Debe MovilBA impedir que dos colectivos vigentes compartan linea + interno?

Sub-tickets

Activity

• wi-cli-venus created · 3w ago
• wi-cli-venus blocked · 3w ago
  Esperando respuesta de Elazar en CLI. P0 = bloqueado en Elazar, no urgente.
• coder04-mba-l-ca progress · 3w ago
  coder04-mba-l-ca | 2026-09-24 03:56 ART. Still blocked on Elazar: decision on 56 overlapping Fecha_Alta/Baja ranges in Para Eugenia #84, administrative event versus occupancy. BWI contains the 2026-09-23 request and no decision; no code/data change is authorized by an answer I can verify.
• audit-mba-l-ca audit-finding · 3w ago
  audit-mba-l-ca | 2026-09-24 04:58 ART. Audit finding and Elazar clarification: MovilBA exists to retire Eugenia's Excel by faithfully importing what it says. Workbook audits test our reading/transformation/import; they are not a request to have Eugenia change an unusual source record to fit an assumed occupancy rule. Distinct lineas may legitimately use the same interno number; overlap must be scoped to the same linea+interno, not interno alone. The cited PROD imported example for linea 12/interno 1 (PMJ020 2025-07-08..2026-02-12; AI009CM from 2026-01-21) was observed as overlapping date ranges, but overlap alone does not establish a workbook defect or justify withholding a faithful import. The ticket's original 56-case count is from its source census, not independently confirmed by this audit (a separate strict consecutive-row check yielded 55; definitions need reconciliation before citing an exact count). Next: coder03 should verify source-to-PROD fidelity and reconcile the 56/55 counting definitions; import correct source data and do not route these records to Eugenia merely for overlap. Leave date semantics as a separate question only if a MovilBA behavior actually requires an occupancy interpretation. No data or status changed by this audit.
• coder03-mba-l-cc unblocked · 3w ago
  coder03-mba-l-cc 2026-09-24 - Unblocked by Elazar's clarification, relayed by audit-mba-l-ca at 05:21 ART. This item was filed as a question about what Fecha_Alta/Fecha_Baja MEAN, to be put to Eugenia. That framing is withdrawn. MovilBA exists to retire the Excel, so an odd overlap in the book is an IMPORT-FIDELITY check, not a request that the stakeholder repair her source: if we imported it faithfully, it is not ours and we move on. Second constraint from the same clarification: the same Nro_Interno under DIFFERENT lineas is normal and is not an overlap at all, so only same-linea overlaps count. The remaining work is therefore a comparison, not a question - do the imported periods reproduce the book's same-linea overlapping internos exactly, neither inventing nor dropping any. NO COUNT IS CITED HERE AND THE TICKET'S 56 IS NOT CONFIRMED. My run over the current book /data-lezama/movilba/libro/2026-09-24-0434-parque-rutap.xlsx (sha256 94c3499d...15fa19, Base header on row 2, 5538 rows with Nro_dominio, 1844 linea+interno groups) returns 479 groups under both an all-pairs and a strict-consecutive-rows definition, difference zero. 479 against this ticket's 56 and audit's 55 is not a reconciliation, it is a broken instrument: I treat a missing Fecha_Baja as 9999-12-31, so every pair of open periods in a group intersects by construction and the open ones dominate the count. That has to be fixed before any number from me is usable, and the two definitions agreeing proves nothing while both share the same bad input. Script at scratchpad/overlap-566.py; it is re-derivable and deliberately not promoted.
• coder03-mba-l-cc titleChanged · 3w ago
  Fidelidad de import: los internos con periodos superpuestos DENTRO DE LA MISMA LINEA tienen que reproducirse exactos contra el libro
• coder03-mba-l-cc descriptionChanged · 3w ago
  was: Pregunta 84 de la hoja Para Eugenia del tracker. Elazar la saco del circuito de Eugenia con un comentario en el libro del 23/09 - literal: 'Too long. Ask me this question since I think I know the answer but I cant read all that' - asi que la celda quedo RETIRADA y la pregunta es para el. Se la hice en CLI el 23/09 21:0x, sin respuesta todavia. Forma corta que se le mando: en la linea 12, interno 1, el libro tiene a PMJ020 del 08/07/2025 al 12/02/2026 y a AI009CM desde el 21/01/2026 todavia abierto - 22 dias con dos colectivos en el mismo interno. Hay 56 casos asi. La pregunta: Fecha_Alta y Fecha_Baja marcan el periodo en que ese colectivo OCUPA ese interno (entonces los 56 son errores a corregir), o son fechas administrativas del expediente que pueden solaparse (entonces no hay nada que arreglar)? Denominadores de la medicion original, que no se reciclan a otra pregunta: 1599 colectivos vigentes con CERO internos repetidos dentro de la misma linea; contando tambien periodos cerrados, 631 combinaciones de linea+interno que usaron dominios distintos, y de esas 26 son el mismo chasis con dos dominios (replate, no reuso), 549 son chasis distintos en epocas que NO se superponen (reuso normal despues de una baja) y 56 son chasis distintos con fechas QUE SE SUPERPONEN. Medido sobre los 5528 renglones del libro del 19/09. La respuesta decide si los 56 son un defecto del libro; hasta que conteste no se toca ninguna fila ni se le lleva el caso a Eugenia.
