OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ebxml-iic message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: Final DRAFT for today's discussion and voting


Title: Minutes last meeting, next call
Mike:
 
as discussed today, that looks OK assuming we have a decent way to
branch-off a thread or not, based on a TestAssertion outcome.
 
Not sure I understand the concern on object hierarchy:
(isn't that similar to having in a prog language a simple instruction invoking a procedure?)
Though I understand parameter scoping would get more difficult.
 
Jacques
 
-----Original Message-----
From: Michael Kass [mailto:michael.kass@nist.gov]
Sent: Monday, April 12, 2004 7:47 AM
To: Jacques Durand; ebXML IIC - main list (E-mail) (E-mail)
Subject: Final DRAFT for today's discussion and voting

Jacques and all,

 

   Here is the URL for the final draft of the Test Framework specification.

 

 

http://xsun.sdct.itl.nist.gov/~mkass/ebxml_iic/IIC_ebXMLTestFramework%20V1.1_04_12_04_final_draft.zip

 

 

  All issues have been addressd that were identified in the last conf call.

I am ready to vote for the document, however I wish to discuss one  item currently in the spec that I believe should be removed ( branching from within a Test Step ).  I believe that doing so breaks the object hierarchy ( and the parameter scooping ) of the Test Framework, and is unnecessary.  I believe that using the natural object hierarchy of :

 

TestSuite

TestCase

Thread
TestStep

 

 makes sense.

 

 I believe that a hierarchy of:

 

TestSuite

TestCase

Thread
TestStep

Thread

 

  does not.  Especially when one thinks of Threads in a BPSS sense ( as a container for a particular set of related actions (i.e. a particular part of a business process ), and thinks of TestSteps as the "atomic" level operation that is simply a PutMessage or GetMessage.  The logical branching from a Test Step destroys this hierarchical relationship, and the readability of a script ( and its natural similarity to a business process definition )

 

Attached is the document for review.

Sorry I couldn't get it out sooner.. worked on it all weekend.

 

Mike

 

 

-----Original Message-----
From: Jacques Durand [mailto:JDurand@us.fujitsu.com]
Sent: Tuesday, April 06, 2004 8:16 PM
To: ebXML IIC - main list (E-mail) (E-mail)
Subject: [ebxml-iic] Minutes last meeting, next call

 

 

All:

Next call:

Time: Monday Apr 12, 2pm PT
Host: Fujitsu
Toll only :  1-512-225-3050
Participant code: 89772

We'll get another iteration of the spec by theend of thsi week and decide whether to submit it to vote
this Monday 12 (vote will be by email, over 1 week)
The new iteration should address the "notes" in the attached minutes.

Mike: can you especially check "note #2" and send a note to the list on current status for this?
we have to follow-up on this to make
sure that we have a clear scheme for terminating a test case (exit statement?)

Regards,

Jacques
<<IIC_April_05_04_minutes.txt>>

 



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]