XML Query/XSL WG Joint Teleconference #627 Minutes 2015-12-15

> DRAFT XML Query/XSL WG Joint Teleconference #627 Agenda 2015-12-15
>
> Andy will chair this meeting.
>
> PLEASE SEND CORRECTIONS AND ADDITIONS AS SOON AS POSSIBLE SO A REVISED
> AGENDA CAN BE PREPARED IN A TIMELY MANNER!
>
> Changes in this agenda:
>
> This week, I'd like us to work on the following items:
> J4.2 (F&O)
> J4.4 (XPath)
> J4.6 (Test Suite)
> Q2.3
>
> Key:
> J = Joint Query/XSLT meeting;
> Q = Query-only meeting
>
> Teleconference connection details
> ---------------------------------
>
> (Member only)
> https://lists.w3.org/Archives/Member/w3c-xsl-query/2015Sep/0014.html
>
> SUMMARY OF DECISIONS
> ====================
> Note to secretary: Please insert a summary of all decisions made
> during this teleconference at this location in the minutes.
>
> Decisions MUST be clear as to their complete intent. The subject of
> the decision should *not* require dereferencing; it is most helpful
> if agenda numbers and bug numbers are accompanied by their titles.
> If a decision is recorded about the resolution of a bug, the
> decision must state exactly what the resolution is (e.g., "Accept
> the solution in comment #5", or "Replace the production for XXX with
> this: <new production>", or the like. It should be possible to
> read a decision months later and be certain what it meant and
> whether or not it was properly applied.
>
> NEW ACTION ITEMS
> ================
> Note to secretary: Please insert all new action items assigned during
> this teleconference at this location in the minutes.
>
> Action items MUST make sense even when they are viewed standalone
> (that is, not read in the context of the agenda item where they were
> assigned). This explicitly requires that all referents (e.g.,
> "this", "that") be made clear. Action items such as "Ralph to
> resolve this problem" are DISALLOWED. It takes a scribe only a few
> additional minutes to ensure that action items are clear, while it
> takes the entire WG considerable effort to constantly research the
> meaning of unclear action items. The subject of the action item
> should *not* require dereferencing; it is most helpful if agenda
> numbers and bug numbers are accompanied by their titles.

ACTION A-627-01: Michael Kay to publish a new F&O with the approved
revised definition of fn:transform

ACTION A-627-02: Michael Dyck to review bug #29277 and clarify what
the text means. ETA 2016-01.

> ===================================================
> Joint XML Query Working Group and XSL Working Group
> ===================================================
>
> J1. Administrative Items
>
> J1.1 Selection of secretary(s)
>
> John has been asked to take the minutes. Norm has been asked to stand
> by.
>
> NOTE: The chair will confirm the secretary prior to the start of the
> meeting.

Chair: Andy
Scribe: Norm
Present: Andy, Norm, Michael Sperberg-McQueen, Michael Kay, Josh, Jonathan,
Michael Dyck, O'Neil, Tim, Carine

> Most recent secretaries (least-recent participants first):
> ----------------------------------------------------------
> (This list identifies the most recent three occasions when each
> participant acted as Secretary for the Task Force. The person whose
> name is at the top of the list should plan to take the minutes at the
> meeting identified by this agenda; if that person is unavailable, then
> the job will fall to the next person on the list, etc.
>
> John Snelson - 2015-07-16, 2015-06-02, 2015-02-11
> Jonathan Robie - 2015-10-29, 2015-05-12, 2015-01-13
> Liam Quin - 2015-10-06, 2015-06-16, 2015-03-10
> Adam Retter - 2015-10-20, 2015-04-21, 2015-01-20
> O'Neil Delpratt - 2015-10-27, 2015-06-23, 2015-03-17
> Michael Sprbrg-McQn - 2015-10-03, 2015-06-09, 2015-03-03
> Mike Kay - 2015-11-10, 2015-07-15, 2015-04-14
> Michael Dyck - 2015-11-17, 2015-09-01, 2015-03-31
> Josh Spiegel - 2015-12-01, 2015-09-08, 2015-03-24
> Tim Mills - 2015-12-08, 2015-09-15, 2015-04-21
> Norm Walsh - 2015-12-15, 2015-09-22, 2015-05-05
>
> J1.2 Call the Roll -- NOT DONE DURING TELECONFERENCE
>
> Note to secretary: During the meeting, take a note of the Webex
> participants, then list in the minutes those present and those who
> sent regrets. All others will be presumed to have been absent
> without regrets.
>
> XML Query WG membership list:
> http://www.w3.org/2000/09/dbwg/details?group=18797&order=org
>
> XSL WG membership list:
> http://www.w3.org/2000/09/dbwg/details?group=19552&order=org
>
> Regrets:
>
> J1.3 Approval of agenda
>
> J1.3.1 Additions, deletions, or corrections?

