- From: Jeni Tennison <jeni@jenitennison.com>
- Date: Thu, 07 Jun 2007 21:17:43 +0100
- To: XProc WG <public-xml-processing-model-wg@w3.org>
Henry S. Thompson wrote: > Innovimax SARL writes: > >> <p:declare-step type="p:split-sequence"> >> <p:input port="source" sequence="yes"/> >> <p:output port="result" sequence="yes"/> >> <p:output port="secondary" sequence="yes"/> >> <p:option name="test" required="yes"/> >> </p:declare-step> >> >> output to the !result until $test is false, then it will output the >> rest of the sequence to !secondary > > I like this, further to our discussion just now. > > Provided we agree the semantics of position() in XPath expressions > evaluated by steps with sequence inputs, and further agree that we > will implement last() 'correctly' in the same cases, I propose we > adopt this, and remove p:matching-documents, p:head and p:tail. > > <p:head count="n"/> becomes <p:split-sequence test="position()<=n"/> > <p:tail count="n"/> becomes <p:split-sequence test="last()-position()<=n"/> Looks good to me, but can we call the ports 'pass' and 'fail' or 'true-result' and 'false-result'? Jeni -- Jeni Tennison http://www.jenitennison.com
Received on Thursday, 7 June 2007 20:17:45 UTC