XML Processing Model WG telcon

04 Feb 2010


Paul Grosso, Alex Milowski, Henry S. Thompson, Toman Vojtech
Norm Walsh, Mohamed Zergaoui
Henry S. Thompson (pro tem)
Henry S. Thompson


HST: Minutes from 2 weeks ago approved
... Regrets for 11 Feb? . . . None

<PGrosso> Paul's attendance on the 11th is at risk.

Possible new issues

<Vojtech> http://www.w3.org/XML/XProc/2010/01/lc-comments/

HST: We believe that the 9 issues listed there (most recent one is 21 January) have been dealt with


TV: Various cases wrt option and variable with the same name:
... 1) Wrt option and variable in the same pipeline, in TV's impl. that's a static error
... 2) A var. in a sub-pipe which appears to shadow an option in a surrounding pipe

HST: 3.2 Scoping of names says. . .


The scope of option and variable names is determined by where they are declared. When an option is declared with p:option (or a variable with p:variable), unless otherwise specified, its scope consists of the sibling elements that follow its declaration and the descendants of those siblings.

HST: The strict solution would be to make XS0004 parallel XS0002

TV: The alternative is to allow shadowing

HST: Do we shadow anywhere else?

TV: Yes, sort of -- counters in nested 'for-each' loops

HST: Quite different

TV: Does XSLT allow shadowing?
... What about a template var. with the same name as a global var?

AM: XSLT does allow shadowing

HST: Yes, global by local, or local by local, but not global with global (!)
... We don't have global vs. local

AM: Question then is -- is it hard to do this later?
... If we say you can't shadow now, will it be a radical change to allow it downstream?

HST: Wouldn't invalidate any existing docs. . .

PROPOSAL: Take the strict route for now

AM: Yes, it would take a lot of rethinking of the doc. to change this. . .

HST: Leave this until we are quorate. . .

Another issue


HST: I'm happy for the editor to do as much or as little as he likes on this -- it's editorial

TV: OK with me

Test Suite

TV: Not much progress -- a few additional tests
... Some multipart tests are still proving challenging
... Waiting for NW to catch up
... Last Call ended on 2 Feb

HST: Cool
... we are nearly done, I think
... We need to focus very hard on whether we are ready to declare victory and go to PR
... It would be nice to have another implementation step up to the testing plate

TV: There is Jim Fuller's work with eXist/XQuery...don't know its status
... I saw signs of a small subset being implemented in PHP

AM: Two complete and some partial-but-positive would be good
... I will track down the state of the eXist work

HST: A report that passes what they've implemented and only fails where they haven't is indeed helpful
... Anonymous results are OK

TV: Standard step library is still broken

HST: My fault


Minutes formatted by David Booth's scribe.perl version 1.134 (CVS log)
$Date: 2010/02/04 17:00:26 $