Merge pull request 'fix(pipeline): decode GIAS extracts as Windows-1252' (#182) from fix/gias-encoding into main

Reviewed-on: #182

backend/app.py

@@ -1047,6 +1047,8 @@ async def get_school_details(request: Request, urn: int):
             "age_range": latest.get("age_range", ""),
             "has_sixth_form": latest.get("has_sixth_form"),
             "nursery_provision": latest.get("nursery_provision"),
+            # The header's Selective flag reads it (lib/schoolFacts).
+            "admissions_policy": latest.get("admissions_policy"),
             "status": latest.get("status"),
             "latitude": latest.get("latitude"),
             "longitude": latest.get("longitude"),

backend/tests/test_school_page_flag_fields.py

@@ -0,0 +1,55 @@
+"""The school page's header flags (nextjs-app/lib/schoolFacts.ts) read seven
+school_info fields, so the detail payload must carry every one.
+
+It lacked admissions_policy, so no school page could flag Selective while its
+search row did: Tiffin and The Grammar School at Leeds showed the tag in
+search and nothing on their own pages.
+"""
+
+import numpy as np
+import pandas as pd
+import pytest
+from fastapi.testclient import TestClient
+
+# The fields schoolFlags() picks from School (FlagFields in lib/schoolFacts.ts).
+FLAG_FIELDS = (
+    "type_group", "admissions_policy", "gender", "religious_denomination",
+    "nursery_provision", "has_sixth_form", "phase",
+)
+
+
+def _school_df() -> pd.DataFrame:
+    return pd.DataFrame([{
+        "urn": 136910, "school_name": "Tiffin School", "phase": "Secondary",
+        "school_type": "Academy converter", "admissions_policy": "Selective",
+        "gender": "Boys", "religious_denomination": "Christian",
+        "nursery_provision": "Not applicable", "has_sixth_form": True,
+        "age_range": "11-18", "local_authority": "Kingston upon Thames",
+        "address": "Queen Elizabeth Road, Kingston upon Thames, KT2 6RL",
+        "postcode": "KT2 6RL", "latitude": 51.41, "longitude": -0.30,
+        "year": 202425, "attainment_8_score": 75.0, "total_pupils": 200,
+        "gias_total_pupils": 1478, "ofsted_grade": np.nan,
+    }])
+
+
+@pytest.fixture()
+def client(monkeypatch):
+    from backend import app as app_module
+
+    monkeypatch.setattr(app_module, "load_school_data", _school_df)
+    monkeypatch.setattr(app_module, "load_latest_school_data", _school_df)
+    monkeypatch.setattr(app_module, "get_supplementary_data", lambda db, urn: {})
+    monkeypatch.setattr(app_module, "_place_registry", None)
+    return TestClient(app_module.app, raise_server_exceptions=False)
+
+
+def test_the_school_page_carries_every_field_its_flags_read(client):
+    resp = client.get("/api/schools/136910")
+    assert resp.status_code == 200, resp.text
+    info = resp.json()["school_info"]
+    assert [f for f in FLAG_FIELDS if f not in info] == []
+
+
+def test_the_school_page_says_a_selective_school_is_selective(client):
+    info = client.get("/api/schools/136910").json()["school_info"]
+    assert info["admissions_policy"] == "Selective"

e2e/tests/journeys.spec.ts

@@ -410,6 +410,29 @@ test('search and the school page agree on how many pupils a secondary has', asyn
   expect(school.total_pupils).toBe(detail.school_info.total_pupils);
 });
 
