movilba
MOVILBA-401
Vehicle chassis must not contradict its propulsion
Done normal
rmresearcher-mba-l-cc
A vehicle chassis must not contradict its own propulsion. db02 measured 15 live vehicles in AGRALE MT 17.0 LE electric/GNC source tuples already linked to diesel chassis 127 before MOVILBA-398c; resolver corrections did not change vehicles. Own a read-only all-manufacturer census using current-period propulsion resolved through catalog labels and linked chassis variant evidence. Establish exact population and source contradictions before proposing any separate vehicle write/GO. Research: /data-lezama/movilba/research/chasis/2026-09-08-agrale-model-mappings.md . No vehicle changes authorized by this item.
Activity
-
db02 reports approved DEV repair applied, md5 7f785d597f7455e99ebabdd658a4ed56: exactly 15 pre-existing Agrale mislinks corrected; this is not the all-manufacturer census. Requested exhaustive read-only export from db02 without manufacturer whitelist, resolving propulsion option labels and separating proven mismatch, ambiguous and unlinked. Census analysis remains owned by coder-mba-l-ca; no additional vehicle repair implied.
-
DEV db02 applied 401:15 incorrect Agrale electric/gas links corrected;402 linked13 previously NULL gas vehicles, total unlinked38->25. All-manufacturer propulsion-name census reports0 detected contradictions but examines only86 linked rows with explicit propulsion markers;2301 of2387 linked rows have neutral catalog names and are NOT established correct by this method. Latest export /data-lezama/import-reports/2026-09-08-movilba-401-censo-contradiccion-propulsion-todas-las-marcas-DEV.tsv md5 555eab5aa5509c6d4f0597afe8bc39e4. 401 artifact shipped in1ffd5465/v5.5.97, deploy typecheck/health/freshness passed.402 applied DB artifact pending next ship.
-
Correction to tested denominator after coder-mba independent check:85 marker-bearing vehicles were tested, not86. One additional linked vehicle lacks source propulsion and is untestable.2387 linked=2301 neutral+85 tested+1 missing propulsion. db02 confirmed LEFT JOIN population preserves all2412 nondeleted vehicles and matches independent count. coder-mba separately established KingLong XMQ6127G natural-gas identity from importer technical page and Yutong ZK6128HG GNC;291 previously neutral GN links are evidence-compatible, not SQL repairs.
-
Independent census review saved /data-lezama/movilba/research/chasis/2026-09-08-propulsion-census-review.md. Summed 92 export groups:2412 live=2387 linked+25 unlinked. db02 independently confirms population and LEFT joins; corrected85 tested, not86. Name-only2301neutral is not correctness. Independently opened Pudong XMQ6127G technical page and Yutong Argentina:291 previously neutral GN links fuel-compatible;6 neutral GN variants remain unresolved. Retired-reference distribution stale after200a/b;6 E5 intersection discrepancy retained. No SQL writes by this review; evidence sent coder04/db02.
-
Completed batched row-by-row evidence overlay using post200ab export MD5 a62a3ff617d0b326a14b81e5f7bceeed. PASS92/92groups,2412/2412vehicles,source columns unchanged,reproducer equality.2197 fuel-compatible with explicit evidence strengths;85 original marker-only;104 linked unresolved;25 unlinked;1 missing source. No proven new contradiction; no SQL changes. Full report and post200ab CSV under /data-lezama/movilba/research/chasis/2026-09-08-propulsion-census-review.md and 2026-09-08-propulsion-census-post200ab-evidence-overlay.csv. One consolidated handoff to coder04/db02; KingLong durable source excerpt included. Audit disposition complete for this export, unresolved identity work not falsely closed.
-
coder-mba-l-cc
-
coder03-mba-l-cc / coder
-
coder03assistant-mba-l-ca
-
researcher-mba-l-cc / solo
-
researcher-mba-l-cc claiming per coder03's 2026-10-08 11:00 reassignment: build chassis-model -> allowed-propulsion table from manufacturer datasheets, then hand the predicate to db03 for a PROD SELECT naming contradicting rows.
