QT4 CG Meeting 076 Minutes 2024-05-07
Table of Contents
1. Administrivia
Approved at meeting 077 on 14 May 2024.
1.1. Roll call
Regrets from NW, JLY. MSM to chair. [7/12]
[ ]Reece Dunn (RD). Regrets.[ ]Sasha Firsov (SF)[X]Christian Grün (CG)[X]Joel Kalvesmaki (JK)[X]Michael Kay (MK)[ ]Juri Leino (JLO)[ ]John Lumley (JLY). Regrets.[X]Dimitre Novatchev (DN)[X]Wendell Piez (WP)[X]Ed Porter (EP)[X]C. M. Sperberg-McQueen (MSM). Scribe. Chair.[ ]Norm Tovey-Walsh (NW). Regrets.
1.2. Approve the agenda
Proposal: Accept the agenda as posted.
MK says that PR #1181 belongs in the general list, not XSLT-focused, as it affects all specs.
CG: can we prioritize 1068?
Accepted as modified.
1.3. Approve minutes of previous meeting
MSM urged the group to look carefully at the status visualizations at the beginning of each week's minutes. The number of open bugs is improving, but (a) the overall trend line is still ascending and (b) the number of issues is still well into triple digits.
Proposal: Accept minutes of the previous meeting.
Accepted.
1.4. Next meeting
The next meeting is scheduled for 14 May. The following meeting is scheduled for 21 May.
MSM noted that he and NW have a conflicting meeting, but having just harangued the group on the subject of improving the burn-down rate, he felt incapable of proposing that we skip the meeting. So: we will meet on 14 May, and NW and MSM will have to sneak out of their other meeting.
1.5. Review of open action items [1/7]
(Items marked [X] are believed to have been closed via email before this agenda was posted.)
[ ]QT4CG-063-06: MK to consider refactoring the declare item type syntax to something like declare record[ ]QT4CG-071-06: NW to clarify the cases that are distinguished by the leading empty string in path segments[ ]QT4CG-072-03: NW to clarify the round-tripping of URIs[ ]QT4CG-073-01: NW to proceed with the records/options proposal and make a PR.[ ]QT4CG-075-01: MK to drop the deterministic option and raise it as a separate issue.[ ]QT4CG-075-02: MK to define sequence-concatenation more formally with links where appropriate[X]QT4CG-075-03: CG to make changes to map:values analagous to MK’s changes to array:values
MK reports he's done half of 075-01.
CG has made a PR satisfying 075-03.
1.6. Review of open pull requests and issues
1.6.1. XSLT focused
The following PRs appear to be candidates for a future XSLT-focused meeting.
- PR #1015: 1013 [XSLT] Clarify
effect of accumulator capture on non-element nodes
- PR #921: 920 Allow xsl:break and
xsl:next-iteration within branch of xsl:switch
- PR #871: Action qt4 cg 027 01 next match
These issues identify the XSLT-focused changes that have been made to the specifications but which have not been established by the community group as the status quo.
- Issue #168: XSLT Extension
Instructions invoking Named Templates
We removed PR #1181 from this list and moved it to the next section.
1.6.2. Substantive PRs
The following substantive PRs were open when this agenda was prepared.
- PR #1186: 1170 Editorial: fn:index-where
- PR #1185: 1179 array:values, map:values → contents
- PR #1184: 1165-Use Unicode-style character references
- PR #1182: 1171 Change predicate
callbacks to allow empty return value
- PR #1148: 1143 Coercion rules:
handle choice types before atomization
less aggressive %-encoding for fn:build-uri
- PR #1098: 566-partial Editorial improvements for parse-uri
- PR #1068: 73 fn:graphemes
- PR #1062: 150bis - revised proposal for fn:ranks
- PR #1027: 150 fn:ranks
We added:
- PR #1181: 296 Allow default-namespace=##any
2. Technical Agenda
2.2. PR #1185: 1179 array:values, map:values → contents
See PR #1185.
MK: the problem is that 'contains' looks at keys, and 'contents' looks at key-values.
MSM: 'value' is overloaded; that's part of theproblem, right?
CG maybe item?
DN: sequence concatenation in undefined order? Yow!
These functions lose the structure! This isn't a great way to get at the parts of an array or map.
MK: Yes. But so many arrays have singleton values that the requirement to move from array to sequence is a very common one.
JK: what's the intended difference between array:content() and array:flatten()? CG: contents() is much simpler.
CG: the motivation for these is to allow functional equivalents for a widely used wildcard syntax.
Various people: would 'flatten' be a better name?
CN: or flattened-value()?
CN: loss of information is the issue. array:squashed(), maps:squashed().
Further consideration needed. Take it offline.
2.3. PR #1184: 1165-Use Unicode-style character references
3. Any other business
None.