W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > August 2006

Re: Source without a step

From: Murray Maloney <murray@muzmo.com>
Date: Tue, 29 Aug 2006 13:03:03 -0400
Message-Id: <>
To: public-xml-processing-model-wg@w3.org
At 12:31 PM 8/29/2006 -0400, Norman Walsh wrote:
>Irrespective of the syntax, it will be possible to specify a source
>without a port, e.g.,
>   <p:input port="document" source="!result"/>
>What does this mean?

Unless I am totally confused, I think that the syntax above references a port,
but does not specify the step. That is what the email subject says, but the
prose above says "specify a source without a port."

My initial preference is #1 -- it is an error.

If we have to allow a default interpretation, then I would prefer #4, but I
am not sure that I could tell you which step!result is referenced in every
example that you could show me.

>Option 1: It's an error and we come back to it later when we're
>talking about abbreviations.
>Option 2: the "result" port of the preceding step. It's an error if
>there's no preceding step.
>Option 3: the "result" port of the preceding step or the parent step
>if there's no preceding step. It's an error if there's no preceding
>or parent step.
>Option 4: The "result" port of the nearest "upstream" step that has
>an output port named "result". It's an error if no such step exists.
>Or something else?
>                                         Be seeing you,
>                                           norm
>Norman Walsh
>XML Standards Architect
>Sun Microsystems, Inc.
Received on Tuesday, 29 August 2006 17:17:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:32:40 UTC