feat(api): filter schools by type group and by faith

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

backend/app.py

@@ -41,6 +41,14 @@ from .data_loader import get_data_info as get_db_info
 from . import flags
 from .places import build_place_index, build_place_registry, places_for_urn
 from .schemas import METRIC_DEFINITIONS, PHASE_GROUPS, RANKING_COLUMNS, SCHOOL_COLUMNS
+from .school_groups import (
+    FAITH_GROUPS,
+    FAITH_KEYS,
+    TYPE_GROUP_KEYS,
+    TYPE_GROUPS,
+    faith_groups_for,
+    type_group_for,
+)
 from .nearby_schools import select_nearby
 from .utils import clean_for_json, convert_to_native
 
@@ -608,6 +616,15 @@ def verify_admin_api_key(x_api_key: str = Header(None)) -> bool:
 
 
 # Input validation helpers
+def _names_in_group(names: pd.Series, in_group) -> set:
+    """The distinct names in a column that a group predicate accepts.
+
+    Evaluated once per distinct name rather than per row, so a filter over
+    every school costs a few dozen lookups.
+    """
+    return {n for n in names.dropna().unique() if in_group(n)}
+
+
 def sanitize_search_input(value: Optional[str], max_length: int = 100) -> Optional[str]:
     """Sanitize search input to prevent injection attacks."""
     if value is None:
@@ -744,6 +761,7 @@ async def get_schools(
     gender: Optional[str] = Query(None, description="Filter by gender (Mixed/Boys/Girls)", max_length=50),
     admissions_policy: Optional[str] = Query(None, description="Filter by admissions policy", max_length=100),
     has_sixth_form: Optional[str] = Query(None, description="Filter by sixth form presence: yes/no", max_length=3),
+    faith: Optional[str] = Query(None, description="Filter by faith group key", max_length=40),
 ):
     """
     Get list of schools with pagination.
@@ -756,6 +774,7 @@ async def get_schools(
     local_authority = sanitize_search_input(local_authority)
     school_type = sanitize_search_input(school_type)
     phase = sanitize_search_input(phase)
+    faith = sanitize_search_input(faith)
     postcode = validate_postcode(postcode)
 
     # Load the pre-computed latest-year snapshot (cached after first request / startup).
@@ -796,6 +815,22 @@ async def get_schools(
             flag = df_latest["age_range"].str.contains("18", na=False)
         df_latest = df_latest[flag if has_sixth_form == "yes" else ~flag]
 
+    # Faith group (backend/school_groups.py). A joint school is in every faith
+    # its label names; a missing religious character is "none". An unknown key
+    # matches nothing rather than being ignored, so a typo cannot show all.
+    if faith:
+        faith_key = faith.lower()
+        if faith_key in FAITH_KEYS and "religious_denomination" in df_latest.columns:
+            column = df_latest["religious_denomination"]
+            matches = column.isin(_names_in_group(column, lambda n: faith_key in faith_groups_for(n)))
+            # _names_in_group skips missing names; a missing religious
+            # character is "No religious character".
+            if faith_key == "none":
+                matches = matches | column.isna()
+            df_latest = df_latest[matches]
+        else:
+            df_latest = df_latest.iloc[0:0]
+
     # Include key result metrics for display on cards
     location_cols = ["latitude", "longitude"]
     result_cols = [
@@ -883,10 +918,17 @@ async def get_schools(
             schools_df["local_authority"].str.lower() == local_authority.lower()
         ]
 
+    # A type group key (backend/school_groups.py), or for an old link a raw
+    # GIAS type label, matched exactly as before.
     if school_type:
-        schools_df = schools_df[
-            schools_df["school_type"].str.lower() == school_type.lower()
-        ]
+        type_key = school_type.lower()
+        if type_key in TYPE_GROUP_KEYS:
+            column = schools_df["school_type"]
+            schools_df = schools_df[
+                column.isin(_names_in_group(column, lambda n: type_group_for(n) == type_key))
+            ]
+        else:
+            schools_df = schools_df[schools_df["school_type"].str.lower() == type_key]
 
     # Compute result-scoped filter values (before pagination).
     # Gender and admissions are secondary-only filters — scope them to schools
@@ -1161,6 +1203,8 @@ async def get_filter_options(request: Request):
             "local_authorities": [],
             "school_types": [],
             "years": [],
+            "school_type_groups": [],
+            "faiths": [],
         }
 
     # Phases: return values from data, ordered sensibly
@@ -1170,6 +1214,18 @@ async def get_filter_options(request: Request):
     genders = clean_filter_values(secondary_df["gender"]) if "gender" in secondary_df.columns else []
     admissions_policies = clean_filter_values(secondary_df["admissions_policy"]) if "admissions_policy" in secondary_df.columns else []
 