+test('a secondary search row compares its Attainment 8 with the LA average', async ({ page }) => {
+  // The comparison vanished unnoticed: the averages were fetched with
+  // force-cache, so one stored failure hid it in that browser for good.
+  // Playwright disables the HTTP cache when it intercepts requests, so this
+  // guards the comparison itself; the unit test pins the cache mode.
+  const la = await (await page.request.get('/api/la-averages')).json();
+  const averages: Record<string, number> = la.secondary?.attainment_8_by_la ?? {};
+  const res = await page.request.get('/api/schools?search=school&phase=secondary&page_size=50');
+  expect(res.ok()).toBeTruthy();
+  const school = ((await res.json()).schools ?? []).find(
+    (s: { attainment_8_score?: number | null; local_authority?: string; school_type?: string }) =>
+      s.attainment_8_score != null && s.local_authority != null && averages[s.local_authority] != null
+      && !/special|pupil referral|alternative provision/i.test(s.school_type ?? ''));
+  test.skip(!school, 'no mainstream secondary with an LA average here');
+
+  await searchByName(page, school.school_name);
+  const link = page.locator(`a[href^="/school/${school.urn}-"]`).first();
+  await expect(link).toBeVisible({ timeout: 15_000 });
+  const stats = link.locator('xpath=ancestor::div[contains(@class, "__rowContent")][1]')
+    .locator('[class*="__line3"]');
+  await expect(stats.getByText(/vs LA avg/)).toBeVisible();
+});
+
 test('a phase outside primary/secondary filters to that phase, not to everything', async ({ page }) => {
   // The search page offers every GIAS phase, but the API only knew the grouped
   // ones and silently dropped the rest — so "Nursery" returned primaries.
@@ -3396,3 +3419,31 @@ test('a selective school is flagged Selective, on its page and in search', async
   const tags = await rowTags(page, school);
   await expect(tags.getByText('Selective', { exact: true })).toBeVisible();
 });
+
+/*
+ * The Admissions section repeated the search rows' old tag logic: "selective"
+ * matched inside "Non-selective", and "None" counted as a faith, so Burntwood
+ * read "this school has a faith-based admissions priority (None)". The
+ * Selective half stayed hidden only because the school page's API did not
+ * send admissions_policy. Data-invariant: the school comes from the API, and
+ * must have an Admissions section to read.
+ */
+test('a non-selective school with no faith makes neither claim in its Admissions section', async ({ page }) => {
+  const res = await page.request.get(
+    '/api/schools?search=school&phase=secondary&faith=none&admissions_policy=non-selective&page_size=20');
+  expect(res.ok()).toBeTruthy();
+  let urn: number | null = null;
+  for (const s of ((await res.json()).schools ?? []).slice(0, 8)) {
+    if (s.religious_denomination !== 'None') continue;
+    const detail = await (await page.request.get(`/api/schools/${s.urn}`)).json();
+    if (detail.admissions) { urn = s.urn; break; }
+  }
+  test.skip(urn == null, 'no non-selective, no-faith secondary with admissions data here');
+
+  await page.goto(`/school/${urn}`);
+  const admissions = page.locator('section#admissions');
+  await expect(admissions).toBeVisible({ timeout: 15_000 });
+  await expect(admissions.getByText('Selective:', { exact: true })).toHaveCount(0);
+  await expect(admissions.getByText('Faith priority:', { exact: true })).toHaveCount(0);
+  await expect(admissions.getByText(/\(None\)/)).toHaveCount(0);
+});

nextjs-app/__tests__/components/HomeView.staleFetch.test.tsx

@@ -1,6 +1,6 @@
 import { act, fireEvent, render, screen } from '@testing-library/react';
 import { HomeView } from '@/components/HomeView';
-import { fetchSchools } from '@/lib/api';
+import { fetchLAaverages, fetchSchools } from '@/lib/api';
 import { primaryFixture } from '../support/schoolFixtures';
 import type { SchoolsResponse, School } from '@/lib/types';
 
@@ -84,3 +84,23 @@ test('failed map requests can be retried by reopening the map', async () => {
   expect(fetchSchools).toHaveBeenCalledTimes(2);
   expect(screen.getByTestId('map')).toHaveTextContent('Retry result');
 });
+
+test('LA averages are not fetched with force-cache, so one failure is not replayed for good', async () => {
+  // force-cache serves any stored response, however old, without asking the
+  // server. A request that failed once (a staging deploy restart, the July
+  // proxy outage) was stored and replayed on every later visit, and the
+  // "vs LA avg" delta vanished from every secondary row in that browser.
+  // The default mode honours the API's Cache-Control and never reuses an
+  // error.
+  params = new URLSearchParams('search=high');
+  const secondary: SchoolsResponse = {
+    ...response('Alpha High'),
+    schools: [{ ...primaryFixture.schoolInfo, school_name: 'Alpha High', phase: 'Secondary', attainment_8_score: 50 }],
+  };
+  render(<HomeView initialSchools={secondary} filters={filters} />);
+  await act(async () => {});
+  expect(fetchLAaverages).toHaveBeenCalled();
+  for (const [options] of jest.mocked(fetchLAaverages).mock.calls) {
+    expect(options?.cache).not.toBe('force-cache');
+  }
+});

