Remove Supabase
The database moved to a container of our own; the platform is gone. This takes out what was left of it — and, where the leftovers were load bearing, moves rather than deletes. Moved, not deleted: supabase/migrations/ -> db/migrations/ the schema's source of truth supabase/build-org.ts -> scripts/build-org.ts lib/supabase/types.ts -> lib/types.ts 52 import sites repointed The bookkeeping needed care. It lived in `supabase_migrations.schema_migrations`, and simply renaming the schema would have left the runner facing an empty table: it would have called all 67 migrations pending and replayed them against a database that is long since current. So the runner now creates `migrationen.schema_migrations` and, once, copies the old rows across — guarded so a second run does nothing and a fresh database skips it entirely. Only then does migration 20260907100000 drop the old schema. Deleted: the CLI config, the seed, the historical schema/function dumps (nothing read them), scripts/umzug-von-supabase.sh (the move is done), and both Supabase packages plus the CLI. Nothing in the application imported them — the build now succeeds with no environment variables at all, which is the proof. Integration tests: six of them signed in through Supabase Auth and asserted against the anon key and the service role. That model is gone, so the tests were not portable — they are deleted. session-context and employee-status-filter already ran on pg and are untouched; om-reporting is ported to a direct connection because it guards a real risk (the reporting line rule exists twice, once in SQL and once in TypeScript). CI: the integration job started a Supabase stack. It now runs a postgres service, applies deploy/db-init and every migration to an empty database — that was the valuable part, and it still holds — then checks that a second run is a no-op, which is what proves the bookkeeping works. Docs: security-review.md audited a service-role key, a cookie adapter and auth.users, none of which exist. Restating findings about removed components would suggest today's system had been reviewed; it has not. It now records what was removed and says a fresh review is due. data-model.md was already marked obsolete and described the pre-OM schema; azure-migration.md was a plan for a route not taken. Both deleted. Verified: npm ci, typecheck, lint, 445 tests, build — all clean without the packages. Integration tests skip cleanly with no database. Migration SQL and the runner are reviewed but NOT executed: no Docker here, and the old instance no longer resolves. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
150
db/migrations/20260727130000_om_cleanup_and_positions.sql
Normal file
150
db/migrations/20260727130000_om_cleanup_and_positions.sql
Normal file
@@ -0,0 +1,150 @@
|
||||
-- Reste des Altmodells entfernen und die Planstellenpflege im OM-Modell
|
||||
-- nachziehen.
|
||||
--
|
||||
-- Die Cut-over-Migration hat die Funktionen des Altmodells mit ihren damals
|
||||
-- bekannten Signaturen entfernt. Ein Teil davon existierte zusätzlich in
|
||||
-- einer jsonb-Variante und ist deshalb stehen geblieben — sichtbar daran,
|
||||
-- dass delete_position und undo_reorg weiterhin in der PostgREST-Schnittstelle
|
||||
-- auftauchen, obwohl die Tabellen, auf denen sie arbeiten, weg sind. Ein
|
||||
-- Aufruf würde erst zur Laufzeit scheitern.
|
||||
|
||||
-- ═══ 1. Übriggebliebene Funktionen des Altmodells ════════════════
|
||||
drop function if exists create_position(jsonb);
|
||||
drop function if exists delete_position(jsonb);
|
||||
drop function if exists delete_position(uuid);
|
||||
drop function if exists staff_position_internally(jsonb);
|
||||
drop function if exists apply_reorg(jsonb);
|
||||
drop function if exists undo_reorg(jsonb);
|
||||
drop function if exists undo_reorg(uuid);
|
||||
|
||||
-- ═══ 2. Reorganisations-Werkbank ═════════════════════════════════
|
||||
-- Sie hat Teams und Abteilungen zwischen Bereichen verschoben — Objekte, die
|
||||
-- es nicht mehr gibt. Im OM-Modell ist eine Reorganisation das Umhängen von
|
||||
-- org_units.parent_id und braucht kein eigenes Szenario-Modell mehr.
|
||||
alter table employee_history drop column if exists reorg_scenario_id;
|
||||
alter table pending_org_changes drop column if exists reorg_scenario_id;
|
||||
drop table if exists reorg_moves;
|
||||
drop table if exists reorg_scenarios;
|
||||
|
||||
-- ═══ 3. Planstellen pflegen ══════════════════════════════════════
|
||||
-- Die alte positions-Tabelle führte nur *offene* Stellen und war damit ein
|
||||
-- eigenes Objekt neben der Person. Im OM-Modell hat jede Person eine
|
||||
-- Planstelle, und eine offene Stelle ist schlicht eine unbesetzte. Anlegen
|
||||
-- und Schliessen sind deshalb Operationen auf om_positions.
|
||||
|
||||
-- set search_path bei jeder Funktion: siehe die folgende Migration, dort steht
|
||||
-- warum. Kurz: heute sind das INVOKER-Funktionen und der Pfad ist harmlos,
|
||||
-- aber sobald eine davon einmal SECURITY DEFINER wird, wäre er es nicht mehr —
|
||||
-- und daran denkt dann niemand.