+    def offered(groups, present):
+        return [{"value": key, "label": label} for key, label, _ in groups if key in present]
+
+    type_groups_present = (
+        {type_group_for(n) for n in df["school_type"].dropna().unique()} - {None}
+        if "school_type" in df.columns else set()
+    )
+    faiths_present = (
+        {f for n in df["religious_denomination"].unique() for f in faith_groups_for(n)}
+        if "religious_denomination" in df.columns else set()
+    )
+
     return {
         "local_authorities": clean_filter_values(df["local_authority"]) if "local_authority" in df.columns else [],
         "school_types": clean_filter_values(df["school_type"]) if "school_type" in df.columns else [],
@@ -1177,6 +1233,8 @@ async def get_filter_options(request: Request):
         "phases": phases,
         "genders": genders,
         "admissions_policies": admissions_policies,
+        "school_type_groups": offered(TYPE_GROUPS, type_groups_present),
+        "faiths": offered(FAITH_GROUPS, faiths_present),
     }
 
 

backend/tests/test_type_and_faith_filters.py

@@ -0,0 +1,100 @@
+"""/api/schools school-type groups and faith filter, and their /api/filters lists."""
+
+import numpy as np
+import pandas as pd
+import pytest
+from fastapi.testclient import TestClient
+
+# urn -> (GIAS school type, GIAS religious character)
+SCHOOLS = {
+    100001: ("Community school", "Does not apply"),
+    100002: ("Voluntary aided school", "Roman Catholic"),
+    100003: ("Academy converter", "Roman Catholic/Church of England"),
+    100004: ("Community special school", None),
+    100005: ("Academy special converter", "Church of England"),
+    100006: ("Other independent school", "Jewish"),
+    100007: ("Miscellaneous", ""),
+}
+
+
+def _schools_df() -> pd.DataFrame:
+    base = {
+        "local_authority": "Testshire", "address": "1 Test Street", "town": "Testtown",
+        "postcode": "TS1 1AA", "age_range": "4-11", "has_sixth_form": None,
+        "gender": "Mixed", "admissions_policy": None, "ofsted_grade": np.nan,
+        "ofsted_date": None, "ofsted_framework": None, "latitude": 51.5,
+        "longitude": -0.1, "year": 202425, "total_pupils": 300,
+        "rwm_expected_pct": np.nan, "attainment_8_score": np.nan, "phase": "Primary",
+    }
+    return pd.DataFrame([
+        {**base, "urn": urn, "school_name": f"School {urn}",
+         "school_type": t, "religious_denomination": r}
+        for urn, (t, r) in SCHOOLS.items()
+    ])
+
+
+@pytest.fixture()
+def client(monkeypatch):
+    from backend import app as app_module
+
+    monkeypatch.setattr(app_module, "load_latest_school_data", _schools_df)
+    monkeypatch.setattr(app_module, "load_school_data", _schools_df)
+    return TestClient(app_module.app, raise_server_exceptions=False)
+
+
+def _urns(client, **params):
+    resp = client.get("/api/schools", params={"page_size": 50, **params})
+    assert resp.status_code == 200, resp.text
+    return sorted(s["urn"] for s in resp.json()["schools"])
+
+
+@pytest.mark.parametrize("key, urns", [
+    ("council", [100001, 100002]),
+    ("academy", [100003]),
+    ("special", [100004, 100005]),
+    ("independent", [100006]),
+    ("Special", [100004, 100005]),
+])
+def test_a_type_group_key_filters_to_its_group(client, key, urns):
+    assert _urns(client, school_type=key) == urns
+
+
+def test_a_raw_type_label_still_filters_exactly(client):
+    assert _urns(client, school_type="Community school") == [100001]
+
+
+@pytest.mark.parametrize("key, urns", [
+    ("roman_catholic", [100002, 100003]),
+    ("church_of_england", [100003, 100005]),
+    ("none", [100001, 100004, 100007]),
+    ("jewish", [100006]),
+    ("Roman_Catholic", [100002, 100003]),
+])
+def test_faith_filters_to_its_group_joint_schools_included(client, key, urns):
+    assert _urns(client, faith=key) == urns
+
+
+def test_an_unknown_faith_returns_nothing(client):
+    assert _urns(client, faith="nonsense") == []
+
+
+def test_type_and_faith_combine(client):
+    assert _urns(client, school_type="special", faith="church_of_england") == [100005]
+
+
+def test_filters_lists_only_groups_present_in_order(client):
+    body = client.get("/api/filters").json()
+    assert body["school_type_groups"] == [
+        {"value": "academy", "label": "State school: academy or free school"},
+        {"value": "council", "label": "State school: council-run"},
+        {"value": "independent", "label": "Independent (fee-paying)"},
+        {"value": "special", "label": "Special school (SEND)"},
+    ]
+    assert body["faiths"] == [
+        {"value": "none", "label": "No religious character"},
+        {"value": "church_of_england", "label": "Church of England"},
+        {"value": "roman_catholic", "label": "Roman Catholic"},
+        {"value": "jewish", "label": "Jewish"},
+    ]
+    # The raw list is still there for anything that reads it.
+    assert "Community school" in body["school_types"]