rutap.v_export_estado_actual: vista DEV-only sin callers, no se porta a PROD y se borra de DEV; resuelve catalogos distinto a MOVILBA-469
rutap.v_export_estado_actual exists only in DEV (PROD to_regclass NULL, pending-prod-changes.md:746-774), is declared in drizzle/schema.ts:2961-3000 and has zero callers in src/ and tools/ (git grep vExportEstadoActualInRutap: declaration only). It is the only DB dependent of rutap.v_vehiculo_historial_actual in DEV. Two facts from the 471 blackboard (db02 + audit, 2026-09-17): (1) an unused object is not ported to PROD and by the no-coexistence rule is dropped from DEV once its deletion has its own evidence (pg_depend at drop time, schema.ts regenerated by introspect.sh, no caller). (2) It resolves catalogs by nombre_mostrable (o_piso, o_aire, o_tramite, o_emision) and tipo from vehiculos_subtipos_transporte.descripcion, contrary to what MOVILBA-469 shipped for the same book columns (valor_opcion for Aire_acondicionado and Tramite, nombre_mostrable for Piso and Tipo_Motor, subtipo codigo for Tipo) - if it were ever pointed at a real export it would disagree with the historico report on the same cells. Resolution: DROP VIEW rutap.v_export_estado_actual in DEV (no CASCADE, no dependents), regenerate schema.ts, remove the pending-prod-changes.md entry that scoped porting it. Not part of 471.
Activity
-
coder03assistant-mba-l-ca | 2026-09-24 04:11:42 ART | DEV movilba_dev and PROD movilba both currently have rutap.v_export_estado_actual, contrary to ticket's DEV-only premise. drizzle/schema.ts still declares it and src/tools search found no caller. Drop/no-coexistence task is not complete; remeasure dependencies and PROD policy before any DDL.
-
coder04-mba-l-ca
-
coder03assistant-mba-l-ca | 2026-10-02 14:17 ART. Fresh read-only census contradicts DEV-only premise: rutap.v_export_estado_actual resolves in PROD movilba and DEV movilba_dev; SELECT count(*) = 2419 PROD, 2455 DEV. Existing MOVILBA-473 already tracks this object, so MOVILBA-667 opens no duplicate child. This census does not establish non-repo/external readers or authorize drop/port.
-
Fresh SELECT-only cross-base check (2026-10-09): rutap.v_export_estado_actual exists in BOTH PROD movilba (2,424 rows) and DEV movilba_dev (2,469 rows); it is not DEV-only. pg_depend reports zero dependent views in either base. Source/tools rg finds no references (including the Drizzle declaration's generated symbol). docs/migrations/pending-prod-changes.md:765-774 still says PROD to_regclass is NULL based on 2026-09-07, so that premise is stale. No DDL or file changes. Before any drop, re-scope whether the now-present PROD/DEV object may be removed; repository scans cannot establish out-of-repo consumers.
-
The ticket's DEV-only premise is stale: fresh PROD and DEV SELECTs show the view present in both, with 2,424/2,469 rows and zero dependent views. Source/tools has no callers, but that cannot rule out external consumers. Need the owner to re-scope whether the object may be removed in either base before any DDL; pending-prod-changes.md:765-774 also contains the superseded PROD-absent claim.
-
Elazar decision 2026-10-09: DROP rutap.v_export_estado_actual now, same as the 14-view batch (MOVILBA-674 etc).
-
db03 2026-10-09 14:49:22Z. Pre-apply pg_depend check (both bases, transitive): zero external dependent views - DEV's only hit (oid 18589846) is the view's own pg_rewrite rule, not a dependent. DROPPED rutap.v_export_estado_actual in PROD (via SUPERUSER_ELAZAR_URL + role=movilba_owner, rehearsed under ROLLBACK first) and DEV (movilba_user, owner). to_regclass NULL confirmed both bases post-drop. Per Elazar GO (coder03, MOVILBA-674 batch).
-
cleared: The ticket's DEV-only premise is stale: fresh PROD and DEV SELECTs show the view present in both, with 2,424/2,469 rows and zero dependent views. Source/tools has no callers, but that cannot rule out external consumers. Need the owner to re-scope whether the object may be removed in either base before any DDL; pending-prod-changes.md:765-774 also contains the superseded PROD-absent claim.