Re: Make p:http-request required?

Norman Walsh wrote:
> / Jeni Tennison <jeni@jenitennison.com> was heard to say:
> |> Having p:http-request and c:http-request may be asking for trouble.
> |
> | I think we should avoid using XML documents to configure a step unless
> | absolutely necessary. I don't think it's necessary in this case.
> 
> Ah, I think that gives me a better understanding of some of your
> earlier comments wrt parameters.
> 
> If we don't use a document to configure the step then parameters would
> be the obvious way to deal with headers, but that would interact badly
> with the decision we just made wrt parameters.
> 
> I have to say that for this step, using a document to configure it
> seems pretty reasonable.

Okaaay... so in what circumstances should steps use 
inputs+options+parameters for configuration and in what circumstances 
should they use a single custom XML input for configuration?

We *could* define the p:xslt step in the same way as the p:http-request 
step. It would look like:

<p:declare-step type="p:xslt">
   <p:input port="source" sequence="no"/>
   <p:input port="config" sequence="no" />
   <p:output port="result" sequence="no" />
</p:declare-step>

where the config input port looks like:

   <c:xslt >
     <c:parameters>
       <c:parameter name="foo" value="bar" />
       ...
     </c:parameters>
     <c:stylesheet>
       <xsl:stylesheet version="1.0">
         ...
       </xsl:stylesheet>
     </c:stylesheet>
   </c:xslt>

Why don't we? What makes it different from the p:http-request step?

It seems that we're making inconsistent design decisions across 
different steps. Both learnability and usability are greatly enhanced by 
consistency.

Cheers,

Jeni
-- 
Jeni Tennison
http://www.jenitennison.com

Received on Tuesday, 8 May 2007 10:05:31 UTC