nextjs-app/__tests__/components/NearbySchoolsSection.test.tsx

@@ -8,7 +8,6 @@
 import { render, screen } from '@testing-library/react';
 
 import {
-  nearbyNoun,
   NearbySchoolsSection,
   shouldRenderNearby,
 } from '@/components/school/NearbySchoolsSection';
@@ -43,13 +42,7 @@ function school(overrides: Partial<NearbySchool> = {}): NearbySchool {
 
 function renderSection(nearby: NearbySchool[]) {
   return render(
-    <NearbySchoolsSection
-      urn={100001}
-      schoolName="Meadowbrook Primary School"
-      phase="Primary"
-      thisMetricValue={72}
-      nearby={nearby}
-    />,
+    <NearbySchoolsSection urn={100001} thisMetricValue={72} nearby={nearby} />,
   );
 }
 
@@ -90,15 +83,10 @@ describe('what the section claims', () => {
   });
 
   it('shows no chips at all when nothing is shared, rather than inventing one', () => {
-    const { container } = render(
-      <NearbySchoolsSection
-        urn={100001}
-        schoolName="Meadowbrook Primary School"
-        phase="Primary"
-        thisMetricValue={72}
-        nearby={[school({ shared: [] }), school({ urn: 100003, shared: [] })]}
-      />,
-    );
+    const { container } = renderSection([
+      school({ shared: [] }),
+      school({ urn: 100003, shared: [] }),
+    ]);
     // The card still carries its distance, name, type and figure — just no
     // claim of likeness.
     expect(container.querySelectorAll('li ul').length).toBe(0);
@@ -106,37 +94,6 @@ describe('what the section claims', () => {
   });
 });
 
-describe('what the lede calls the set', () => {
-  it.each([
-    ['Primary', 'primary schools'],
-    ['Middle deemed primary', 'primary schools'],
-    ['Secondary', 'secondary schools'],
-    ['Middle deemed secondary', 'secondary schools'],
-    ['All-through', 'all-through schools'],
-    // GIAS phase 6. Its candidates span the whole secondary group, so no
-    // single noun fits and it takes the honest general one.
-    ['16 plus', 'schools and colleges'],
-    ['', 'schools'],
-    [null, 'schools'],
-  ])('calls a %s school\'s neighbours "%s"', (phase, expected) => {
-    expect(nearbyNoun(phase)).toBe(expected);
-  });
-
-  it('never calls a sixth form college\'s neighbours primary schools', () => {
-    render(
-      <NearbySchoolsSection
-        urn={100001}
-        schoolName="Barnet Sixth Form College"
-        phase="16 plus"
-        thisMetricValue={null}
-        nearby={[school(), school({ urn: 100003 })]}
-      />,
-    );
-    expect(screen.getByText(/Other schools and colleges near Barnet Sixth Form College/)).toBeInTheDocument();
-    expect(screen.queryByText(/primary schools/)).not.toBeInTheDocument();
-  });
-});
-
 describe('cards', () => {
   it('links each school to its canonical slug', () => {
     renderSection([school(), school({ urn: 100003, school_name: 'Oakfield Primary School' })]);

