|
This is a known issue, here is an ancient reference
from seven years ago:
I remember as it did cause an interoperability issue with
two ebMS2 implementations I was working with in 2004. These
were custom implementations that never participated in any Drummond test.
We could post an errata for ebMS2. We also have a UPA
violation with the ebMS 2.0 XSD that we discussed a year ago, did we ever
post the patched XSD ?
And we could think about an errata for ebMS 3 Core, it
has a few minor issues too.
Pim
The Drummond Group has been
conducting interoperability tests for ebMS2 for 8 years now (which has included
Cleo and Oracle), and as far as my reasearch has shown, we have no history of
this being an interop issue -- at least one that required any sort of special
"consensus item" discussion and resolution.
Sander Fieten
wrote:
Hi
guys,
there’s a error in the example in section 2.1.2 of the V2 ebMS
spec with regard to the value for Content-id. The example shows the cid
[incorrectly] without the angle brackets (“<“ “>”) in the start
parameter of the Content-Type MIME header field of the multi part. The
Content-Id MIME header of the related part however has the cid enclosed in
brackets.
Should we correct this? It seems that it can lead to
confusion (see the mail below).
Regards,
Sander
------
Forwarded Message
From: Andy Evett <AEvett@cleo.com>
Date: Mon, 5 Oct 2009
15:24:51 +0200
To: Julien R <jurenit.jurenit@gmail.com>, Michael O'Connell
<mike@flame.co.za>, Sander Fieten <sander@fieten-it.com>
Cc: <ebxml-dev@lists.ebxml.org>
Subject:
RE: [ebxml-dev] EbMS interopability: Content-type start parameternot
matching content-id
Hi all,
Cleo chose to follow the
example in the ebMS specification. LexiCom does accept incoming messages
with or without the enclosing '<' and '>' in the start parameter, but
for outgoing messages LexiCom will leave off the enclosing
characters.
I do see though that even the ebMS specification itself is
inconsistent as examples in Appendix B show the syntax suggested. If
this is causing interoperability issues, please contact support-ec@cleo.com and
we will look at adding a formatting option.
Regards,
Andy
Evett
Cleo Communications
-----Original Message-----
From:
Michael O'Connell [mailto:mike@flame.co.za]
Sent: Monday, October
05, 2009 8:08 AM
To: Sander Fieten
Cc: Julien R; ebxml-dev@lists.ebxml.org
Subject: Re:
[ebxml-dev] EbMS interopability: Content-type start parameternot matching
content-id
Hi Julien, Sander
According to the Content-ID and
Message-ID Uniform Resource Locators
Spec http://www.ietf.org/rfc/rfc2392.txt the use of
content-id <> is
defined
as:
A "cid" URL is
converted to the corresponding Content-ID
message
header [MIME] by
removing the "cid:" prefix, converting the
%
encoded character to
their equivalent US-ASCII characters,
and
enclosing the remaining
parts with an angle bracket pair,
"<"
and ">".
For example, "cid:foo4%25foo1@bar.net" corresponds
to
Content-ID: <foo4%25foo1@bar.net>
Also,
Content-ID:
<messagepackage-123@example.com
<mailto:messagepackage-123@example.com>
>
Looks a lot like a MS-Word interpreted paste of a email-like
string.
Hence, perhaps, the mailto:
attribute?
Regards,
Michael O'Connell
Flame Computing
(FMS)
On Mon, 2009-10-05 at 14:39 +0200, Sander Fieten wrote:
>
Hi Julien,
>
> I think you’re right about that both the Lexicom
product and the
> example in the ebMS v2 specification do not comply
with RFC 2387. The
> cid given as the start id should exactly match the
one used in the
> Content-Id MIME header. So in both cases the value of
start parameter
> should be enclosed in “<“ and
“>”.
>
> I assume the value “SOAP” is here as an example and
was not in the
> real message as “SOAP” does not comply with RFC 822
regarding the
> formatting of message id’s?
>
>
Regards,
> Sander
>
>
> On 05/10/2009 14:13, "Julien
R" <jurenit.jurenit@gmail.com>
wrote:
>
> Hi
all,
>
>
We are testing the EbMS
product Lexicom (Cleo communications)
>
and found the following
issue:
>
>
The Content-type start
parameter value doesn't exactly match
>
the content-id value: The <
and > symbols are not present in
>
the content-id.
Example:
>
>
Content-Type:
multipart/related; type="text/xml";
>
boundary="--------CLEOebXML.Boundary.1253712627668.yradnuoB.LMXbeOELC--------";
start=SOAP
>
>
First lines of body:
>
----------CLEOebXML.Boundary.1253712627668.yradnuoB.LMXbeOELC--------
>
Content-Id:
<SOAP>
>
Content-Type: text/xml
>
>
We think Lexicom doesn't
follow RFC 2387 because the
>
content-id and start parameter
should match exactly.
>
>
We have also checked the EbMS
v2.0 specification and it
>
contains the following example
in section 2.1.2 message
>
package:
>
Content-Type:
multipart/related; type="text/xml";
>
boundary="boundaryValue"; start=messagepackage-123@example.com
>
>
--boundaryValue
>
Content-ID: <messagepackage-123@example.com
>
<mailto:messagepackage-123@example.com>
>
>
>
Here the start and content-id
parameters also do not match. We
>
believe this is a mistake in
the EbMS specifications. Further
>
in Appendix B there is an
example where the start and
>
content-id do match
exactly.
>
>
I want to know how this should
be implemented and how other
>
EbMS products implement
this.
>
>
Best regards,
>
Julien
>
------ End of
Forwarded
Message
---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that generates
this mail. Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
|