None proposed.

Accepted.

> J1.4 Review of recent minutes and administrative items
>
> J1.4.1 Minutes of meeting #626 (Tim)
> https://lists.w3.org/Archives/Public/public-xsl-query/2015Dec/0011.html

Accepted.

> J1.5 Review of action items
>
> NOTE: Please be prepared to respond quickly on your action items!
>
> J1.5.1 Outstanding action items
>
> ACTION A-579-18: Jonathan to write change proposal to support update
> of
> maps and arrays. (See agenda item Q2.1.1)
>
> Status: Pending decision on future of XQUF31

Closed without action.

XQuery Update 3.1 will be removed from the agenda due to lack of
resources and interest.

> ACTION A-585-04: Jonathan to write a concrete proposal for XQuery
> Update
> 3.1 that makes preservation of node identity implementation defined at
> the persistence layer.
> See
> https://lists.w3.org/Archives/Member/w3c-xsl-query/2014Nov/0012.html
>
> Status: Pending decision on future of XQUF31

Closed without action.

XQuery Update 3.1 will be removed from the agenda due to lack of
resources and interest.

> ACTION A-622-02: Josh Spiegel to propose changes to XQueryX to support
> new grammar for empty expressions in braces and return to group with a
> proposal.
>
> Status: Pending; ETA 2015-12-15

Completed.

> ACTION A-623-01 Andy to do what's needed to republish the XQuery CR.
>
> Status: Pending; ETA not specified.

Completed.

> ACTION A-624-05 on Andy to request publication of Serialization 3.1
> as revised CR.
>
> Status: Pending; ETA not specified.

Completed.

> ACTION A-624-06 on Andy to request publication of XPath 3.1
> as revised CR.
>
> Status: Pending; ETA not specified.

Completed.

> ACTION A-625-06 on O’Neil to fix bug 29258 by splitting the tests into
> multiple versions.
>
> Status: Pending; ETA 2015-12-15

Completed.

> ACTION A-625-08 on Jonathan to propose a solution for bug 29246.
>
> Status: Pending; ETA 2015-12-15

Completed; needs discussion.

> ACTION A-626-01: Mike to fix F&O example to use correct XPath.
> Bug 29310 - [FO31] Invalid and ambiguous XPath expression in normative
> part of section 5.4.3 fn:substring.
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29310
>
> Status: Pending; ETA not specified.

Continued. New ETA: this week.

> ACTION A-626-02: JRobie to make a proposal by defining 1. external
> functions, 2. host language functions. Bug 29277 - [XP31]
> Evaluating function calls does not mention evaluation of dynamic or
> static function calls that have no FunctionBody
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29277
>
> Status: Pending; ETA not specified.

Completed.

> ACTION A-626-03: MKay to fix bug 27908 by modifying the test case to
> use
> an invalid URI. Bug 27908 - K2-SeqDocFunc-12
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=27908
>
> Status: Pending; ETA not specified.

Continued.

> ACTION A-626-04: JRobie to add 'switch' to the list of expressions in
> section 2.3.4 Errors and Optimization which must not raise a dynamic
> error in respect of subexpressions occurring in a branch that is not
> selected, and must not return the value delivered by a branch unless
> that branch is selected. ALSO add a note to clarify that in "if
> (condition) then 'z' else error()" z cannot be returned without
> evaluating the condition. Bug 29320 - [XQ31] Switch expressions
> should have same error semantics as typeswitch.
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29320
>
> Status: Pending; ETA not specified.

Completed.

