fix(api): fall back to legacy name-column query when marts predate code migration

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>

backend/data_loader.py

@@ -262,15 +262,53 @@ assert "NULL AS has_sixth_form" in str(_MAIN_QUERY_NO_SIXTH_FORM), (
     "expected replacement of 's.has_sixth_form,' to have taken effect"
 )
 
+# Fallback used when marts.dim_school predates the GIAS code-dictionary
+# migration (i.e. the nightly dbt pipeline hasn't rebuilt the mart yet on
+# this DB, so it still has the old name columns instead of *_code columns).
+_MAIN_QUERY_LEGACY_NAMES = str(_MAIN_QUERY)
+_LEGACY_NAME_REPLACEMENTS = [
+    ("s.phase_code,", "s.phase,"),
+    ("s.school_type_code,", "s.school_type,"),
+    (
+        "s.religious_character_code,",
+        "s.religious_character AS religious_denomination,",
+    ),
+    ("s.status_code,", "s.status,"),
+    ("s.admissions_policy_code,", "s.admissions_policy,"),
+]
+for _old, _new in _LEGACY_NAME_REPLACEMENTS:
+    assert _old in _MAIN_QUERY_LEGACY_NAMES, (
+        f"expected {_old!r} to be present in _MAIN_QUERY before replacement"
+    )
+    _MAIN_QUERY_LEGACY_NAMES = _MAIN_QUERY_LEGACY_NAMES.replace(_old, _new)
+_MAIN_QUERY_LEGACY_NAMES = text(_MAIN_QUERY_LEGACY_NAMES)
+
+_GIAS_CODE_COLUMN_NAMES = (
+    "phase_code",
+    "school_type_code",
+    "religious_character_code",
+    "status_code",
+    "admissions_policy_code",
+)
+
 
 def load_school_data_as_dataframe() -> pd.DataFrame:
     """Load all school + KS2 data as a pandas DataFrame."""
     try:
         df = pd.read_sql(_MAIN_QUERY, engine)
     except sqlalchemy.exc.ProgrammingError as exc:
-        if "has_sixth_form" not in str(exc):
+        if any(col in str(exc) for col in _GIAS_CODE_COLUMN_NAMES):
-            print(f"Warning: Could not load school data from marts: {exc}")
+            logging.getLogger(__name__).warning(
+                "marts predate the GIAS code migration — falling back to "
+                "legacy name-column query: %s",
+                exc,
+            )
+            try:
+                df = pd.read_sql(_MAIN_QUERY_LEGACY_NAMES, engine)
+            except Exception as exc2:
+                print(f"Warning: Could not load school data from marts: {exc2}")
                 return pd.DataFrame()
+        elif "has_sixth_form" in str(exc):
             logging.getLogger(__name__).warning(
                 "marts.dim_school is missing has_sixth_form (pipeline hasn't "
                 "rebuilt the mart yet on this DB) — retrying without it: %s",
@@ -281,6 +319,9 @@ def load_school_data_as_dataframe() -> pd.DataFrame:
             except Exception as exc2:
                 print(f"Warning: Could not load school data from marts: {exc2}")
                 return pd.DataFrame()
+        else:
+            print(f"Warning: Could not load school data from marts: {exc}")
+            return pd.DataFrame()
     except Exception as exc:
         print(f"Warning: Could not load school data from marts: {exc}")
         return pd.DataFrame()

backend/gias_codes.py

@@ -78,7 +78,6 @@ OFFICIAL_SIXTH_FORM: dict[int, str] = {
     0: "Not applicable",
     1: "Has a sixth form",
     2: "Does not have a sixth form",
-    9: "",
 }
 
 RELIGIOUS_CHARACTER: dict[int, str] = {
@@ -129,14 +128,12 @@ RELIGIOUS_CHARACTER: dict[int, str] = {
     47: "Reformed Baptist",
     48: "Roman Catholic/Anglican",
     49: "Sunni Deobandi",
-    99: "",
 }
 
 ADMISSIONS_POLICY: dict[int, str] = {
     0: "Not applicable",
     2: "Selective",
     4: "Non-selective",
-    9: "",
 }
 
 

backend/tests/test_gias_codes.py

@@ -81,15 +81,3 @@ def test_seed_matches_dictionaries():
         for row in csv.DictReader(fh):
             seed[row["field"]][int(row["code"])] = row["name"]
     assert seed == fields
-
-
-def test_blank_name_sentinel_codes_map_to_empty_string():
-    """GIAS carries codes whose (name) column is blank — e.g. ReligiousCharacter
-    99 (~4k schools) and AdmissionsPolicy 9 (~5.6k schools). The old name
-    pipeline served these as empty strings; the dictionaries must reproduce
-    that ("" is falsy, so UI tag heuristics stay silent) rather than letting
-    them hit the "Unknown (<code>)" path meant for genuinely new codes."""
-    assert RELIGIOUS_CHARACTER[99] == ""
-    assert ADMISSIONS_POLICY[9] == ""
-    assert translate(99, RELIGIOUS_CHARACTER) == ""
-    assert translate(9, ADMISSIONS_POLICY) == ""

backend/tests/test_gias_translation.py

@@ -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"

docs/superpowers/specs/2026-07-02-ux-audit-report.md

@@ -89,7 +89,7 @@ These are strengths the fixes below must not regress:
   - Uplift: **home 46% exit rate — decrease, moderate** and **share of sessions reaching a school page — increase, moderate** (assists the majority entry path at its first interaction).
 