nextjs-app/__tests__/components/secondaryAdmissionsTag.test.tsx

@@ -0,0 +1,58 @@
+/**
+ * The Admissions section's Selective and faith notes.
+ *
+ * It tested the policy with includes('selective'), which "Non-selective"
+ * passes, and excluded only "Does not apply" from the religious character, so
+ * Burntwood (no religious character, recorded as "None") read "this school
+ * has a faith-based admissions priority (None)". The Selective half never
+ * fired only because the school page's API did not send admissions_policy.
+ */
+
+import { screen, within } from '@testing-library/react';
+import type { School } from '@/lib/types';
+import { secondaryFixture } from '../support/schoolFixtures';
+import { renderSecondarySchoolDetail } from '../support/renderSchoolDetail';
+
+jest.mock('@/lib/analytics', () => ({
+  track: jest.fn(),
+  getNavigationSource: () => 'direct',
+}));
+jest.mock('@/components/PerformanceChart', () => ({
+  PerformanceChart: () => <div data-testid="performance-chart" />,
+}));
+jest.mock('@/components/AdmissionsTrendChart', () => ({
+  __esModule: true,
+  default: () => <div data-testid="admissions-trend-chart" />,
+}));
+jest.mock('@/components/SchoolHeroMap', () => ({
+  SchoolHeroMap: () => <div data-testid="hero-map" />,
+  __esModule: true,
+}));
+
+function admissionsOf(info: Partial<School>) {
+  renderSecondarySchoolDetail({ ...secondaryFixture, schoolInfo: { ...secondaryFixture.schoolInfo, ...info } });
+  return within(screen.getByRole('heading', { name: 'Admissions' }).closest('section')!);
+}
+
+describe('Admissions section notes', () => {
+  it('notes the entrance test for a selective school', () => {
+    const section = admissionsOf({ admissions_policy: 'Selective', religious_denomination: 'None' });
+    expect(section.getByText('Selective:')).toBeInTheDocument();
+  });
+
+  it('does not call a non-selective school selective', () => {
+    const section = admissionsOf({ admissions_policy: 'Non-selective', religious_denomination: 'None' });
+    expect(section.queryByText('Selective:')).not.toBeInTheDocument();
+  });
+
+  it.each(['None', 'Does not apply'])('claims no faith priority when the register says %p', (religious_denomination) => {
+    // Not "Non-selective": a misread Selective would win and hide this case.
+    const section = admissionsOf({ admissions_policy: 'Not applicable', religious_denomination });
+    expect(section.queryByText('Faith priority:')).not.toBeInTheDocument();
+  });
+
+  it('notes the religious character of a faith school', () => {
+    const section = admissionsOf({ admissions_policy: 'Non-selective', religious_denomination: 'Church of England' });
+    expect(section.getByText('Faith priority:')).toBeInTheDocument();
+  });
+});

nextjs-app/__tests__/lib/lastDistanceOffered.test.ts

@@ -157,6 +157,13 @@ describe('compareToCutoff', () => {
 });
 
 describe('describeCutoffAbsence', () => {
+  it('does not read "Non-selective" as selective', () => {
+    // A substring test matched "selective" inside "Non-selective".
+    const s = describeCutoffAbsence({ localAuthority: 'Wandsworth', admissionsPolicy: 'Non-selective' });
+    expect(s).not.toMatch(/entrance test/);
+    expect(s).toMatch(/^Wandsworth has not published/);
+  });
+
   it('explains a selective school by how it admits, not as missing data', () => {
     const s = describeCutoffAbsence({ localAuthority: 'Kent', admissionsPolicy: 'Selective' });
     expect(s).toContain('entrance test');

nextjs-app/__tests__/lib/schoolFacts.test.ts

@@ -59,6 +59,8 @@ describe('schoolFlags', () => {
   it('never prints Non-selective, and never a no-faith value', () => {
     expect(labels({ admissions_policy: 'Non-selective', religious_denomination: 'Does not apply' })).toEqual([]);
     expect(labels({ admissions_policy: 'Not applicable', religious_denomination: 'None' })).toEqual([]);
+    // PlaceView's rule: GIAS sometimes says "Not applicable" for no faith too.
+    expect(labels({ religious_denomination: 'Not applicable' })).toEqual([]);
   });
 
   it('prints the religious character as the register records it', () => {

nextjs-app/components/HomeView.tsx

@@ -358,10 +358,14 @@ export function HomeView({ initialSchools, filters, totalSchools, howItWorks, ed
     return () => controller.abort();
   }, [resultsView, searchParams, initialSchools.schools]);
 
-  // Fetch LA averages when secondary or mixed schools are visible
+  // Fetch LA averages when secondary or mixed schools are visible. Default
+  // cache mode, never force-cache: force-cache replays any stored response
+  // without asking the server, so one failed request (a deploy restart) hid
+  // every "vs LA avg" delta in that browser for good. The API's Cache-Control
+  // already lets the browser reuse a good answer for five minutes.
   useEffect(() => {
     if (!isSecondaryView && !isMixedView) return;
-    fetchLAaverages({ cache: 'force-cache' })
+    fetchLAaverages()
       .then(data => setLaAverages(data.secondary.attainment_8_by_la))
       .catch(() => {});
   }, [isSecondaryView, isMixedView]);

nextjs-app/components/school/NearbySchools.module.css

@@ -1,8 +1,7 @@
 .heading { font-family: var(--font-display); font-size: 1.4rem; letter-spacing: -0.4px; margin: 0; }
-.lede { margin: 0.5rem 0 1.25rem; color: var(--text-secondary); max-width: 64ch; }
 .caption { margin: 1rem 0 0; font-size: 0.72rem; color: var(--text-muted); }
 