-
researcher-mba-l-cc 2026-10-08T11:01:30-03:00. Chassis-model -> allowed-propulsion table built from the live PROD chasis catalog (33 rows, MOVILBA-597's 2026-10-06 export) + manufacturer datasheets. Full table + predicate: /data-lezama/movilba/evidencia/MOVILBA-401/2026-10-08-chasis-modelo-propulsion-permitida.md. 27/33 chasis_id are single-propulsion (23 diesel, 4 GNC, 2 electric), each with a manufacturer-source citation (reused Agrale D17/H17/G17/E17/E6/D12/SB17/R15 evidence already approved for MOVILBA-398/365; new citations for KING LONG XMQ6127G and YUTONG ZK6128HG from the 2026-09-08 propulsion-census note). 2 exclusions, do not treat as violations if found mismatched: - chasis_id 135 (AGRALE MT 17, bare) — MOVILBA-597's own 2026-09-08 blackboard decision deliberately keeps this id multi-propulsion/generic. Not a single-propulsion id by design. - chasis_id 136 (KING LONG XMQ6127E) — ran WebSearch today, found no King Long source distinguishing "E" as a fuel-type suffix (only diesel XMQ6127 variants surfaced, Cummins engine). Do NOT assume electric by analogy with 117's "G"=gas pattern. Flagging as unconfirmed, not asserting a propulsion. Handoff: db03 fills the exact opciones.id values for GO/GNC/ELECTRICO+ELECTRICO-CABA (per catalogos.md's fixed mapping) into the predicate in the doc and runs it read-only against PROD current-period vehicles, chasis_id NOT IN (135,136). Expected output is a short dominio+chasis+propulsion contradiction list, same defect class as the 15 Agrale mislinks MOVILBA-398c already corrected, but for the other 31 chasis ids — never censused this way before. No vehicle write authorized by this item, per its own description.
-
db03-mba-l-cc 2026-10-08. SELECT-only PROD, predicate from researcher-mba-l-cc's table, opciones.id filled live: GO=124, GNC=157, ELECTRICO(+ELECTRICO-CABA collapse)=37 (grupo_nombre='motor', campo_nombre='unidad_propulsora', borrado_en IS NULL; column is vehiculos.motor_unidad_propulsora_id, not vdc - corrected from the handed predicate). chasis_id 135/136 excluded per the table. Result: 1 contradiction (not zero). AE016ZA | chasis_id 142 | SCANIA 280 B 4X2 (diesel-only, medium confidence) | vehiculo's recorded propulsion = GNC (id 157) Same defect class as the 15 Agrale mislinks (MOVILBA-398c). No vehicle write done - read-only per scope note. Needs same GO/decision path as 398c before any fix.
-
researcher-mba-l-cc 2026-10-08T11:12:00-03:00. Re: db03's 1 contradiction (AE016ZA, chasis_id 142, SCANIA 280 B 4X2, motor=GNC). WebSearch confirms Scania's 280hp K-series engine (9L, 5-cyl) is genuinely offered as BOTH diesel and GNC in Argentina - the GNC variant is a real, deployed CABA product (tested with CABA transporte + linea 132, 4 roof-mounted T4 tanks, 868L capacity). This is the SAME engine platform as catalog ids 115 (K 280 B4x2 diesel) and 124 (K 280 B4x2 GNC), which the catalog already splits correctly by propulsion. Read: chasis_id 142's text ("280 B 4X2", no "K" prefix) is most likely a MOVILBA-597-shaped model-IDENTITY ambiguity, not a genuine propulsion-data error on AE016ZA's vehicle row. The vehicle's own GNC value is plausible and consistent with a real Scania product; what's more likely wrong is the CHASIS LINK itself - AE016ZA may belong under 124 (K 280 B4x2 GNC), not under the diesel-flagged 142. My original table correctly flagged 142 only medium-confidence precisely because it lacked an explicit diesel/GNC marker, unlike 115/124's split. Recommendation: before writing anything, db03/coder03 check whether 142 is a duplicate/near-alias of 115 or 124 in MOVILBA-597's own chasis-model text-matching work (same ticket already tracks 72 model-text pairs) rather than treating this as a new propulsion-vs-chasis defect requiring a vehicle-side fix. If 142 turns out to be a distinct, genuinely diesel-only model, then AE016ZA is a real contradiction and the fix is on the vehicle/chasis link, same shape as the 15 Agrale cases MOVILBA-398c corrected. No vehicle write authorized by this item either way.
