Minutes of XSLT WG Telcon 31 March 2016

Please find the minutes of XSLWG telcon 31 March 2016 below.

NOTE: In the discussion about Bug 29431 - [XSLT30] item-separator applicability, I am not sure I captured MSMcQ's comments entirely, see "NOTE FROM SCRIBE" below.

Summary of new action items:
ACTION-2016-03-31-001: new call to be scheduled on testing for April 1st, 2016, 18:00PM CET
ACTION-2016-03-31-002: MKay to make his test driver available to test writers
ACTION-2016-03-31-003: ABr to allow running a full test from the commandline using Exselt
ACTION-2016-03-31-004: MKay to add some examples to bug # 29431 on item-separator applicability, in particular if the result is a sequence of nodes/elements or a sequence of atomic items and/or mixed result

There were issues with connecting through the WebEx interface, new instructions are:

1) Using the direct link WebEx, the website asks for a code (this is new), use: 648 089 xxx (this is the meeting number, it will be different each time)
2) Then it will ask for a password and your email and name. Password will be in the Agenda (member-only list)
3) After giving these details, WebEx will start the interface.

Cheers,
Abel

> From: Michael Kay [mailto:mike@saxonica.com]
> Sent: Thursday, March 31, 2016 12:14 PM
> To: XSL Working Group
> Subject: AGENDA XSLT WG Telcon 31 March 2016
> 
> Sharon reminded me that I agreed to chair this week: sorry for the late
> agenda, I have been enjoying the sunshine in Devon.
> 
> The clocks are back in sync.
> 
> The XSL WG will meet on Thursday, 31 March 2016 at the START TIME as
> follows:
> 
>   09:00 a.m. Pacific Standard US/Canada,
>   10:00 a.m. Mountain Standard US/Canada,
>   11:00 a.m. Central Standard US/Canada,
>   12:00 Eastern Standard US/Canada,
>   17:00 UK, and
>   18:00 for France, Germany, Sweden and Belgium
> 
> Joining WebEx calls:
> 
> 
> AGENDA:
> 
> Michael will chair this week.
> 
> 1 Roll call - Need manual roll call

Roll call: Anders, MSMcQ, Charles, Luis, Michael Kay (chairing), Sharon (listening), Abel

> 
> 2 Assign minute-taker - volunteers?

Abel volunteers ;)

> 
> 3 Approval of previous minutes (10 March):
> 
> https://lists.w3.org/Archives/Member/w3c-xsl-wg/2016Mar/0005.html

Minutes approved with no comments.

> 
> 4. Review of actions
> 
> ACTION-2015-11-19-002 MHK to ensure that the build system creates up-to-
> date namespace documents  for the XSLT and MAP namespaces.
>    status? Still pending?

Still pending

> ACTION 2016-03-03-001: Charles Foster to develop an RNG schema for XSLT
> 3.0 (either from scratch or on the basis of Mohammed Zergaoui's old RNG
> schema for XSLT 2.0).

Still pending

> 
> 
> ACTION-2016-03-10-001: Carine to work with the comms team to see our
> options for what https://www.w3.org/TR/xslt should result in.

DONE
See https://lists.w3.org/Archives/Public/public-xsl-wg/2016Mar/0011.html

To be discussed next week when Carine is on the call.

> 
> ACTION-2016-03-10-002: Abel to send mail to his favourite lists announcing
> new versions of the test development and submission info.

Still pending


> 
> ACTION-2016-03-10-003: Abel to set up a conference call to to discuss test
> writing assignments and mechanics. Tentatively usual telcon time Friday 11
> March.

Done

ACTION-2016-03-31-001: new call to be scheduled on testing for April 1st, 2016, 18:00PM CET

New call to be done Friday, April 1st, 2016, 18:00PM GMT+2 (CET)

New action emerging from that last call:

ACTION-2016-03-31-002: MKay to make his test driver available to test writers
ACTION-2016-03-31-003: ABr to allow running a full test from the commandline using Exselt

> 
> ACTION-2016-03-10-004: Mike Kay to develop his proposal on executing a
> transformation into actual spec text for review.

Still pending

> 
> ACTION-2016-03-10-005: Mike Kay to introduce the term "vacuous" for
> empty in xsl:on-empty.

