RE: [ebxml-msg] Discussion: payload reference for use in SOAP body. Survey of options before writing this up.

["Dale Moberg" <dmoberg@cyclonecommerce.com>]

Title: Message

 I was leaning toward originally, just a bare fragment identifier --

 

uriref="#invoiceA"

 

Supposedly we can always use xml:id and not have to consult the schema to find information items with ID types.

 

 

But I think you misunderstand the escaping. (Or else I do.) I think that the escaping is done on characters in the content-id value.

 

For example, starting with

 

content-id: <foo%bar@something.com>

 

to construct the cid URI part of the URI-reference we remove the brackets

 

foo%bar@something.com

 

we escape reserved characters (the percent sign is reserved for indicating hex escapes so we have to escape the percent sign)

foo%25bar@something.com

 

 

 

We then append the scheme

 

cid:foo%25bar@something.com

 

And that is the URI part.

 

Now add the fragment (start is indicated by the hash mark #)

 

cid:foo%25bar@something.com#xmlns(s12=http://www.w3.org/2003/05/soap-envelope) xpointer(/s12:envelope/s12:body)

 

That is the URI-reference, absolute path format.

 

You are doing the escaping backwards it seems to me. We have to escape some of the cid's character values because a cid can have characters in it that are not legal in URIs.

 

Whew.

 

 

 

 

-----Original Message-----
From: Matthew MacKenzie [mailto:mattm@adobe.com]
Sent: Friday, May 21, 2004 9:14 AM
To: ebxml-msg@lists.oasis-open.org
Subject: Re: [ebxml-msg] Discussion: payload reference for use in SOAP body. Survey of options before writing this up.

Ok, so in the case of the SOAP Body, do we allow an arbitrary CID, e.g.

cid:some@crap#xmlns(s12=http://www.w3.org/2003/05/soap-envelope) xpointer(/s12:envelope/s12:body) 

<some@crap%??xmlns%??=http%??%??%??www.w3.org%??2003%??05%??soap-envelope%??xpointer%??s12%??envelope%??s12%??body%??

I don't really like this....

How about we use xml:id?

xmlid:invoiceA

...

<soap:Body>

<Invoice xml:id="invoiceA" />

</soap:Body>

On May 21, 2004, at 1:32 AM, Dale Moberg wrote:

From RFC  2392

     Notes: In Internet mail messages, the addr-spec in a Content-ID

     [MIME] or Message-ID [822] header is enclosed in angle brackets

     (<>).  Since addr-spec in a Message-ID or Content-ID might contain

     characters not allowed within a URL; any such character (including

     "/", which is reserved within the "mid" scheme) must be hex-encoded

     using the %hh escape mechanism in [URL].

   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>

   Reversing the process and converting URL special characters to their

   % encodings produces the original cid.

[I think there is a typo. I think it should read

     Content-ID: foo4%foo1@bar.net

%25 == "%"]

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

From: Matthew MacKenzie [mailto:mattm@adobe.com]

Sent: Thursday, May 20, 2004 6:26 PM

To: Dale Moberg

Cc: ebxml-msg@lists.oasis-open.org; Pete Wenzel

Subject: Re: [ebxml-msg] Discussion: payload reference for use in SOAP body. Survey of options before writing this up.

hex-escaped-cid-value-minus-opening-and-closing-brackets

??

On May 20, 2004, at 8:44 PM, Dale Moberg wrote:

I am following up with the namespace qualification for the URI-reference fragment using an xpointer scheme-based approach,

CID:hex-escaped-cid-value-minus-opening-and-closing-brackets#xmlns(s12=http://www.w3.org/2003/05/soap-envelope) xpointer(/s12:envelope/s12:body) 

[This is how the fragment indicates the namespace qualification for soap12. Note this doesn't mean that the prefix in the soap message has to be s12 but only that whatever prefix was actually used is associated with the URI given in the xmlns() pointer-part.]

This is from the http://www.w3.org/TR/xptr-xmlns/ which basically writes up the above piece of syntax for the xmlns() pointer-part.. The namespace value has to be escaped. The framework allows an optional whitespace string between the items in the fragment's pointer-parts.

___________________________

Matthew MacKenzie

Senior Architect

IDBU Server Solutions

Adobe Systems Canada Inc.

http://www.adobe.com/products/server/

mattm@adobe.com

+1 (506) 871.5409

___________________________

Matthew MacKenzie

Senior Architect

IDBU Server Solutions

Adobe Systems Canada Inc.

http://www.adobe.com/products/server/

mattm@adobe.com

+1 (506) 871.5409