Validation Protocol: Cross-Field, Async, and Conditional validation - #59

…tack-ai#7768) (objectstack-ai#7813)

* feat(spec): Field.number gains useGrouping presentation hint (objectstack-ai#7768)

FieldSchema gains an optional `useGrouping: boolean` (Option A, ruled
2026-08-11 on objectstack-ai#7768, maintainer veto window open) so an authored number
field can opt out of Intl.NumberFormat's digit grouping without losing
numeric semantics -- the fix for years (Field.number({ scale: 0, min: 1900 }))
rendering as "2,026" that downstream apps have worked around three times by
converting to Field.text (hotcrm-heimao#35/objectstack-ai#40/objectstack-ai#59).

No default is declared: absent defers to the renderer (interim heuristic
today, locale default eventually -- objectui#4033's contract, not this
package's). Threads through Field.number(...) automatically via the
existing FieldInput shape, same as scale/min.

Also: liveness ledger classifies the key `planned` (objectui#4033 is the
pending consumer); authorable-surface/data.json, field.mdx and
state-counts.md regenerated to match.

* chore(spec): regenerate field docs/authorable-surface/liveness after main merge

Wholesale regen (gen:docs, gen:schema's authorable-surface projection,
gen:liveness-counts) to re-materialize artifacts that drifted from commits
main picked up since this branch's last merge — the `internal` field key
(objectstack-ai#7728) and the `flows` translation surface's planned entries (objectstack-ai#7763).
check:generated: 13/13 green; check:liveness: green.

---------

Co-authored-by: Claude <noreply@anthropic.com>

…aggregate × field-type table (#16685)

Decision batch #80 (2026-09-08) holds ruling #11152 - booleans aggregate as
numbers on every backend, no per-aggregate exception - over batch #59's
blanket "every other pair refused", which never named booleans. The four
arithmetic / order rows gain the BOOLEAN_VALUE_TYPES members; the module
TSDoc and the table's pending changeset no longer publish the refusal; the
pins hold the boolean rows both literally and against AGGREGATION_CASES.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016N6xmWt5hYm94ffVEwGH8x

…AggregationFunction × FieldType) dataset measures are refused against (objectstack-ai#16353) (objectstack-ai#16684)

* feat(spec): declare the aggregate × field-type compatibility matrix (objectstack-ai#16353)

Export AGGREGATE_FIELD_TYPE_COMPATIBILITY and isAggregateCompatibleWithFieldType
from @objectstack/spec/data: the one table the dataset compiler and the lint
rule refuse dataset measures against. Rows follow the director ruling
(decision batch objectstack-ai#59), resolved against the full FieldType membership through
the field-value semantic classes; pinned literally in the test.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01F8SRGcf2eKTK7RRpWCGxwf

* chore(spec): pin the aggregate × field-type table in the api-surface and export-origins baselines (objectstack-ai#16353)

Regenerated by `check:generated --fix` after a full spec build: the two
stale shards (api-surface/data.json, export-origins/data.json) each gain the
two new exports and nothing else.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01F8SRGcf2eKTK7RRpWCGxwf

* fix(spec): fail-closed shape guard on isAggregateCompatibleWithFieldType; correct the published grounds for the boolean and time rows (objectstack-ai#16353)

Contract-review patch round. The predicate now refuses any non-string input
(a property-key lookup alone coerced ['count'] / { toString } to a member
spelling); pinned. The TSDoc and changeset no longer claim booleans are the
divergence class - objectstack-ai#11152 has every backend answer them as numbers - and
record that row, plus the min/max refusal over the string classes (objectstack-ai#15768
types them as a supported 'string' result), as overrides of existing
opinions referred to the maintainer. The time justification names SQLite's
canonical TEXT form (objectstack-ai#3994). No row changed.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01F8SRGcf2eKTK7RRpWCGxwf

---------

Co-authored-by: Claude <noreply@anthropic.com>

…aggregate × field-type table (objectstack-ai#16685) (objectstack-ai#16750)