• coder03-mba-l-cc titleChanged · 3w ago
  Debe MovilBA impedir que dos colectivos vigentes compartan linea + interno?
• coder03-mba-l-cc assigned · 2w ago
  db03-mba-l-cc
• coder03-mba-l-cc note · 2w ago
  2026-09-28 13:30 ART, coder03-mba-l-cc. Elazar routed this to Eugenia, not to himself: written into 'Para Eugenia' as question 91, state PENDIENTE. Re-measured in PROD before writing it, because the claim is about OUR data: 1602 colectivos with estado ALTA carrying both linea and interno, and ZERO repeated (linea_id, interno) combinations. Close to the 1599 counted on the 19/09 book, and it means a uniqueness rule would reject none of today's data. Positive control on the same query: 1745 distinct combinations exist overall, so the zero is a real zero and not an empty instrument. One thing the question had to name explicitly: vehicles given de BAJA keep their last interno, and among ALL live VDC rows there are 531 repeated combinations - 4 dominios share linea 138 interno 424. Any rule must therefore be scoped to vigentes only, or it fires on history.
• coder03-mba-l-cc blocked · 2w ago
  Awaiting Eugenia's answer to question 91 in the tracker.
• coder03-mba-l-cc assigned · 2w ago
  db04-mba-l-ca
• coder03-mba-l-cc note · 1w ago
  Stakeholder feedback, tracker online 2026-10-05 13:24 (Eugenia/Fabri), relayed by coder03: Estado: 'Segun lo conversado recientemente, no debe tener restriccion al dar de alta o modificacion que exista vehiculos con el mismo interno en la misma linea.' Verificado (describes current app): 'no puede repetirse el interno > se tiene que dar de baja y despues alta.' New row (Euge y Fabri): 'Puedo cargar un dominio con un interno ya vinculado a otro dominio (LID108)'. DECIDED: remove the same-linea+interno restriction on alta and modificacion.
• coder03-mba-l-cc unblocked · 1w ago
• coder03-mba-l-cc assigned · 1w ago
  coder04-mba-l-ca / coder
• coder03assistant-mba-l-ca note · 1w ago
  coder03assistant-mba-l-ca | 2026-10-05 13:55 ART 2026-10-05-1324 online download (SHA 20b5ca07…), Pendientes!J11, MOVILBA-566 #91, carries this text in the Estado cell: 'Segun lo conversado recientemente, no debe tener restriccion al dar de alta o modificacion que exista vehiculos con el mismo interno en la misma linea.' It conflicts with the existing Respuesta 'Lo rechaza el sistema.' and the download's Verificado text 'Verificado, no puede repetirse el interno > se tiene que dar de baja y despues alta.' I have recorded this source statement here, not copied it or internal analysis into a stakeholder-facing cell; the candidate leaves existing Estado/Respuesta unchanged and merges the new Verificado cell. The conflict remains for item handling.
• coder04-mba-l-ca progress · 1w ago
  coder04-mba-l-ca | 2026-10-05 13:55 ART. Stakeholder answer recorded: allow duplicate same-linea+interno on Alta and Modificacion (LID108). Source census: no duplicate lookup/error path in the Alta form, edit form, actions, or services; createColectivo/updateColectivo only validate that interno is present. Tracked DB schema has chk_vdc_interno_activo = borrado_en IS NOT NULL OR interno IS NOT NULL, a non-null CHECK, not a uniqueness rule; no same-linea+interno unique index appears in Drizzle/migrations. An installed trigger outside tracked definitions is not yet ruled out; asked db03 for read-only PROD/DEV confirmation before claiming end-to-end behavior.
• coder04-mba-l-ca completed · 1w ago
  coder04-mba-l-ca | 2026-10-05 14:00 ART. Stakeholder ruling: allow duplicate vigente colectivos with the same linea+interno on Alta and Modificacion. Source census found no duplicate check in forms, actions, or createColectivo/updateColectivo. db03 read-only inspected PROD and DEV: no unique index, constraint, trigger, or installed function enforces the pair; chk_vdc_interno_activo only requires interno for active rows, and the two table triggers are unrelated archive/soft-delete behavior. No code removal was needed; requested behavior is already allowed.