fix(api): fall back to legacy name-column query when marts predate code migration
PR Checks / Frontend Typecheck + Tests (pull_request) Successful in 9m41s
PR Checks / Backend Smoke (pull_request) Successful in 6s
PR Checks / Build Backend (no push) (pull_request) Successful in 19s
PR Checks / Build Frontend (no push) (pull_request) Successful in 47s
PR Checks / Build Pipeline (no push) (pull_request) Successful in 10s
PR Checks / AI Code Review (Claude) (pull_request) Failing after 3m8s

Closes the deploy window flagged by CI review — the backend now works
against both the old (name) and new (code) mart schemas.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Tudor
2026-07-09 14:48:06 +01:00
co-authored by Claude Fable 5
parent 84dfc6c1bb
commit 4b75152ee0
2 changed files with 104 additions and 11 deletions
+52
View File
@@ -42,3 +42,55 @@ def test_missing_code_columns_are_a_noop():
out = translate_gias_code_columns(df)
assert out.iloc[0]["phase"] == "Primary"
assert out.iloc[0]["status"] == "Open"
def test_load_school_data_survives_premigration_marts(monkeypatch):
"""Real prod state until the nightly pipeline first rebuilds the mart with
the GIAS code columns: marts.dim_school still has the old name columns
(phase, school_type, religious_character, status, admissions_policy)
instead of the new *_code columns. The first query raises UndefinedColumn
on s.phase_code; load_school_data_as_dataframe must retry with the
legacy name-column query rather than swallow the error and return (and
then have load_school_data cache) an empty DataFrame."""
import sqlalchemy.exc
from backend import data_loader
data_loader._df_cache = None
data_loader._df_latest_cache = None
good_df = pd.DataFrame(
[
{
"urn": 1,
"school_name": "Legacy School",
"phase": "Primary",
"school_type": "Academy",
"status": "Open",
}
]
)
calls = []
def fake_read_sql(query, con):
calls.append(query)
if len(calls) == 1:
raise sqlalchemy.exc.ProgrammingError(
"(psycopg2.errors.UndefinedColumn) column s.phase_code does not exist",
None,
None,
)
return good_df.copy()
monkeypatch.setattr(data_loader.pd, "read_sql", fake_read_sql)
try:
df = data_loader.load_school_data_as_dataframe()
finally:
data_loader._df_cache = None
data_loader._df_latest_cache = None
assert len(calls) == 2, "must retry with the legacy name-column query variant"
assert calls[1] is data_loader._MAIN_QUERY_LEGACY_NAMES
assert not df.empty
assert df["phase"].iloc[0] == "Primary"
assert df["status"].iloc[0] == "Open"