|
||||
create or replace function next_position_number()
|
||||
returns text language sql stable
|
||||
set search_path = public, pg_temp
|
||||
as $$
|
||||
select '6' || lpad((coalesce(max(substring(position_number from 2)::bigint), 0) + 1)::text, 7, '0')
|
||||
from om_positions
|
||||
where position_number ~ '^6[0-9]{7}$';
|
||||
$$;
|
||||
|
||||
comment on function next_position_number() is
|
||||
'Nächste freie Planstellennummer im Nummernkreis 6xxxxxxx.';
|
||||
|
||||
create or replace function create_position(payload jsonb)
|
||||
returns uuid language plpgsql
|
||||
set search_path = public, pg_temp
|
||||
as $$
|
||||
declare
|
||||
v_org_unit_id uuid := (payload->>'org_unit_id')::uuid;
|
||||
v_job_title text := nullif(trim(payload->>'job_title'), '');
|
||||
v_is_chief boolean := coalesce((payload->>'is_chief')::boolean, false);
|
||||
v_valid_from date := coalesce(nullif(payload->>'valid_from','')::date, current_date);
|
||||
v_job_id uuid;
|
||||
v_position_id uuid;
|
||||
v_unit_name text;
|
||||
begin
|
||||
perform require_hr_admin();
|
||||
|
||||
select name into v_unit_name from org_units where id = v_org_unit_id;
|
||||
if v_unit_name is null then
|
||||
raise exception 'Die Organisationseinheit existiert nicht.';
|
||||
end if;
|
||||
if v_job_title is null then
|
||||
raise exception 'Es muss eine Tätigkeit angegeben werden.';
|
||||
end if;
|
||||
|
||||
-- Der Unique-Index würde das ebenfalls abfangen, aber mit einer Meldung,
|
||||
-- die in der Oberfläche nichts erklärt.
|
||||
if v_is_chief and exists (
|
||||
select 1 from om_positions
|
||||
where org_unit_id = v_org_unit_id and is_chief and valid_to is null
|
||||
) then
|
||||
raise exception 'Für % besteht bereits eine Leitungsplanstelle.', v_unit_name;
|
||||
end if;
|
||||
|
||||
-- Gleiche Tätigkeit, ein Katalogeintrag: sonst stehen "Schlosser:in" und
|
||||
-- "Schlosser" nebeneinander und jede Auswertung nach Tätigkeit ist wertlos.
|
||||
select id into v_job_id from jobs where lower(title) = lower(v_job_title);
|
||||
if v_job_id is null then
|
||||
insert into jobs (code, title)
|
||||
values ('J' || lpad((select count(*) + 1 from jobs)::text, 4, '0'), v_job_title)
|
||||
returning id into v_job_id;
|
||||
end if;
|
||||
|
||||
insert into om_positions (position_number, org_unit_id, job_id, is_chief, valid_from)
|
||||
values (next_position_number(), v_org_unit_id, v_job_id, v_is_chief, v_valid_from)
|
||||
returning id into v_position_id;
|
||||
|
||||
insert into audit_log (actor_user_id, actor_name, action, target_label, details)
|
||||
values (auth.uid(), current_actor_name(), 'Planstelle angelegt',
|
||||
v_job_title || ' (' || v_unit_name || ')',
|
||||
'Gültig ab ' || v_valid_from || case when v_is_chief then ', Leitung' else '' end);
|
||||
|
||||
return v_position_id;
|
||||
end;
|
||||
$$;
|
||||
|
||||
create or replace function delete_position(payload jsonb)
|
||||
returns void language plpgsql
|
||||
set search_path = public, pg_temp
|
||||
as $$
|
||||
declare
|
||||
v_position_id uuid := (payload->>'position_id')::uuid;
|
||||
v_label text;
|
||||
v_hat_historie boolean;
|
||||
begin
|
||||
perform require_hr_admin();
|
||||
|
||||
select j.title || ' (' || u.name || ')' into v_label
|
||||
from om_positions p
|
||||
join jobs j on j.id = p.job_id
|
||||
join org_units u on u.id = p.org_unit_id
|
||||
where p.id = v_position_id;
|
||||
if v_label is null then
|
||||
raise exception 'Die Planstelle existiert nicht.';
|
||||
end if;
|
||||
|
||||
if exists (select 1 from position_assignments where position_id = v_position_id and valid_to is null) then
|
||||
raise exception 'Die Planstelle ist besetzt und kann nicht entfernt werden.';
|
||||
end if;
|
||||
|
||||
select exists (select 1 from position_assignments where position_id = v_position_id)
|
||||
into v_hat_historie;
|
||||
|
||||
-- Eine Planstelle, auf der einmal jemand sass, wird geschlossen statt
|
||||
-- gelöscht: sonst verschwindet mit ihr die Besetzungshistorie, und in der
|
||||
-- Personalakte klafft eine Lücke.
|
||||
if v_hat_historie then
|
||||
update om_positions set valid_to = current_date where id = v_position_id;
|
||||
else
|
||||
delete from om_positions where id = v_position_id;
|
||||
end if;
|
||||
|
||||
insert into audit_log (actor_user_id, actor_name, action, target_label, details)
|
||||
values (auth.uid(), current_actor_name(),
|
||||
case when v_hat_historie then 'Planstelle geschlossen' else 'Planstelle gelöscht' end,
|
||||
v_label, null);
|
||||
end;
|
||||
$$;
|
||||
|
||||
grant execute on function next_position_number() to anon, authenticated, service_role;
|
||||
grant execute on function create_position(jsonb) to anon, authenticated, service_role;
|
||||
grant execute on function delete_position(jsonb) to anon, authenticated, service_role;
|
||||
Reference in New Issue
Block a user