> J1.5.2 Completed action items
>
> ACTION A-624-03 on O'Neil to fix test suite re bug 29249
> (Serialization-text-11).
>
> Status: Closed; See
> https://lists.w3.org/Archives/Public/public-xsl-query/2015Dec/0010.html
>
> ACTION A-624-04 on JRobie to evaluate comment 12 of Bug 29277
> ([XP31] Evaluating function calls does not mention evaluation of
> dynamic
> or static function calls that have no FunctionBody)
>
> Status: Closed; Discussed by WG and superceded by A-626-02
>
> ACTION A-625-01 on Andy to discuss the future of update 3.1 with Liam.
>
> Status: Closed; Decided to remove update 3.1 and full text 3.1 from
> agenda due to lack of interest and resources. Will reinstate at a
> later
> date if there is sufficient interest, i.e. at least two participants
> to develop the specs and create implementations.
>
> J1.6 Update on related activities
>
> J1.6.1 Update on XSLT WG activities (Sharon, MikeK)

Mike: Sending fn:transform bug back over the fence to XQuery. We have
a proposal for which we'd like approval from the joint working groups.

Andy: That's first on the agenda.

> J1.6.2 Update on EXPath/EXQuery CG activities (Mike)

No report.

> J1.7 Future Meetings
>
> J1.7.1 QT Meeting Schedule
>
> [V43] Joint XML Query/XSL WG meeting schedule
> https://lists.w3.org/Archives/Member/w3c-xsl-query/2015Jul/0038.html

Andy: I propose to cancel the meeting of 22 December. We resume in
the new year on 5 Jan.

> J1.7.2 Future F2F Meetings
>
> J1.7.2.1 Other future F2F meetings
>
> None scheduled.
>
> J2. Documents For Review
>
> (none)
>
> J3. Maintenance Issues
>
> J3.0 The XML Query WG and XSLT WG do not intend to maintains specs
> earlier than the 3.0 suite of documents; when the 3.1 suite reaches
> Recommendation, maintenance of the 3.0 documents will be terminated.
>
> This agenda item is reserved for reporting errors and other problems
> reported against the 3.0 (later, 3.1) suite of specifications.
>
> J3.0.1 Concerns raised by the XSLT working group in regards to
> potential
> completion delays of XSLT 3.0 if XPath 3.0 is made obsoleted by 3.1
>
> Status: Pending the outcome of Abel's review of 3.0 bugs/errata.
> Any change to the status of the 3.0 specs requires explicit XSL WG
> consultation.
>
> J3.1 Bugzilla Bugs filed against the 3.0 suite of joint XML Query WG
> and
> XSLT WG documents
>
> J3.1.1 Bugzilla Bug 24568, Is the type system really a lattice? Or
> just
> a partially ordered set?
> http://www.w3.org/Bugs/Public/show_bug.cgi?id=24568
>
> Status: Fixed in Data Model 3.1; not expected to be addressed in Data
> Model 3.0
>
> J3.1.2 Bugzilla Bug 25231, [F+O 3.0 ONLY], ISO 8601, and XSD 1.1
> http://www.w3.org/Bugs/Public/show_bug.cgi?id=25231
>
> Status: Fixed in F&O 3.1; not expected to be addressed in F&O 3.0
>
> J3.1.3 Bugzilla Bug 26788, [XPath30] XQuery-only errors in XPath 3.0
> specification, like XQST0134"
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=26788
>
> Status: Fixed in XPath 3.1; not expected to be addressed in XPath 3.0
>
> J4. Active Specification Development
>
> ** Note: Full Text 3.1 has been removed from the WG agenda.
> If, in the future, there is sufficient interest (i.e. at least two
> independent participants to develop the spec and implementations),
> then this can be brought back on the agenda.
>
> J4.1 Data Model 3.1 (Norm)
>
> J4.1.0 Internal working draft
> http://www.w3.org/XML/Group/qtspecs/specifications/xpath-datamodel-31/html/Overview.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xpath-datamodel-31/html/Overview-diff.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xpath-datamodel-31/src/changes.txt
>
> (none)
>
> J4.2 Functions and Operators 3.1 (Mike Kay)

Andy notes that the agenda has been renumbered following the removal
of XQuery Update 3.1.

> J4.2.0 Internal working draft
> http://www.w3.org/XML/Group/qtspecs/specifications/xpath-functions-31/html/Overview.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xpath-functions-31/html/Overview-diff.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xpath-functions-31/src/changes.txt
>
> J4.2.1 Bug 29146 - [FO31] fn:transform options update
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29146
>
> Status: Pending outcome of XSL WG's discussion on this.

Proposal is in editor's draft of the spec. Mike summarizes.

Substantive changes almost exclusively to the invocation of XSLT 3.0.

The XSLT WG requests approval to make this the status quo.

Andy: Any objection?

None heard.

