QT4 CG Meeting 121 Minutes 2025-05-13

The following PRs are open but have merge conflicts or comments which suggest they aren’t ready for action.

• PR #1942: 37 Support sequence, array, and map destructuring declarations
• PR #1883: 882 Replace fn:chain by fn:compose
• PR #1283: 77b Update expressions
• PR #1062: 150bis revised proposal for fn:ranks

The following PRs are editorial, small, or otherwise appeared to be uncontroversial when the agenda was prepared. The chairs propose that these can be merged without discussion. If you think discussion is necessary, please say so.

• PR #1988: 1986 Drop obsolete notes on error reporting

Proposal: merge without discussion.

Accepted.

The following substantive PRs were open when this agenda was prepared.

• PR #1991: 835 Add built-in named record types to static context
• PR #1990: Update schema-for-xslt40.xsd
• PR #1989: 1983 QName literals in node constructors
• PR #1987: 1985 Tidy up namespace terminology
• PR #1984: 882 Drop fn:chain
• PR #1982: 1981 Ambiguity with qname literals and pragmas
• PR #1977: 1889 Tidy up handling of HTML serialization version, default to HTML5
• PR #1959: 1953 (part) XSLT Worked example using methods to implement atomic sets
• PR #1894: Additional examples to fn:chain - in a new branch
• PR #1888: 366 xsl:package-location

See PR #1989

Appropriate for XQuery and with a backwards incompatibility. Removes the option of using a string literal, replacing with QName literal. Uses a forward reference to PR1987 on namespace defaults.

• MK: What is unresolved is what we do for PI or namespace constructors
  • which currently are a string literal.
  • We could allow a Qname but error if it is in a namespace.

ACTION: QT4CG-021-01: MK to raise a new issue which may propose a subset of the QNameLiteral which doesn't permit a namespace.

• JLO: I like the proposal - avoiding the unreserved name string. Undecided on the processing
• WP: I think it's about the error message that come

Proposal: accept this PR.

Accepted.

See PR #1982

• MK: An incompatibility which now requires whitespace between # and pragma, but forbids it in a QNameLiteral.
• JWL: Is pragma the least disruptive?
• MK: Yes and most people using pragmas probably have whitespace already.

Proposal: accept this PR.

Accepted.

See PR #1977

• MK: Clarifying the rules and moving the default to HTML5 and de-emphasising earlier conventions.
  • We no longer give (global) default values for (most of?) the serialisation parameters.
  • XML version default is 1.0, HTML is 5
• CG: Questions on the PR may have already been answered. We need to look at the defaults for fn:serialize()

ACTION: QT4CG-121-02: MK to fix removal of default values in the options for fn:serialize()

Proposal: accept the PR with the above action.

Accepted.

See PR #1987

• MK: Now with a set of defined term named rules, (E.g. default type namespace rule…) which are then referenced for use cases.

Proposal: accept this PR

Accepted.

See PR #1990

• MK: Detailed work by Priscilla Walmsley.

MK has checked the detail, with only a few minor errors.

• JK: Is there a test suite?
• MK: yes - there are a set of tests of various forms.
• JK: can we add new tests to check for the new material?
• MK: the schema for XSLT is not normative.

Some discussion on the relation between spec and schema

Proposal: accept this PR

Accepted with thanks to Priscilla!

See PR #1888

• JK: PR mostly for comments and probably not yes for acceptance.
  • … As an XSLT programmer, xsl:use-package is not like xsl:import etc, and it doesn't give me control of the resources from the program. This is left to the implementation/execution.
  • … To fix this we add a xsl:package-location element which describes the potential paths and types of resources.
• JWL: Do we need unique priorities rather than order?

Some discussion including from JLO.

• MK: Archive-path might be confused as path within the archive. Also where do URI formats come from.
• JK: is an href enough to negotiate with package services?