-.top { display: flex; align-items: flex-start; justify-content: space-between; gap: 1rem; }
+.top { display: flex; align-items: center; justify-content: space-between; gap: 1rem; margin-bottom: 1.25rem; }
 .arrows { display: flex; gap: 0.5rem; flex: none; }
 .arrow { width: 44px; height: 44px; display: grid; place-items: center; cursor: pointer; border: 1px solid var(--border-strong); border-radius: 999px; background: var(--bg-card); color: var(--brand); }
 .arrow:hover:not(:disabled) { border-color: var(--brand); background: var(--brand-bg); }
@@ -15,10 +14,10 @@
 .scroller { display: grid; grid-auto-flow: column; grid-auto-columns: calc((100% - 1.8rem) / 3); gap: 0.9rem; overflow-x: auto; scroll-snap-type: x mandatory; padding: 2px; margin: -2px; list-style: none; scrollbar-width: none; -ms-overflow-style: none; }
 .scroller::-webkit-scrollbar { display: none; }
 @media (max-width: 820px) { .scroller { grid-auto-columns: calc((100% - 0.9rem) / 2); } }
-/* Touch widths (MOBILE.md): the arrows would take 96px from a 328px card and
-   crush the lede into four lines, for a control swiping already provides. They
-   go, and the documented right-edge fade carries the affordance — lifting at
-   the end of the travel, where there is nothing more to hint at. */
+/* Touch widths (MOBILE.md): the arrows would take 96px from a 328px card, for
+   a control swiping already provides. They go, and the documented right-edge
+   fade carries the affordance — lifting at the end of the travel, where there
+   is nothing more to hint at. */
 @media (max-width: 640px) {
   .top { display: block; }
   .arrows { display: none; }

nextjs-app/components/school/NearbySchoolsCarousel.tsx

@@ -4,7 +4,7 @@
  * The scroller and its arrows.
  *
  * `children` are the server-rendered cards and `header` the server-rendered
- * heading and lede: both stay server components, passed through, so this file
+ * heading: both stay server components, passed through, so this file
  * owns a DOM ref and nothing else. That is what keeps all six links in the
  * initial HTML — a carousel that mounted cards on click would put four of the
  * six beyond a crawler and beyond a reader with no JavaScript.

nextjs-app/components/school/NearbySchoolsSection.tsx

@@ -13,8 +13,10 @@
  * reached on their behalf.
  *
  * There is deliberately no "how these are chosen" panel: the method is already
- * visible in the lede, the chips and the distances. The single caption line is
- * not a method note — it is the one thing a card cannot self-correct.
+ * visible in the chips and the distances. The single caption line is not a
+ * method note — it is the one thing a card cannot self-correct.
+ *
+ * Nor is there a lede: "Other primary schools near X" only restated the heading.
  */
 
 import Link from 'next/link';
@@ -32,28 +34,6 @@ export function shouldRenderNearby(nearby?: NearbySchool[] | null): boolean {
   return (nearby?.length ?? 0) >= MINIMUM;
 }
 
-/**
- * What the lede calls the set of schools it is showing.
- *
- * Derived from the school's own GIAS phase rather than the template it renders
- * with, because those disagree for "16 plus" (GIAS phase 6): a sixth-form
- * college renders the primary template — computeSchoolFlags tests for the
- * substring "secondary" — while the backend correctly matches it against the
- * secondary group. Taking the noun from the template would print "Other primary
- * schools near <sixth form college>" above a row of secondaries.
- *
- * A 16-plus school's candidates span the whole secondary group, so no single
- * noun fits and it gets the honest general one.
- */
-export function nearbyNoun(phase: string | null | undefined): string {
-  const text = (phase ?? '').trim().toLowerCase();
-  if (text === 'all-through') return 'all-through schools';
-  if (text === '16 plus') return 'schools and colleges';
-  if (text.includes('secondary')) return 'secondary schools';
-  if (text.includes('primary')) return 'primary schools';
-  return 'schools';
-}
-
 function metricLabel(key: string): string {
   return key === 'attainment_8_score' ? 'Attainment 8' : 'Reading, writing & maths';
 }
@@ -65,25 +45,16 @@ function formatMetric(value: number | null, key: string): string {
 
