- From: Norman Walsh <ndw@nwalsh.com>
- Date: Thu, 24 Apr 2008 07:46:49 -0400
- To: public-xml-processing-model-wg@w3.org
- Message-ID: <m24p9rv5jq.fsf@nwalsh.com>
/ Innovimax SARL <innovimax@gmail.com> was heard to say:
| let's say that we have (p:declare-step[1] | p:import[2]) in
| p:declare-step[3]
|
| if [3] is an atomic step (non sub-pipeline declared), what do mean a [1] ?
| in case [1] is a declaration of atomic step ? in case [1] is a declaration
| of a pipeline ?
|
| and what about having [2] in [3] when [3] is an atomic step ?
Bleh. I think it was a mistake to put import and declare-step in the
signature. I think we should change p:declare-step to:
<p:declare-step
name? = NCName
type? = QName
psvi-required? = boolean
xpath-version? = string>
(p:input |
p:output |
p:option |
p:log |
p:serialization)*,
((p:import | p:declare-step)*,
subpipeline)?
</p:declare-step>
That is: you should only be able to use p:import and p:declare-step when
you're defining a pipeline.
Be seeing you,
norm
--
Norman Walsh <ndw@nwalsh.com> | The first step towards madness is to
http://nwalsh.com/ | think oneself wise.--Fernando De Rojas
Received on Thursday, 24 April 2008 11:47:29 UTC