Summary of technical discussion/decisions XSL WG telcon 2016-08-25

The XSL WG met on Thursday, 25 August 2016 for 60 minutes at the START TIME, a summary of the minutes and decisions of that telcon follows:

> 7. XSLT Draft
> 
> The diff version is at the usual location.
> 
>  Published CR: http://www.w3.org/TR/2015/CR-xslt-30-20151119/
> 
> Internal version:
> https://www.w3.org/XML/Group/qtspecs/specifications/xslt-
> 30/html/Overview.html
> W3C Candidate Recommendation 30 June 2016
> 
> 8. Spec bugs
> 
> 29472  [XSLT30] Optionally allow xsl:stream to process a document 
> *without* streaming  2016-03-10
>  https://www.w3.org/Bugs/Public/show_bug.cgi?id=29472
>      Status: closed by Abel; reopened by Sharon to have WG approve 
> necessary actions.  Just a check to see if everyone is happy with 
> current status.

Chair asks whether there are objections for closing this bug.
DECISION: bug can be closed as INVALID

ABr closed the bug (during the meeting)

> 
> 29692 xsl:strip-space and packages
> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29692
> <https://www.w3.org/Bugs/Public/show_bug.cgi?id=29692>
>    Status: Changes have been put into spec; bug marked resolved - bug 
> should not be closed until text is reviewed in next draft.

Not discussed during this meeting.

> 
> 29732 [xslt30] Streamable stylesheet functions: shallow-descent example
>    https://www.w3.org/Bugs/Public/show_bug.cgi?id=29732
>          See MK action item.

Linked to action item ACTION-2016-07-21-002

The WG will wait until MKay comes with a proposal.

Note (requested by MKay): consider the scenario where declaring a streaming variable, but not using it, should not be illegal.

> 
> 29733 [xslt 3.0] Why you can't pass streamed nodes to the second 
> argument of a streamable stylesheet function
>    https://www.w3.org/Bugs/Public/show_bug.cgi?id=29733
>          See MK action item

Linked to ACTION-2016-07-21-003
Action item was done just before the meeting, to be discussed at next telcon

> 
> 29747 [XSLT30] Allow the attribute xsl:stream/@streamable with a default of
> "yes"     19:58:45
>     https://www.w3.org/Bugs/Public/show_bug.cgi?id=29747
>     Check - review that all are OK with current status.

MSMcQ argues (summarized in IRC):
"the proposal in the original description of 29747 is fine.  The supporting argument by MK is of more concern, as it once again conflates the declarative and imperative interpretations." 

He expresses his concern about imperative vs declarative, also in other parts of the spec with regards to streamability.

MK explains that we are deliberately fuzzy about what exactly streaming constitutes, or means for a processor.
He further explains that setting streamable="yes" is more of a hint than a request.

SA explains that there will always an interoperability issue, where processors can stream more than we can specify. Which is why we allow processors to do more than the streamability rules allow.

CONCLUSION: There is consensus on the principle that is described in the description of the bug. After some discussion, there was consensus on the proposal of comment#1 as well.

DECISION: the WG agreed to replace xsl:stream with xsl:source-document, as mentioned in the penultimate paragraph in bug 29747, comment#1 with reverted @streamable (defaulting to no, as other @streamable attributes)

ACTION-2016-08-25-002: MKay to implement changing xsl:stream into xsl:source-document, on the principle of comment#0 and as further described in comment #1 of Bug 29747

> 
> 29752  [XSLT30]two accumulator examples using count(tokenize(., 
> '\s+')) respectively count(tokenize(., '\W+')) to count words give odd 
> results
> 2016-07-25
>      New - never discussed

To be discussed next week

> 
> 29763  [XSLT30] overriding xsl:param with xsl:variabe or vice versa 
> should be disallowed
>      New - never discussed

The WG discussed this bug and agreed to the proposal outlined in comment#3

DECISION: proposal on fixing the default visibility for xsl:param when using a package in section 

ACTION-2016-08-25-001: MKay to implement comment#3 of bug 29763 into the spec on fixing default visibility of xsl:param for used packages

> 
> 29768  [xslt3.0] Streamability of map functions
>      New - never discussed

DECISION: The proposal in the description of bug 29768  was accepted by the WG

ACTION-2016-08-25-003: MKay to fix streamability rules for map functions as described in comment#0 of bug 29768

> 
> 29790 [xslt30] Sample stylesheet for xml-to-json conversion uses a 
> reserved namespace
> 
> 9. Test bugs
> 
> From bugzilla awaiting bug ABr desires to be discussed.

ABr mentioned that no additional discussion is needed for the time being, and that several WG members are working on tests

Received on Thursday, 1 September 2016 14:30:42 UTC