← Prev in month ← Prev in thread
Next in thread → Next in month →

Re: [sca-assembly] NEW ISSUE (v 1.2): Global domain channels creationinconsistent with pattern established by definitions.xml

From
Mike Edwards <>
Date
2011-02-15T09:31:16+00:00
ID
Thread
Re: [sca-assembly] NEW ISSUE (v 1.2): Global domain channels creationinconsistent with pattern established by definitions.xml
Logged as: http://www.osoa.org/jira/browse/ASSEMBLY-257

Yours, Mike

Dr Mike Edwards
 Mail Point 146, Hursley
Park

STSM
 Winchester, Hants SO21
2JN

SCA & Services
Standards
 United Kingdom

Co-Chair OASIS SCA
Assembly TC
 

IBM Software Group
  

Phone:
+44-1962 818014
  

Mobile:
+44-7802-467431 (274097)
  

e-mail:

  

 

 

From:
Eric Johnson <>

To:
OASIS SCA Assembly <>

Date:
14/02/2011 23:54

Subject:
[sca-assembly] NEW ISSUE (v 1.2): Global
domain channels creation inconsistent with pattern established by definitions.xml

Title: Global domain channels creation inconsistent with
pattern established by definitions.xml

Target: Assembly 1.2 WD 03 

Description:

The Assembly specification defines a "definitions.xml" file that
can be used to declare/define constructs that are shared across multiple
composites. This file currently encompasses implementation types, binding
types, policy intents, XML Schemas, WSDLs, and so on. Quite inconsistently
however, global domain channels (GDCs) are defined when deployed as part
of any composite deployed directly to the domain (not as part of implementation.composite).
The current version of the spec doesn't justify this inconsistency.

Besides the inconsistency, this triggers a number of related problems (which
could possibly be solved in a different way):

It obliges the "assembler" who defines channels
in a composite to know up front whether their composite will be deployed
directly to the domain, or rather be used as part of implementation.composite.
If the former, their "private" channels may in fact risk name
collisions with other already defined GDCs.

Since GDCs are presumed to be shared across numerous composites
& components, their lifecycle & scope is presumably distinct from
the composites that use the GDCs, and therefore the model should encourage
them to be deployed separately.

When tooling goes to examine a domain contribution to
understand the domain-level consequences of deployment, it will now have
to examine all of the composites in the contribution, in addition to what
it has to do now - scanning the definitions.xml file.Proposal:

(In abstract form) Remove the text from section 5.8 that describes how
channels defined in a composite directly contributed to the domain define
GDCs. Add text which describes how they can be declared via definitions.xml.

-Eric.

Unless stated otherwise above:

IBM United Kingdom Limited - Registered in England and Wales with number
741598. 

Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
3AU
← Prev in month ← Prev in thread
Next in thread → Next in month →