You are viewing a plain text version of this content. The canonical link for it is here.
Posted to soap-dev@xml.apache.org by Wouter Cloetens <wc...@raleigh.ibm.com> on 2000/11/24 13:27:53 UTC
Use of URN syntax
The samples and included functionality of Apache SOAP use (by
convention, I gather) URN identifiers as namespace identifiers in
target object URIs. E.g. the service manager interface uses
"urn:xml-soap-service-management-service". Encoded beans tend to use
the target object URN, followed by a colon and another identifying
string.
However, RFC2141 (http://www.ietf.org/rfc/rfc2141.txt), which defines
the URN syntax, defines a URN as the string "urn:", followed by a
namespace id of 1 to 32 alphanumeric characters or dashes (a dash isn't
legal as the first character), followed by another colon, followed by
an identifying string of one or more characters, with special
characters escaped in URL encoding style.
So the service manager URN quoted above is invalid for two reasons: the
namespace id is longer than 32 characters, and the second colon and
final identifying part is missing.
The RFC further defines rules for comparing URNs, but it looks like the
Apache SOAP code does not apply these, and simply does a case-sensitive
string compare, instead of a case-insensitive comparison of normalised
URNs.
I don't think it's high priority, but I propose that we fix this at
some point in time, before some interoperability issue comes up...
bfn, Wouter
--
http://www.workspot.net/~zombie/soap/
My opinions are irrelevant. They will be assimilated by my employer.
Re: Use of URN syntax
Posted by Sanjiva Weerawarana <sa...@watson.ibm.com>.
Hi Wouter,
Thanks for the clarification .. I am the src of of this mess - I started
using invalid URN URIs without checking the correct syntax.
We should change all the URIs in the samples/docs/code to be consistent.
Also, using the proper format will help us sort out the EJB URI stuff
I think - we can put the "instance id" after the second colon. However,
the instance ID problem needs to be solved in a first class way I
think.
Sanjiva.
----- Original Message -----
From: "Wouter Cloetens" <wc...@raleigh.ibm.com>
To: <so...@xml.apache.org>
Sent: Friday, November 24, 2000 7:27 AM
Subject: Use of URN syntax
> The samples and included functionality of Apache SOAP use (by
> convention, I gather) URN identifiers as namespace identifiers in
> target object URIs. E.g. the service manager interface uses
> "urn:xml-soap-service-management-service". Encoded beans tend to use
> the target object URN, followed by a colon and another identifying
> string.
>
> However, RFC2141 (http://www.ietf.org/rfc/rfc2141.txt), which defines
> the URN syntax, defines a URN as the string "urn:", followed by a
> namespace id of 1 to 32 alphanumeric characters or dashes (a dash isn't
> legal as the first character), followed by another colon, followed by
> an identifying string of one or more characters, with special
> characters escaped in URL encoding style.
>
> So the service manager URN quoted above is invalid for two reasons: the
> namespace id is longer than 32 characters, and the second colon and
> final identifying part is missing.
>
> The RFC further defines rules for comparing URNs, but it looks like the
> Apache SOAP code does not apply these, and simply does a case-sensitive
> string compare, instead of a case-insensitive comparison of normalised
> URNs.
>
> I don't think it's high priority, but I propose that we fix this at
> some point in time, before some interoperability issue comes up...
>
> bfn, Wouter
> --
> http://www.workspot.net/~zombie/soap/
> My opinions are irrelevant. They will be assimilated by my employer.
>
Re: Use of URN syntax
Posted by Sanjiva Weerawarana <sa...@watson.ibm.com>.
Hi Wouter,
Thanks for the clarification .. I am the src of of this mess - I started
using invalid URN URIs without checking the correct syntax.
We should change all the URIs in the samples/docs/code to be consistent.
Also, using the proper format will help us sort out the EJB URI stuff
I think - we can put the "instance id" after the second colon. However,
the instance ID problem needs to be solved in a first class way I
think.
Sanjiva.
----- Original Message -----
From: "Wouter Cloetens" <wc...@raleigh.ibm.com>
To: <so...@xml.apache.org>
Sent: Friday, November 24, 2000 7:27 AM
Subject: Use of URN syntax
> The samples and included functionality of Apache SOAP use (by
> convention, I gather) URN identifiers as namespace identifiers in
> target object URIs. E.g. the service manager interface uses
> "urn:xml-soap-service-management-service". Encoded beans tend to use
> the target object URN, followed by a colon and another identifying
> string.
>
> However, RFC2141 (http://www.ietf.org/rfc/rfc2141.txt), which defines
> the URN syntax, defines a URN as the string "urn:", followed by a
> namespace id of 1 to 32 alphanumeric characters or dashes (a dash isn't
> legal as the first character), followed by another colon, followed by
> an identifying string of one or more characters, with special
> characters escaped in URL encoding style.
>
> So the service manager URN quoted above is invalid for two reasons: the
> namespace id is longer than 32 characters, and the second colon and
> final identifying part is missing.
>
> The RFC further defines rules for comparing URNs, but it looks like the
> Apache SOAP code does not apply these, and simply does a case-sensitive
> string compare, instead of a case-insensitive comparison of normalised
> URNs.
>
> I don't think it's high priority, but I propose that we fix this at
> some point in time, before some interoperability issue comes up...
>
> bfn, Wouter
> --
> http://www.workspot.net/~zombie/soap/
> My opinions are irrelevant. They will be assimilated by my employer.
>