* feat(spec): accept boolean / toggle for sum / avg / min / max in the aggregate × field-type table (objectstack-ai#16685)

Decision batch objectstack-ai#80 (2026-09-08) holds ruling objectstack-ai#11152 - booleans aggregate as
numbers on every backend, no per-aggregate exception - over batch objectstack-ai#59's
blanket "every other pair refused", which never named booleans. The four
arithmetic / order rows gain the BOOLEAN_VALUE_TYPES members; the module
TSDoc and the table's pending changeset no longer publish the refusal; the
pins hold the boolean rows both literally and against AGGREGATION_CASES.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016N6xmWt5hYm94ffVEwGH8x

* docs(spec): retract the d.ts byte-identity claim in the boolean-members changeset; derive the flag-case vocabulary pin from AggregationFunction (objectstack-ai#16685)

Contract-review patch round. The changeset claimed dist/*.d.ts was
byte-identical; it is not - the rewritten module TSDoc ships in
dist/data/index.d.ts. It now states what holds: the exported declarations
are unchanged, the private BOOLEAN_AGGREGATE_FIELD_TYPES constant is absent
from the bundle, and api-surface / export-origins are untouched. The test
header claims only what the cross-pin reaches (the boolean axis) and the
flag-case vocabulary is derived from AggregationFunction.options instead of
a literal six-member list.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016N6xmWt5hYm94ffVEwGH8x

* docs(spec): state the boolean-members constant's absence as measured — dist/*.d.ts and the bundles' export lists, not the bundle (objectstack-ai#16685)

Contract-review patch round 2. The constant does ship inside the bundles as
a non-exported binding, so "absent from the bundle" over-claimed; the
changeset now says only what was measured: absent from dist/*.d.ts and
from the bundles' export lists.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016N6xmWt5hYm94ffVEwGH8x

---------

Co-authored-by: Claude <noreply@anthropic.com>

…at it

Amendment by addition, in ADR-0058's own idiom: a new blockquoted block after
Amendment II.2 recording the maintainer ruling (decision batch #59, 2026-09-06)
that per-row `previous` on a predicate write may serve a row-invariant-in-effect
rewrite, with MULTI_UPDATE_HOOK_KEY_DIVERGENCE (#14099) as the engine mechanism
that makes it safe and the two shapes the rule does not admit.

The superseded 2026-08 D3 sentence is left standing as the dated record and
carries a forward pointer to the new block, so the AGENTS.md directive-13 grep
lands on the pointer at the line that would otherwise read as the live rule.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MkQhmuuJAVDjmeWNixwDDH

…view F1-F4)

Four prose corrections from the delta contract review's non-blocking findings.
No behaviour change, no contract change, no new argument.

F1 — ADR-0058 Amendment II.3's ruling paragraph cited the date and the decision
batch but not the recording comment. It now names comment `5560086928`, the
comment on this card that records the maintainer reply the block quotes.

F2 — round 3's rewrap left a stub line (`matched). So an`) mid-paragraph. The
paragraph is rewrapped to the block's own idiom; the prose is word-identical.

F3 — the block said it amends D3's closing SENTENCE. It amends the bullet's
last two: the "rewrite *conditioned* on the row is out of contract" sentence is
superseded for the in-place / same-key-set case alongside the "not so a rewrite
can be aimed" one. Now "closing sentences".

F4 — the changeset attributed the ruling to the director seat. The MAINTAINER
ruled; the director seat recorded it. This text ships to consumers inside the
package's CHANGELOG.md, so the misattribution was published. Now "Maintainer
ruling (recorded by the director seat, decision batch #59, 2026-09-06)".

Level re-derived rather than inherited: `packages/spec`'s files[] carries
src/**/*.zod.ts and dist, while docs/adr/** is in no package's files[]. This
round moves no published carrier and changes nothing behavioural, so the
existing `@objectstack/spec: minor` stands.

Claude-Session: https://claude.ai/code/session_01MkQhmuuJAVDjmeWNixwDDH
Co-authored-by: Claude <noreply@anthropic.com>