 export function NearbySchoolsSection({
   urn,
-  schoolName,
-  phase,
   thisMetricValue,
   nearby,
 }: {
   urn: number;
-  schoolName: string;
-  /** The school's own GIAS phase, not the template it renders with. */
-  phase: string | null | undefined;
   thisMetricValue: number | null;
   nearby?: NearbySchool[] | null;
 }) {
   if (!shouldRenderNearby(nearby)) return null;
   const schools = nearby as NearbySchool[];
-
-  // One card matched on phase alone, so the section may not claim the set
-  // shares an intake with this school.
   const metricKey = schools[0].metric_key;
-  const noun = nearbyNoun(phase);
 
   return (
     <Section id="nearby">
@@ -91,12 +62,9 @@ export function NearbySchoolsSection({
         count={schools.length}
         labelledBy="nearby-schools-heading"
         header={
-          <div>
-            <h2 id="nearby-schools-heading" className={styles.heading}>
-              Other schools nearby
-            </h2>
-            <p className={styles.lede}>{`Other ${noun} near ${schoolName}.`}</p>
-          </div>
+          <h2 id="nearby-schools-heading" className={styles.heading}>
+            Other schools nearby
+          </h2>
         }
       >
         {schools.map((school) => (

nextjs-app/components/school/PrimarySchoolSections.tsx

@@ -154,8 +154,6 @@ export function PrimarySchoolSections({
       {/* Last: it is where the reader goes next, not part of this school. */}
       <NearbySchoolsSection
         urn={schoolInfo.urn}
-        schoolName={schoolInfo.school_name}
-        phase={schoolInfo.phase}
         thisMetricValue={flags.latestResults?.rwm_expected_pct ?? null}
         nearby={nearbySchools}
       />

nextjs-app/components/school/SecondaryAdmissionsSection.tsx

@@ -7,7 +7,7 @@
  */
 
 import type { School, SchoolAdmissions, SchoolAdmissionDistance } from '@/lib/types';
-import { formatPercentage } from '@/lib/utils';
+import { formatPercentage, hasReligiousCharacter, isSelective } from '@/lib/utils';
 import { Section, sectionStyles as styles } from './sectionShared';
 import {
   describeCutoff, describeCutoffAbsence,
@@ -33,10 +33,8 @@ export function SecondaryAdmissionsSection({
   const featureOn = admissionDistance !== undefined;
   // Moved with this section from SecondarySchoolDetailView, its only consumer.
   const admissionsTag = (() => {
-    const policy = schoolInfo.admissions_policy?.toLowerCase() ?? '';
-    if (policy.includes('selective')) return 'Selective';
-    const denom = schoolInfo.religious_denomination ?? '';
-    if (denom && denom !== 'Does not apply') return 'Faith priority';
+    if (isSelective(schoolInfo.admissions_policy)) return 'Selective';
+    if (hasReligiousCharacter(schoolInfo.religious_denomination)) return 'Faith priority';
     return null;
   })();
 

nextjs-app/components/school/SecondarySchoolSections.tsx

@@ -148,8 +148,6 @@ export function SecondarySchoolSections({
       {/* Last: it is where the reader goes next, not part of this school. */}
       <NearbySchoolsSection
         urn={schoolInfo.urn}
-        schoolName={schoolInfo.school_name}
-        phase={schoolInfo.phase}
         thisMetricValue={flags.latestResults?.attainment_8_score ?? null}
         nearby={nearbySchools}
       />

nextjs-app/components/school/lastDistanceOffered.ts

@@ -16,7 +16,7 @@
  */
 
 import type { SchoolAdmissionDistance } from '@/lib/types';
-import { formatCutoffDistance, formatMiles, formatEntryYear } from '@/lib/utils';
+import { formatCutoffDistance, formatMiles, formatEntryYear, isSelective } from '@/lib/utils';
 
 export interface CutoffDisplay {
   /** Headline figure, e.g. "0.31 miles". */
@@ -193,8 +193,7 @@ export function describeCutoffAbsence({
   admissionsPolicy,
   admissionsHistory = [],
 }: AbsenceInput): string {
-  const policy = (admissionsPolicy ?? '').toLowerCase();
-  if (policy.includes('selective')) {
+  if (isSelective(admissionsPolicy)) {
     return 'Places at this school are ranked by the entrance test rather than by '
       + 'distance, so no cut-off distance applies.';
   }

nextjs-app/lib/schoolFacts.ts

@@ -7,7 +7,7 @@
  */
 
 import type { School } from './types';
-import { hasNurseryClasses, hasReligiousCharacter, singleSexLabel } from './utils';
+import { hasNurseryClasses, hasReligiousCharacter, isSelective, singleSexLabel } from './utils';
 
