See also: IRC log
-> http://www.w3.org/XML/XProc/2014/11/12-agenda
Accepted.
-> http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2014Nov/0004.html
Norm thanks Jim for taking minutes.
<jfuller> tbey were terrible
Accepted.
Proposed: 19 Nov 2014 does anyone have to give regrets?
No regrets heard.
<scribe> No progress reported.
Vojtech: Two things. If you have PIs or comments, you need to have a p:inline, but what does it mean
<jfuller> https://xquery.github.io/xproc-specification/langspec/xproc20/head/diff.html#p.inline
Some discussion of what it means to have multiple top-level elements.
Norm: No, I don't think we should
do that.
... For exactly the reasons that Vojtech raised, I think if we
have multiple top-level elements, you need the p:inline
wrapper.
Jim: I thought comments and PIs would always throw an error.
Vojtech: No, if there's only one top-level element, they're ok.
Jim: So instead of foreign element(s), you're saying foreign element (not plural) and go for the singular.
Vojtech: The spec of p:inline is any element, but we also allow PIs and comments around it.
Norm: I think if you have multiple documents, you need the wrapper.
Vojtech: At the moment, if you
use p:inline then comments and PIs are ignored outside the
p:inline elements.
... If you remove the wrapper, then they become part of the
document.
... Another thing, at the moment the exclude-inline-prefixes
attribute is on p:inline and p:input.
... But I think the p:inline should be optional in the other
places as well. On output, on connections, etc.
Norm: I agree.
Jim: I didn't propose that because I was trying to make the change very narrow.
Norm: I think if the p:inline can be left out in one context, it should be legal to leave it out everywhere that it can occur.
Jim: What about expand-text?
Norm: I think that needs to be available on all of the possible parents of p:inline.
<scribe> ACTION: ACTION A-257-01 Jim to revise the p:inline proposal to cover all of the places where p:inline can occur and to handle expand-text as well as exclude-inline-prefixes. [recorded in http://www.w3.org/2014/11/12-xproc-minutes.html#action01]
-> https://ndw.github.io/specification/
Norm thinks he made it work.
<jfuller> +1
<jfuller> yes
Norm: Consensus is that we should do this. The editor asserts it's entirely editorial. No technical changes except putting p:template and p:in-scope-namespaces into the steps document.
Vojtech: There are also references to p:try and such. They are still in the language spec.
Norm: Yes.
Vojtech: The document could also
be called something else; the compound steps are also "standard
steps".
... I would expect them in the steps document, but that would
probably not be as easy to do.
Norm: But if we moved them all into the steps document, we'd really be back to a single document...
Vojtech: I see that for
p:declare-step or p:pipeline, but things like p:viewport or
p:choose are more like standard steps that we provide out of
the box.
... It's really just atomic steps.
... The main spec is compound steps and the rest of the
language.
Norm: That's right.
Vojtech: On one level all these things like viewport and for-each are really just steps.
Norm: So should we wait on splitting...
Vojtech: No, but maybe we should
consider moving viewport, for-each, etc. into the steps
document.
... The second document is really a library of atomic steps;
maybe that's ok.
<jfuller> I think its ok as well
Proposal: Accept the two-spec version as the new consensus draft.
No objections heard.
Accepted.
Norm: I will make a pull request and make it so.
Norm: It's a useful collection of first-time-user-concerns.
<jfuller> http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2014Nov/0007.html
Norm: I propose that we make one-or-more issues out of these and see that we address them.
The WG walks through Liam's message, muses about the changes.
None heard.
Adjourned.