search

xproc / 3.0-steps

Repository for change requests to the standard step library and for official extension steps
10 stars 7 forks source link

The result port on p:compare is not primary? #549

Closed ndw closed 4 months ago

ndw commented 6 months ago

That surprises me. Was that intentional, or an oversight?

gimsieke commented 6 months ago

Maybe this was intentional. In 1.0, the (single) output port result was explicitly designated as primary="false". But I’m not sure what the 1.0 reasoning for this was.

xml-project commented 6 months ago

Oh, suprise here too. And my implementation has "result" as primary. May be it is just a copy-paste from 1.0, but given the other steps it would feel more natural to me, to have "result" as primary.

xatapult commented 6 months ago

I definitely think it should be primary.

ndw commented 6 months ago

I can't recall what he rationale was in 1.0. Possibly that it was "destructive" in the sense that it doesn't return the source document modified, it returns c:result marked true or false. But we fixed most of those in 3.0. I think it's very likely that Achim is correct and it's just an oversight.