 /** The search filter's type groups (backend/school_groups.py), without its
  *  parenthesised notes: fees have a flag of their own. */
@@ -54,7 +54,7 @@ export function schoolFlags(school: FlagFields): SchoolFlag[] {
   const phase = school.phase?.trim().toLowerCase();
 
   if (school.type_group === 'independent') condition('Fee-paying');
-  if (school.admissions_policy?.trim().toLowerCase() === 'selective') condition('Selective');
+  if (isSelective(school.admissions_policy)) condition('Selective');
   const singleSex = singleSexLabel(school.gender);
   if (singleSex) condition(singleSex);
   if (hasReligiousCharacter(school.religious_denomination)) {

nextjs-app/lib/utils.ts

@@ -923,12 +923,22 @@ export function isSpecialSchool(school: { school_type?: string | null }): boolea
 }
 
 /**
- * Whether GIAS records a religious character. "None" and "Does not apply" are
- * the register's two ways of saying it has none, and neither is a faith.
+ * Whether GIAS records a religious character. "None", "Does not apply" and
+ * "Not applicable" are the register's ways of saying it has none; the place
+ * table (PlaceView's NO_FAITH) reads the same three.
  */
 export function hasReligiousCharacter(value: string | null | undefined): boolean {
   const v = value?.trim().toLowerCase() ?? '';
-  return v !== '' && v !== 'none' && v !== 'does not apply';
+  return v !== '' && v !== 'none' && v !== 'does not apply' && v !== 'not applicable';
+}
+
+/**
+ * Whether GIAS records the school as selective. Exact: "Non-selective"
+ * contains "selective", so a substring test read every comprehensive as
+ * selective. (The register files a partly selective school as non-selective.)
+ */
+export function isSelective(admissionsPolicy: string | null | undefined): boolean {
+  return admissionsPolicy?.trim().toLowerCase() === 'selective';
 }
 
 /**

pipeline/dags/school_data_pipeline.py

@@ -106,9 +106,12 @@ print(f'Validation passed: {{count}} GIAS rows')
         """,
     )
 
+    # Marts fed by annual EES staging models are rebuilt by the EES DAG, even
+    # when they join dim_school. Selecting them here fails in any database
+    # where that DAG hasn't run (pipeline/tests/test_dag_selectors.py).
     dbt_build = BashOperator(
         task_id="dbt_build",
-        bash_command=f"cd {PIPELINE_DIR}/transform && {DBT_BIN} build --profiles-dir . --target production --select stg_gias_establishments+ stg_gias_links+ gias_code_names+ --exclude int_ks2_with_lineage+ int_ks4_with_lineage+",
+        bash_command=f"cd {PIPELINE_DIR}/transform && {DBT_BIN} build --profiles-dir . --target production --select stg_gias_establishments+ stg_gias_links+ gias_code_names+ --exclude int_ks2_with_lineage+ int_ks4_with_lineage+ stg_ees_ks4_destinations+ stg_ees_ks5_destinations+",
     )
 
     sync_typesense = BashOperator(
@@ -143,7 +146,7 @@ with DAG(
 
     dbt_build_ofsted = BashOperator(
         task_id="dbt_build",
-        bash_command=f"cd {PIPELINE_DIR}/transform && {DBT_BIN} build --profiles-dir . --target production --select stg_ofsted_inspections+ int_ofsted_latest+ fact_ofsted_inspection+ dim_school+",
+        bash_command=f"cd {PIPELINE_DIR}/transform && {DBT_BIN} build --profiles-dir . --target production --select stg_ofsted_inspections+ int_ofsted_latest+ fact_ofsted_inspection+ dim_school+ --exclude stg_ees_ks4_destinations+ stg_ees_ks5_destinations+",
     )
 
