Skip to content

chore: release (alpha) - #459

Open
supabase-releaser[bot] wants to merge 1 commit into
mainfrom
changeset-release/main
Open

chore: release (alpha)#459
supabase-releaser[bot] wants to merge 1 commit into
mainfrom
changeset-release/main

Conversation

@supabase-releaser

@supabase-releaser supabase-releaser Bot commented Sep 7, 2026

Copy link
Copy Markdown
Contributor

This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will be updated.

⚠️⚠️⚠️⚠️⚠️⚠️

main is currently in pre mode so this branch has prereleases rather than normal releases. If you want to exit prereleases, run changeset pre exit on main.

⚠️⚠️⚠️⚠️⚠️⚠️

Releases

@supabase/pg-delta@1.0.0-alpha.50

Minor Changes

  • cb9a90b: feat(pg-delta): apply accepts a pre-extracted source fact base for the fingerprint gate

    apply() re-extracted the target on every gated call so the fingerprint could
    be compared to the plan's source. Callers that already extracted the target they
    planned from — and still hold exclusive write access — can now pass that raw
    FactBase as sourceFactBase. The gate reconstructs the same managed view and
    skips the extract; a mismatched base still fails. Default behaviour (re-extract)
    is unchanged when the option is absent. planSchemaFiles returns the planning
    extract as targetFactBase so a library caller that still holds exclusive write
    access can pass it; schema apply keeps the live re-extract because it does not
    hold that guarantee across shadow load.

    sourceFactBase is only valid when nothing else can write to the target between
    plan and apply; the caller is asserting that. fingerprintGate: false remains
    the escape hatch that drops the check entirely.

  • ce61f01: Empty-target baselines can exhaust PostgreSQL's lock table in one
    transaction. Call estimateLockTableBudget on the apply target, then
    optionally splitPlan({ maxLocks }) so each segment tries to stay under
    that many lock slots. apply honors the marks and does not refuse.

    CLI: --max-locks <n> (plan / apply / schema apply) and --split-to-fit
    (apply / schema apply; probes the target). Extra COMMITs are only safe
    when nothing else reads the target; off by default, so existing plans
    and the corpus are unchanged.

Patch Changes

  • 678e4c6: Skip default-privilege rows scoped to system or per-session temp schemas (ALTER DEFAULT PRIVILEGES … IN SCHEMA pg_temp_N) at extract time. Such a row keyed a defaultPrivilege fact on a schema that exists on no target and no plan produces, so planning against a database carrying one failed in buildActionGraph with missing requirement: … consumes schema:pg_temp_N.
  • b17d6a9: A PG16+ CREATEROLE non-superuser cannot replay GRANT <role> TO <self> WITH ADMIN OPTION (SQLSTATE 0LP01); CREATE ROLE already recreates that membership. Extract stays a catalog dump. resolveProfile now probes applier capability by default (omitted or true) and projects those self-ADMIN memberships — plus existing FDW ACLs — out of the managed view for plan, schema apply, diff, and schema export. Pass { restrictToApplier: false } / --no-restrict-to-applier (plan / schema apply) for an unrestricted view (plan-here / apply-as-more-privileged). prove reconstructs the plan artifact's capability and does not re-probe the clone. A superuser probe excludes nothing. Bare plan() stays unrestricted when capability is omitted.
  • 1c2d4cb: fix(pg-delta): the shadow loader's DML observation no longer probes tables in the active policy's assumed (platform) schemas
  • 8365dc2: fix(pg-delta): do not plan a duplicate CREATE for a platform event trigger hidden by owner on one side only (CLI-2341)
  • a982dfa: Keep checked-out PostgreSQL clients protected by an error listener until
    release.

@github-actions
github-actions Bot force-pushed the changeset-release/main branch 5 times, most recently from a5635be to ea5c41d Compare September 8, 2026 16:26
@avallete
avallete enabled auto-merge September 8, 2026 18:14
@github-actions
github-actions Bot force-pushed the changeset-release/main branch from ea5c41d to c3d9145 Compare September 9, 2026 05:22
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant