Re: Non-primary port connection between non-contiguous steps


I agree with you. Context-dependent names for p:input etc. would have made this kind of misunderstanding impossible.

While we're at bashing the spec :-) : Within p:declare-step, having container elements that enclose p:input, p:output, p:option, p:log and p:serialization on the one hand and the remaining content on the other hand (which could have been named e.g. "head" and "body") would have made it much easier to hand-edit step declarations without inadvertently mixing up instances from both groups, at least in my experience.


Am 06.06.2012 13:25, schrieb Imsieke, Gerrit, le-tex:
> Yves, you are not alone.
> This context dependence of p:input/p:output element semantics was hard for me to understand initially. Only after I had grasped it finally, I saw that there was indeed a kind of caveat in the spec:
> Then I thought for some time that the *declaration* and the *connection* use of p:input should have been discerned by giving context-dependent names to the elements (e.g., p:declare-input and p:connect-input).
> I think it might facilitate things for new users. It will be interesting to learn about the input/output naming discussions from the spec editors.

Dr. Yves Forkl - Softwareentwicklung
SRZ, Bessemerstr. 83-91, 12103 Berlin | Firmengruppe:
tel +49 30 75301-335 | fax +49 30 75301-11335

Satz-Rechen-Zentrum Hartmann+Heenemann GmbH&Co. KG
Sitz Berlin | AG Charlottenburg | HRA 8089
Komplementärin Satz-Rechner-Betriebsgesellschaft mbH
Sitz Berlin | AG Charlottenburg | HRB 4905
Geschäftsführer: Walter Fock

Received on Thursday, 7 June 2012 08:55:57 UTC