The proposal is accepted.

ACTION A-627-01: Michael Kay to publish a new F&O with the approved
revised definition of fn:transform

Some discussion of how this plays out in publishing specs wrt to normative
references.

> J4.3 Serialization 3.1 (Michael Sperberg-McQueen, Andy)
>
> J4.3.0 Internal working draft
> http://www.w3.org/XML/Group/qtspecs/specifications/xslt-xquery-serialization-31/html/Overview.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xslt-xquery-serialization-31/html/Overview-diff.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xslt-xquery-serialization-31/src/changes.txt
>
> (none)
>
> J4.4 XPath 3.1 (Jonathan)
>
> J4.4.0 Internal working draft
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-31/html/xpath-31.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-31/html/xpath-31-diff.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-31/src/changes.txt
>
> J4.4.1 Bug 29277 - [XP31] Evaluating function calls does not mention
> evaluation of dynamic or static function calls that have no
> FunctionBody
>
> Status: Pending completion of A-626-02

Jonathan: I was prepared to make the changes, but it appears that they've
already been made.

Some discussion of the bug.

Do XSLT defined functions fall into category one or category two?

Some disagreement. Clarification: this is about functions defined by
an xsl:function element.

Jonathan: XPath doesn't know about XSLT-defined functions.

Jonathan and Michael discuss what's expected.

Jonathan: XSLT could say that these functions are invoked "as if" they were
FunctionBodies.

MKay: In XPath 2.0, this was defined at a slightly higher level of
abstraction where it worked. By being more specific about a couple of
cases, and saying nothing about other cases, we've lost something.

...But the 2.0 spec says nothing about the possibility of functions
being context dependent.

...I think we've simply got the wrong level of abstraction.

Jonathan: I'd like to avoid a major rewrite.

MikeK/MDyck conclude that there's some very fuzzy wording here.

Various proposals for always passing the context discussed.

Abel Braaksma joins the call for this topic.

More discussion of the meaning of "host language dependent function".

Jonathan: I think the XPath spec can remain unchanged. The XSLT spec
says that the context passed in is ignored.

... What I'm really trying to avoid is having the XPath spec say how
XSLT or any other language evaluates things.

CMSMcQ (in IRC): I think we can solve this either (a) by XSLT saying
“the context is passed, but must be ignored” or (b) by XPath saying
“the context is in F iff the host language says it’s in F” I believe
JR is suggesting that (a) would be nicer.

General agreement that this sentence is difficult to interpret:

  F's implementation is invoked with the converted argument values
  using the contexts it is associated with in F. If these contexts are
  absent in F, it is associated with SC and DC.

Jonathan (in IRC): let me try two rewordings and see if i am getting
this right

  "F's implementation is invoked with the converted argument values
  using the contexts it is associated with in F => F's implementation
  is invoked using the static and dynamic context of F itself."

and

  "If these contexts are absent in F, it is associated with SC and DC
  => If F does not have a static and dynamic context of its own, the
  SC and DC of the function call are used."

Abel: I think this just moves the scary bits ahead with different words.

Some discussion of how closures are formed.

Michael Kay (in IRC):  F's implementation is invoked, supplying:
  (a) the converted argument values
  (b) depending on the specification controlling the function in
  question, the focus for evaluation of F's implementation is either
  (I) SC and DC (the static and dynamic context of the function call),
  or (II) absent.
  The result of evaluating F's implementation is either a value or an
  error. If the result is a value then it is converted to the declared
  return type of the function by applying the function conversion
  rules. The result of evaluating the function call is this converted
  value.

Michael Dyck offers to investigate and report back.

ACTION A-627-02: Michael Dyck to review bug #29277 and clarify what
the text means. ETA 2016-01.

