Merge pull request 'fix(sec): derive fiscal year end from the issuer's own 10-K' (#3) from fix/sec-fundamentals-parity-gaps into main

Reviewed-on: #3

app/services/sec_facts_parser.py

@@ -179,15 +179,19 @@ def parse_snapshots(
 ) -> ParseResult:
     """Build snapshot rows for ``accessions`` (those with facts + filing meta).
 
-    ``fiscal_year_end`` is the issuer's ``submissions.fiscalYearEnd`` (MMDD) and
+    ``fiscal_year_end`` is the issuer's declared ``submissions.fiscalYearEnd``
-    is what makes period identity independent of SEC's unreliable fy/fp fields
+    (MMDD) and seeds period identity (see ``_period_identity``), making it
-    (see ``_period_identity``). Omitting it falls back to the old fy/fp behaviour.
+    independent of SEC's unreliable fy/fp fields. It is only a hint: the issuer's
+    own 10-K period ends override it (see ``resolve_fiscal_year_end``). With
+    neither available, the old fy/fp behaviour is used.
 
     ``skipped_filings`` = no row produced (missing facts/meta or no usable period
     identity); ``field_issues`` = a row was produced but a field is null/ambiguous.
     Callers must not use field issues as failed-row coverage.
     """
     cik = f"{int(companyfacts['cik']):010d}"
+    # The declared value is only a hint; the issuer's own 10-Ks are authoritative.
+    fiscal_year_end = resolve_fiscal_year_end(filings, fiscal_year_end)
     by_accn = _index_by_accession(companyfacts)
     result = ParseResult()
     for accn in accessions:
@@ -291,6 +295,29 @@ def _parse_one(
     return row, ("ambiguous shares outstanding" if ambiguous else None)
 
 
+def resolve_fiscal_year_end(
+    filings: dict[str, FilingMeta], declared: str | None
+) -> str | None:
+    """The issuer's fiscal-year-end MMDD, preferring its own 10-K period ends.
+
+    ``submissions.fiscalYearEnd`` is *not* reliable: Franklin Resources (BEN)
+    declares 1231 while every one of its 10-Ks ends 09-30. Trusting it put BEN's
+    fiscal Q1 (Dec) 0 days from the claimed year end — matching no quarter band —
+    and labelled its fiscal Q2 (Mar) as Q1, colliding two periods on one key and
+    destroying the quarter chain.
+
+    A 10-K's reportDate **is** the fiscal year end by definition, so it wins
+    whenever one is available; the declared value is only a fallback for an issuer
+    with no annual filing in the set. The most recent 10-K is used, so an issuer
+    that changed its year end is measured against its current calendar.
+    """
+    annual = [m.report_date for m in filings.values() if m.form.startswith("10-K")]
+    if annual:
+        latest = max(annual)
+        return f"{latest.month:02d}{latest.day:02d}"
+    return declared
+
+
 def _period_identity(
     meta: FilingMeta, fiscal_year_end: str | None
 ) -> tuple[int | None, str | None]:

tests/unit/test_sec_facts_parser.py

@@ -494,3 +494,47 @@ def test_weighted_average_shares_falls_back_to_the_basic_and_diluted_concept():
                                datetime(2026, 11, 1, 10, tzinfo=UTC), "10-Q")}
     res = parse_snapshots(companyfacts, filings, {"X"}, fiscal_year_end="1231")
     assert res.rows[0].weighted_avg_diluted_shares == 500_000
+
+
+# -- fiscal year end resolution (submissions.fiscalYearEnd is unreliable) -----
+
+from app.services.sec_facts_parser import resolve_fiscal_year_end  # noqa: E402
+
+
+def test_the_issuers_own_10k_overrides_a_wrong_declared_year_end():
+    # Franklin Resources declares 1231 while every 10-K ends 09-30. Trusting the
+    # declaration labelled its fiscal Q2 (Mar) as Q1, colliding with the real
+    # fiscal Q1 (Dec) and destroying the quarter chain.
+    filings = {
+        "K": _meta("2025-09-30", "10-K"),
+        "Q": _meta("2025-12-31"),
+    }
+    assert resolve_fiscal_year_end(filings, "1231") == "0930"
+
+
+def test_declared_year_end_is_used_when_no_annual_filing_is_present():
+    assert resolve_fiscal_year_end({"Q": _meta("2026-03-31")}, "1231") == "1231"
+    assert resolve_fiscal_year_end({}, None) is None
+
+
+def test_a_wrong_declared_year_end_no_longer_collides_two_periods():
+    """End to end: BEN's Dec and Mar quarters must land on distinct keys."""
+    def _q(accn, start, end, val):
+        return _dur(start, end, val, accn, fp="Q1")
+
+    companyfacts = {
+        "cik": 38777,
+        "facts": {"us-gaap": {"Revenues": {"units": {"USD": [
+            _q("Q1", "2025-10-01", "2025-12-31", 2327),   # fiscal Q1
+            _q("Q2", "2025-10-01", "2026-03-31", 4622),   # fiscal Q2 YTD
+        ]}}}},
+    }
+    filings = {
+        "K": _meta("2025-09-30", "10-K"),
+        "Q1": _meta("2025-12-31"),
+        "Q2": _meta("2026-03-31"),
+    }
+    res = parse_snapshots(companyfacts, filings, {"Q1", "Q2"}, fiscal_year_end="1231")
+    keys = {(r.fiscal_year, r.fiscal_period) for r in res.rows}
+    assert len(keys) == 2, f"periods collided on one key: {keys}"
+    assert keys == {(2026, "Q1"), (2026, "Q2")}