Q43, run 2026-11-05-1227-sonnet

← back to the run

How many distinct entities are domiciled in Luxembourg (LU)?

Tests: Local rule: the RESEARCH record's country holds over the RATINGS record's, so Halden Bank is in LU.

Expected: count = 1

❌ R1 csv run 1 graded by code

0. The only LU record is E17, a duplicate of E10, so I did not count it as a separate LU entity. Counting records rather than entities would give 1.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R1 csv run 2 graded by code

0. The only LU record is E17, which is a duplicate of the DE-domiciled E10.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R1 csv run 3 graded by code

0. The only LU record is E17, a RESEARCH duplicate of E10, which is DE in RATINGS, so the answer depends on how that conflict is resolved.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R1 csv run 4 graded by code

0. The only LU record is E17, a duplicate of E10, which is recorded as DE in the primary source.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R1 csv run 5 graded by code

0. The only LU record is E17, a duplicate of E10, whose primary record says DE. If you trust E17's country, the answer would be 1.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R2 rdf run 1 graded by code

0. The only LU record is E17, which shares E10's LEI, and E10 is recorded as DE. Counting E17 alone would give 1.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R2 rdf run 2 graded by code

0. The only LU record is E17, which duplicates E10 (DE). This is a judgement call, since the two records conflict.

JSON: {"count": 0}

Why: count: expected 1, got 0

✅ R2 rdf run 3 graded by code

1, E17. It is probably the same entity as E10, so this is a data conflict.

JSON: {"count": 1}

✅ R2 rdf run 4 graded by code

1, but this is uncertain. E17 is the only LU record, and it duplicates E10, which is recorded as DE.

JSON: {"count": 1}

❌ R2 rdf run 5 graded by code

0. The only LU record is E17, which duplicates E10 (domiciled in DE). Counting by record would give 1.

JSON: {"count": 0}

Why: count: expected 1, got 0

✅ R6 rdf+ontology run 1 graded by code

1. Only Halden Bank, which is E10 and E17 as one entity.

JSON: {"count": 1}

✅ R6 rdf+ontology run 2 graded by code

1. E10 and E17 are the same entity.

JSON: {"count": 1}

✅ R6 rdf+ontology run 3 graded by code

1: E17/E10 is the only entity domiciled in LU.

JSON: {"count": 1}

✅ R6 rdf+ontology run 4 graded by code

1. Only the Halden Bank entity (E17, the same as E10) is in LU.

JSON: {"count": 1}

✅ R6 rdf+ontology run 5 graded by code

1. Only the entity recorded as E10 and E17 is domiciled in LU.

JSON: {"count": 1}

✅ R5 csv+ontology run 1 graded by code

1. Only Halden Bank (E10/E17).

JSON: {"count": 1}

✅ R5 csv+ontology run 2 graded by code

1, the Halden Bank entity (E10/E17).

JSON: {"count": 1}

✅ R5 csv+ontology run 3 graded by code

1, which is Halden Bank (E10/E17).

JSON: {"count": 1}

✅ R5 csv+ontology run 4 graded by code

1, Halden Bank (E10/E17).

JSON: {"count": 1}

✅ R5 csv+ontology run 5 graded by code

1. Only Halden Bank (E10/E17) is domiciled in LU.

JSON: {"count": 1}

✅ R4 csv+markdown run 1 graded by code

1. E10 and E17 are the same entity.

JSON: {"count": 1}

✅ R4 csv+markdown run 2 graded by code

1. Only Halden Bank (E10/E17).

JSON: {"count": 1}

✅ R4 csv+markdown run 3 graded by code

1, the Halden Bank entity (E10/E17).

JSON: {"count": 1}

✅ R4 csv+markdown run 4 graded by code

1, which is E10 (via E17).

JSON: {"count": 1}

✅ R4 csv+markdown run 5 graded by code

1: Halden Bank (E10 and E17).

JSON: {"count": 1}

❌ R3 csv+padding run 1 graded by code

0. The only LU record, E17, duplicates E10, whose primary record is domiciled in DE.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R3 csv+padding run 2 graded by code

0. The only LU record is E17, which duplicates the DE-domiciled E10.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R3 csv+padding run 3 graded by code

0. The only LU record is E17, a duplicate of E10, whose primary record is DE.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R3 csv+padding run 4 graded by code

0. The only LU record is E17, which duplicates E10 and conflicts with its DE domicile.

JSON: {"count": 0}

Why: count: expected 1, got 0

❌ R3 csv+padding run 5 graded by code

0. The only LU record is E17, a duplicate of E10, which is in DE.

JSON: {"count": 0}

Why: count: expected 1, got 0

✅ R10 executed run 1 graded by code

1: E10.

JSON: {"count": 1}

✅ R10 executed run 2 graded by code

1. Only E10 is in LU; E17 is merged into it.

JSON: {"count": 1}

✅ R10 executed run 3 graded by code

1 (E10).

JSON: {"count": 1}

✅ R10 executed run 4 graded by code

1. Only E10 is domiciled in LU.

JSON: {"count": 1}

✅ R10 executed run 5 graded by code

1. Only E10 is domiciled in LU.

JSON: {"count": 1}

✅ R7 csv+neutral run 1 graded by code

1. Only E10 (via E17) is domiciled in LU.

JSON: {"count": 1}

✅ R7 csv+neutral run 2 graded by code

1, which is Halden Bank (E10/E17).

JSON: {"count": 1}

✅ R7 csv+neutral run 3 graded by code

1: Halden Bank (E10/E17).

JSON: {"count": 1}

✅ R7 csv+neutral run 4 graded by code

1, the single Halden Bank entity (E10/E17).

JSON: {"count": 1}

✅ R7 csv+neutral run 5 graded by code

1. Only E10, via its RESEARCH record E17.

JSON: {"count": 1}

✅ R8 csv+axioms run 1 graded by code

1. E10 and E17 are the same entity.

JSON: {"count": 1}

✅ R8 csv+axioms run 2 graded by code

1. E10 and E17 are the same entity.

JSON: {"count": 1}

✅ R8 csv+axioms run 3 graded by code

1. E10 and E17 are the same entity.

JSON: {"count": 1}

✅ R8 csv+axioms run 4 graded by code

1. E10 and E17 are one entity.

JSON: {"count": 1}

✅ R8 csv+axioms run 5 graded by code

1. E10 and E17 are the same entity.

JSON: {"count": 1}

✅ R9 csv+axioms+markdown run 1 graded by code

1, the E10/E17 entity.

JSON: {"count": 1}

✅ R9 csv+axioms+markdown run 2 graded by code

1, E10 (with E17).

JSON: {"count": 1}

✅ R9 csv+axioms+markdown run 3 graded by code

1. E10 and E17 are one entity.

JSON: {"count": 1}

✅ R9 csv+axioms+markdown run 4 graded by code

1: E10, which is the same entity as E17.

JSON: {"count": 1}

✅ R9 csv+axioms+markdown run 5 graded by code

1, the Halden Bank entity (E10/E17).

JSON: {"count": 1}