RE: April CCXML 9_2_3_B: handling of invalid event name in <send> [cc] ISSUE-719

Petr,

We are in the process to address all ISSUES related to IR. The goal is to finalize the test-suite during the PR transition.
Please explicitly confirm that you accept the proposed resolution or after one week we will consider it implicitly accepted.
If you need clarifications, please ask them very soon.

Paolo Baggia
Author of CCXML-IR Plan

ISSUE-734:

We reviewed this ISSU-719 (see separate email) and also ISSUE 734.

The behaviour is correct and coherent with CCXML specification, to clarify
it CCXML PR will include the following clarification:

9.3.5: error.send.failed

This error event MUST be thrown when a <send> could not be completed for a reason not covered by another error.send.* event [[[(for example, due to an invalid event name specified on <send>)]]].

=================================
From: RJ Auburn <rj@voxeo.com> 
Date: Thu, 29 Jul 2010 10:51:38 -0400
To: W3C Voice Browser Working Group <w3c-voice-wg@w3.org> 
Message-Id: <092E4380-2EE1-4176-8D77-03EDB42FB9B0@voxeo.com> 

> 
> Tracked as ISSUE-734. 
> 
> 	RJ
> 
> On Jul 26, 2010, at 4:59 AM, Petr Kuba wrote:
> 
>> Hello www-voice,
>> 
>> In 9_2_3_B, the following code is used:
>> 
>> 	<send targettype="'ccxml'" target="session.id"
>> 		name="'.name_prepare_new'"/>
>> 
>> Since the name value is invalid (starts with a dot) the <send> results in throwing an error event.
>> 
>> The problem is that the test expects error.send.* event to be thrown. We believe that error.semantic should be thrown in this case where the attribute value is incorrect.
>> 
>> See the Specification, Section 9.3.2 error.semantic:
>> 
>> "This error event MUST be thrown when there is a semantic error in a CCXML element ( e.g. passing an incorrect value for an attribute, etc.)."
>> 
>> The error.semantic is used in every situation where attribute value is incorrect so why not here?
>> 
>> On the other hand, the only usable error.send.* event would be "general" error.send.failed (9.3.5):
>> 
>> "This error event MUST be thrown when a <send> could not be completed for a reason not covered by another error.send.* event."
>> 
>> Please clarify on this.
>> 
>> Thanks,
>> Petr
>> 
>> -- 
>> Petr Kuba, Project Manager
>> OptimSys, s.r.o
>> kuba@optimsys.cz
>> Tel: +420 541 143 065
>> Fax: +420 541 143 066
>> http://www.optimsys.cz

Received on Friday, 27 August 2010 12:25:07 UTC