacques> Right. Bad wording. I think we should ban "requestor" from all this section, as this is also too restrictive on the possible test harness configurations.

Or I am missing something :).
Thank you,
- serm