-
db03-mba-l-cc 2026-10-08. Verdict for AE016ZA/chasis_id 142, SELECT-only PROD. Book (rutap.rutap_rows, snapshot 11, all 3 historial rows for this vehicle, filas 1826/2269/5157): chasis_marca=SCANIA, chasis_modelo='280 B 4x2', unidad_propulsora=GNC. Consistent across every row, not a one-off typo. Vehicle's resolved link: chasis_id=142 (catalog label '280 B 4X2'), numero_serie_chasis=9BSK4X200K3942253. Grafia match is EXACT (case-insensitive) between book raw text and catalog label - this is NOT a resolver mislink, unlike 398c's Agrale cases where the book text matched a DIFFERENT catalog row than the one linked. Catalog coverage gap, not a mislink: 115 ('K 280 B4x2 diesel') and 124 ('K 280 B4x2 GNC') both carry an explicit K-prefix + fuel suffix in their own labels. 142 ('280 B 4X2') carries neither - no K, no fuel marker. Researcher's diesel-only call on 142 was inferred by elimination ("parallels 115/124 split, no GNC marker"), not sourced from any manufacturer spec naming THIS chasis_id specifically. Verdict: genuine book contradiction against an UNDER-EVIDENCED catalog restriction, not our-side mislink. Same genus as 135 (AGRALE MT17 bare) being deliberately left multi-propulsion rather than forced to one fuel - 142 likely needs the same treatment, or needs its own manufacturer source before being asserted diesel-only. Does not meet the bar for an Eugenia tracker cell (catalog/research question, not a book-text ambiguity for her to resolve) - leaving to coder03/researcher to decide whether 142 gets reclassified as ambiguous-propulsion (like 135) or researched further.
-
researcher-mba-l-cc 2026-10-08T11:14:00-03:00. Verdict on AE016ZA/chasis_id 142 per db03's book pull (snapshot 11, 3 historial rows consistent, chasis_modelo verbatim '280 B 4x2', propulsion GNC). Reclassifying chasis_id 142 from "diesel only, medium confidence" to AMBIGUOUS/multi-propulsion, same genus as 135. Basis: WebSearch today confirms Scania's 280hp K-series platform (9L, 5-cyl) is genuinely sold as both diesel and GNC in Argentina - the GNC conversion was deployed with CABA transporte on linea 132 (4 roof T4 tanks, 868L, tested in CABA/Cordoba/Salta/Mendoza). My original diesel-only call on 142 was inference-by-elimination from the label's missing K-prefix/fuel-suffix, not a sourced spec for 142 itself - db03 is right that this is under-evidenced, not a typo or resolver mislink. The book is internally consistent (3/3 historial rows agree) and matches a real manufacturer product, so this is NOT an Eugenia question (catalog-first rule doesn't apply - there's no catalog/book text mismatch here, both say GNC). Updated table and predicate at /data-lezama/movilba/evidencia/MOVILBA-401/2026-10-08-chasis-modelo-propulsion-permitida.md: chasis_id 142 moves to the exclusion list alongside 135/136 (3 excluded of 33, not 2). AE016ZA is not a contradiction - no tracker cell. Net result of the full census: 0 confirmed contradictions among the 30 single-propulsion chasis_id, 3 excluded as ambiguous/unconfirmed (135, 136, 142), pending further evidence on 136/142 if anyone wants to narrow them later. No vehicle write, no tracker write.
-
researcher-mba-l-cc 2026-10-08T14:14:25-03:00. Closing per coder03's request. Census of 30 single-propulsion chasis_id against PROD current-period vehicles: 0 confirmed propulsion contradictions. 3 excluded as ambiguous/multi-propulsion by manufacturer evidence and db03's SELECT verification (135, 136, 142) - none are defects, no vehicle write authorized or needed. Full table and predicate at /data-lezama/movilba/evidencia/MOVILBA-401/2026-10-08-chasis-modelo-propulsion-permitida.md.
bug
5w ago by wi-cli-venus
1w ago
2026-10-08 17:14