> J4.5 Grammar Test Applets (Michael Dyck)
>
> J4.5.1 Bugzilla Bug 13796, [Applet30] \u escaped characters in grammar
> applets
> http://www.w3.org/Bugs/Public/show_bug.cgi?id=13796
>
> Status: Pending
>
> J4.6 QT3 test suite (O'Neil)
>
> J4.6.1 Status report on test suite, harness, tools
>
> J4.6.2 Status of 3.1 tests
>
> Status: Coverage is almost complete.
>
> J4.6.3 Publication of test suite.
>
> Status: Ready to discuss
>
> J5. Progression of Work
>
> J5.1 XSLT 3.0, XQuery 3.1, XQueryX 3.1, XPath 3.1, XDM 3.1, F&O 3.1,
> Serialization 3.1
>
> J5.1.1 Current status
>
> * In Candidate Recommendation; awaiting implementation reports
>
> J5.1.2 Dependencies inhibiting progression
>
> * Completion of test suite minimum-required-to-declare-victory
>
> * Sufficient implementation experience and reports thereof
>
> * Formal response to all comments, responses accepted by commenters
>
> * PR Transition Request and authorization, PR pub request
>
> * REC Transition Request and authorization, REC pub request
>
> J5.2 Update 3.0
>
> J5.2.1 Current Status
>
> * In Last Call Working Draft
>
> J5.2.2 Require owner of XQUF 3.0 Test Suite
>
> J5.2.3 Dependencies inhibiting progression
>
> * CR Transition Request and authorization, CR pub request
>
> * Completion of test suite minimum-required-to-declare-victory
>
> * Sufficient implementation experience and reports thereof
>
> * Formal response to all comments, responses accepted by commenters
>
> * PR Transition Request and authorization, PR pub request
>
> * REC Transition Request and authorization, REC pub request
>
> J6. Any other business
>
> J6.1 Consideration of future work
>
> J6.1.1 QT 3.2
>
> The WGs decided that we will finish the 3.1 documents presently under
> way,
> but undertake no new work. Maintenance work should be planned upon for
> a
> period of time after finishing the specifications.
>
> J6.1.2 FT 3.1
>
> Currently inadequate interest and resources
>
> J6.1.3 Update 3.1
>
> Currently inadequate interest and resources
>
> J7. Adjourn
>
> =======================
> XML Query Working Group
> =======================
>
> Q1. The XML Query WG does not intend to maintains specs earlier than
> the
> 3.0 suite of documents; when the 3.1 suite reaches Recommendation,
> maintenance of the 3.0 documents will be terminated.
>
> Q1.0 This agenda item is reserved for reporting errors and other
> problems reported against the 3.0 (later, 3.1) XML Query WG suite of
> specifications.
>
> Status: Pending completion of ACTION A-592-02.
>
> Q1.1 Bugzilla Bugs filed against the 3.0 suite of XML Query WG
> documents
>
> (none)
>
> Q2. Active Specification Development
>
> Q2.1 XQuery Update Facility 3.0 (John)
>
> Q2.1.0 Internal working draft
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-update-30/html/Overview.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-update-30/html/Overview-diff.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-update-30/src/changes.txt
>
> * The WG needs to obtain implementation experience with this document.
>
> Q2.2 XQuery Update Facility 3.0 Requirements and Use Cases (Andy)
> See
> http://www.w3.org/XML/Group/qtspecs/requirements/xquery-update-30/html/Overview.html
>
> * The WGs need to be certain that this document is completed
> (red/yellow/green bullets on requirements, all the use cases intended
> to
> be published) and a decision made to advance it to WG Note.
>
> Q2.3 XQuery 3.1 (Jonathan)
>
> Q2.3.0 Internal working draft
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-31/html/xquery-31.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-31/html/xquery-31-diff.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xquery-31/src/changes.txt
>
> Q2.3.1 Bug 29246 - [XQ31] context item declarations
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29246
>
> Status: Pending completion of A-625-08

Jonathan: I attempted to resolve this in comment #7; I wonder if
Michael Dyck agrees with my proposal.

Some discussion of the proposal and what Michael Dyck intended.

Jonathan makes some editorial changes on the fly. Michael Dyck
expresses reluctant consent.

Jonathan agrees to continue editorial improvements if he sees fit.

Agreement to close bug.

> Q2.4 XQueryX 3.1 (Josh)
>
> Q2.4.0 Internal working draft
> http://www.w3.org/XML/Group/qtspecs/specifications/xqueryx-31/html/Overview.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xqueryx-31/html/Overview-diff.html
> http://www.w3.org/XML/Group/qtspecs/specifications/xqueryx-31/src/changes.txt
>
> (none)
>
> Q3. Test Suite Development
>
> Q3.1 Update Facility 3.0 Test Suite (????)
>
> * Need test suite developer/manager
>
> * Need tests
>
> Q4. Any other business

None heard.

> Q5. Adjourn

                                        Be seeing you,
                                          norm

--
Norman Walsh
Lead Engineer
MarkLogic Corporation
Phone: +1 512 761 6676
www.marklogic.com