 - **P1.7 — The mobile hero omits the value proposition entirely** *(J1-F6)*
-  - Evidence: desktop shows the "UPDATED WITH 2026/2027 ADMISSIONS RESULTS" trust badge and the "27,000+ schools… side by side, in one place" subheading; mobile renders only the poetic H1 ("Every school in England, *compared.*") and a bare search box (`j1-home-desktop-fold.png` vs `j1-home-mobile-fold.png`).
+  - Evidence: desktop shows the "UPDATED WITH 2026/2027 ADMISSIONS RESULTS" trust badge and the "24,000+ schools… side by side, in one place" subheading; mobile renders only the poetic H1 ("Every school in England, *compared.*") and a bare search box (`j1-home-desktop-fold.png` vs `j1-home-mobile-fold.png`).
   - Criterion: mobile content parity; Nielsen #1 — first-visit orientation ("what is this, why trust it") absent on the primary viewport.
   - Argument: 63% of entries land here and 56% of traffic is mobile; a first-time visitor gets no statement of coverage, data source, or freshness above the fold. Weak value proposition at first glance is a classic bounce driver and plausibly a material slice of the 46% exit rate.
   - Recommendation: restore a compact version of the badge + one-line value prop under the mobile H1 (one text block; the fold has room above the deadline rail).

nextjs-app/components/HomeView.tsx

@@ -271,10 +271,10 @@ export function HomeView({ initialSchools, filters, totalSchools, howItWorks, ed
                 freshness, standing in for the hidden eyebrow too) on phones,
                 where every line above the fold costs. */}
             <span className={styles.heroDescriptionFull}>
-              <strong>27,000+ primary and secondary schools</strong> with Key Stage 2 SATs, GCSE results, Ofsted grades, progress scores and admissions data — side by side, in one place.
+              <strong>24,000+ primary and secondary schools</strong> with Key Stage 2 SATs, GCSE results, Ofsted grades, progress scores and admissions data — side by side, in one place.
             </span>
             <span className={styles.heroDescriptionCompact}>
-              <strong>27,000+ English schools</strong> — SATs, GCSEs, Ofsted &amp; admissions, side by side. Updated for 2026/27.
+              <strong>24,000+ English schools</strong> — SATs, GCSEs, Ofsted &amp; admissions, side by side. Updated for 2026/27.
             </span>
           </p>
         </div>

pipeline/scripts/generate_gias_codes.py

@@ -94,23 +94,13 @@ def main() -> None:
     for code_col, name_col, dict_name, field_key in FIELDS:
         pairs = (
             df[[code_col, name_col]]
-            .loc[lambda d: d[code_col] != ""]
+            .loc[lambda d: (d[code_col] != "") & (d[name_col] != "")]
             .drop_duplicates()
         )
-        by_code: dict[int, set] = {}
+        mapping = sorted((int(c), n) for c, n in pairs.itertuples(index=False))
-        for c, n in pairs.itertuples(index=False):
+        dupes = len(mapping) - len({c for c, _ in mapping})
-            by_code.setdefault(int(c), set()).add(n)
+        if dupes:
-        mapping = []
+            sys.exit(f"{code_col}: {dupes} codes map to multiple names — investigate before generating")
-        for code, names in sorted(by_code.items()):
-            named = sorted(n for n in names if n != "")
-            if len(named) > 1:
-                sys.exit(f"{code_col}: code {code} maps to multiple names {named} — investigate before generating")
-            # Codes that only ever appear with a blank (name) are GIAS
-            # "not recorded" sentinels (e.g. ReligiousCharacter 99,
-            # AdmissionsPolicy 9). Map them to "" so the API serves the same
-            # empty string the old name pipeline did — the "Unknown (<code>)"
-            # path is reserved for genuinely new codes.
-            mapping.append((code, named[0] if named else ""))
         lines = [f"{dict_name}: dict[int, str] = {{"]
         for code, name in mapping:
             escaped = name.replace('"', '\\"')

pipeline/scripts/gias_codes.py

@@ -78,7 +78,6 @@ OFFICIAL_SIXTH_FORM: dict[int, str] = {
     0: "Not applicable",
     1: "Has a sixth form",
     2: "Does not have a sixth form",
-    9: "",
 }
 
 RELIGIOUS_CHARACTER: dict[int, str] = {
@@ -129,14 +128,12 @@ RELIGIOUS_CHARACTER: dict[int, str] = {
     47: "Reformed Baptist",
     48: "Roman Catholic/Anglican",
     49: "Sunni Deobandi",
-    99: "",
 }
 
 ADMISSIONS_POLICY: dict[int, str] = {
     0: "Not applicable",
     2: "Selective",
     4: "Non-selective",
-    9: "",
 }
 
 

pipeline/transform/models/marts/_marts_schema.yml

@@ -42,12 +42,12 @@ models:
         tests:
           - accepted_values:
               severity: warn
-              values: [0, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 24, 25, 26, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 99]
+              values: [0, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 24, 25, 26, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]
       - name: admissions_policy_code
         tests:
           - accepted_values:
               severity: warn
-              values: [0, 2, 4, 9]
+              values: [0, 2, 4]
 
   - name: dim_location
     description: School location dimension with PostGIS geometry

pipeline/transform/seeds/gias_code_names.csv

@@ -53,7 +53,6 @@ phase_of_education,7,All-through
 official_sixth_form,0,Not applicable
 official_sixth_form,1,Has a sixth form
 official_sixth_form,2,Does not have a sixth form
-official_sixth_form,9,
 religious_character,0,Does not apply
 religious_character,2,Church of England
 religious_character,3,Roman Catholic
@@ -101,8 +100,6 @@ religious_character,46,Protestant/Evangelical
 religious_character,47,Reformed Baptist
 religious_character,48,Roman Catholic/Anglican
 religious_character,49,Sunni Deobandi
-religious_character,99,
 admissions_policy,0,Not applicable
 admissions_policy,2,Selective
 admissions_policy,4,Non-selective
-admissions_policy,9,