Action still pending
Bug number: 29455

> 
> 5 Other administrative business
> 
> 5.1 Review of agenda - items to be added to Agenda?

Nothing to be added

> 
> 5.2 Telcon schedule -

Back to normal schedule:

April 7th, 14th, 21st, 28th telcon as planned at normal time (90 mins?)

> 
> 5.3 Report on joint work

Nothing to mention

> 6. Spec bugs
> From bugzilla (12 open bugs)

Bug #29431
https://www.w3.org/Bugs/Public/show_bug.cgi?id=29431
[XSLT30] item-separator applicability

We decided to go over the resolution in comment#2 once again. Michael Kay walked us through it.

Abel asks:

Typically, is this what we are saying? 
Raw Result (without seps) 
-> Adding separators (build-tree)
-> Serialization (processing xsl:character-maps etc) 
and the API and/or xsl:output can be used to decide which of the three stages you want applied.

MKay agrees

MSMcQ asks whether we can say more about the raw result being delivered by the API or otherwise and stipulates that in almost most cases he uses an XSLT processor only from the commandline.

MKay suggests that getting the Raw Result is out of scope for the spec (it's an API matter). In the commandline scenario, there is no way to get the Raw Result when you are serializing anyway, in which case it matters little.

NOTE FROM SCRIBE: I am not sure I captured MSMcQ's remarks / comments completely and/or correctly here.

ACTION-2016-03-31-004: MKay to add some examples to bug # 29431 on item-separator applicability, in particular if the result is a sequence of nodes/elements or a sequence of atomic items

DECISION: comment#2 accepted as resolution, with comments in comment#3, and the action (ACTION-2016-03-31-003) from these minutes.


Bug #29455
https://www.w3.org/Bugs/Public/show_bug.cgi?id=29455
[XSLT30] Clarify/update rules of xsl:on-empty and xsl:on-non-empty with respect to sequences of empty strings or arrays

We briefly went over comments from earlier discussions to recollect our train of thought on this.

DECISION: this bug is accepted with resolution comment#5 and "vacuous" as the chosen term for *irrelevant items*

Bug #29461
https://www.w3.org/Bugs/Public/show_bug.cgi?id=29461
[xslt30] Clarify effect of document("") when the source stylesheet is not available at run-time

ABr asks what happens if you do not have a location of the source of the stylesheet (i.e., if it comes from a database).
MSMcQ answers that if you load it from a stream, than there won't be location information available, but you should (generically speaking) take the base uri of the processor, if one exists

ABr asks whether there *must* be a static base uri present, MKay answers that this is allowed, in which case function calls dependent on this will raise an error

MSMcQ asks in what situations there will be no meaningful URI at all, ABr answers that that may happen in SaaS scenarios or in dynamically compiled assemblies (in .NET for instance).

ABr asks whether we can add a line that explains that the "static base URI" can be absent.

MSMcQ: asks to make this a *Note* which points to the relevant section: https://www.w3.org/TR/xpath-30/#static_context (or the XPath 3.1 section which has "absent" defined). There is a generic error in XP30: https://www.w3.org/TR/xpath-30/#ERRXPST0001

MKay remarks that in FO30 and FO31 the notion of static base uri being absent is in: https://www.w3.org/TR/xpath-functions-30/#func-static-base-uri

ABr summarizes that we need to cater for these scenarios as least:"
1) source document is not available, but the URI is
2) source URI is not available, either deliberately (though API) or by limitation of environment

MKay summarizes this as "Two questions", with all four answers allowed (though not all "recommended"), that the text should cover:
1) Is the source stylesheet available? Answer can be yes or no
2) Is the base uri available? Answer can be yes or no

DECISION: The WG accepts the general idea explained in comment#3 with taking into account the minutes of today's telcon


Bug #29544
https://www.w3.org/Bugs/Public/show_bug.cgi?id=29544
[xslt30] xsl:strip-space and parse-xml

These functions are from FO30 and are not listed under the items for xsl:strip-space, but xsl:strip-space does not mention this.

DECISION: the WG accepted the proposal in comment#0


Bug #29513
https://www.w3.org/Bugs/Public/show_bug.cgi?id=29513
[XSL30] Definition of "potentially consuming" in Glossary not complete

Marked EDITORIAL

Received on Friday, 1 April 2016 12:18:50 UTC