     sync_typesense_ofsted = BashOperator(

pipeline/plugins/extractors/tap-uk-gias/tap_uk_gias/gias_csv.py

@@ -0,0 +1,26 @@
+"""Read a GIAS extract from the raw bytes of the download.
+
+GIAS writes its CSVs in Windows-1252 and sends no charset, so `resp.text`
+leaves requests to guess the codec. On 3 Oct 2026 it guessed windows-1250 and
+"à" became "ŕ". Decode the bytes ourselves instead.
+"""
+
+from __future__ import annotations
+
+import io
+
+import pandas as pd
+
+GIAS_ENCODING = "cp1252"
+
+
+def read_gias_csv(content: bytes, logger=None) -> pd.DataFrame:
+    """Every column as a string; a blank cell stays ''."""
+    # Windows-1252 leaves five bytes undefined. One stray byte must not stop
+    # the daily refresh of every school, so it becomes U+FFFD and is logged.
+    text = content.decode(GIAS_ENCODING, errors="replace")
+    undecodable = text.count("�")
+    if undecodable and logger is not None:
+        logger.warning("%d byte(s) in the GIAS extract could not be decoded as %s",
+                       undecodable, GIAS_ENCODING)
+    return pd.read_csv(io.StringIO(text), dtype=str, keep_default_na=False)

pipeline/plugins/extractors/tap-uk-gias/tap_uk_gias/tap.py

@@ -7,6 +7,8 @@ from datetime import date, timedelta
 from singer_sdk import Stream, Tap
 from singer_sdk import typing as th
 
+from tap_uk_gias.gias_csv import read_gias_csv
+
 GIAS_URL_TEMPLATE = (
     "https://ea-edubase-api-prod.azurewebsites.net"
     "/edubase/downloads/public/edubasealldata{date}.csv"
@@ -74,9 +76,6 @@ class GIASEstablishmentsStream(Stream):
 
     def get_records(self, context):
         """Download GIAS CSV and yield rows."""
-        import io
-
-        import pandas as pd
         import requests
 
         today = date.today()
@@ -94,12 +93,7 @@ class GIASEstablishmentsStream(Stream):
 
         resp.raise_for_status()
 
-        df = pd.read_csv(
-            io.StringIO(resp.text),
-            encoding="latin-1",
-            dtype=str,
-            keep_default_na=False,
-        )
+        df = read_gias_csv(resp.content, self.logger)
 
         for _, row in df.iterrows():
             record = row.to_dict()
@@ -126,9 +120,6 @@ class GIASLinksStream(Stream):
 
     def get_records(self, context):
         """Download GIAS links CSV and yield rows."""
-        import io
-
-        import pandas as pd
         import requests
 
         today = date.today()
@@ -146,12 +137,7 @@ class GIASLinksStream(Stream):
 
         resp.raise_for_status()
 
-        df = pd.read_csv(
-            io.StringIO(resp.text),
-            encoding="latin-1",
-            dtype=str,
-            keep_default_na=False,
-        )
+        df = read_gias_csv(resp.content, self.logger)
 
         for _, row in df.iterrows():
             record = row.to_dict()

pipeline/tests/test_dag_selectors.py

@@ -0,0 +1,98 @@
+"""Every scheduled dbt build must only build models whose parents exist.
+
+The daily GIAS build selects `stg_gias_establishments+`, so any mart that joins
+dim_school joins the daily build too. When such a mart also reads a staging
+model that only the manually triggered EES DAG builds, the daily build fails in
+any database where that DAG has not run since. Sync and cache invalidation then
+never run either. The destinations marts did this from late August 2026.
+
+The graph is read from the model SQL, because CI has no dbt.
+"""
+import re
+from collections import defaultdict
+from pathlib import Path
+
+import pytest
+
+PIPELINE = Path(__file__).resolve().parents[1]
+MODELS = PIPELINE / 'transform' / 'models'
+DAG_FILE = PIPELINE / 'dags' / 'school_data_pipeline.py'
+
+REF = re.compile(r"ref\(\s*'([a-z0-9_]+)'\s*\)")
+DBT_BUILD = re.compile(r'dbt_build\w*\s*=\s*BashOperator\(.*?build --profiles-dir \. --target production ([^"]+)"', re.S)
+DAG_ID = re.compile(r'dag_id="([a-z0-9_]+)"')
+
+DAILY = 'school_data_daily'
+
+# dim_school reads int_ofsted_latest only when the relation exists
+# (adapter.get_relation), so a missing table is not a failure.
+OPTIONAL_PARENTS = {'int_ofsted_latest'}
+
+
+def model_parents():
+    """{model: models it refs}. Seeds are left out: they are loaded once and always exist."""
+    sql = {p.stem: p.read_text() for p in MODELS.rglob('*.sql')}
+    return {name: set(REF.findall(text)) & set(sql) for name, text in sql.items()}
+
+
+def downstream(node, children):
+    seen, stack = {node}, [node]
+    while stack:
+        for child in children[stack.pop()]:
+            if child not in seen:
+                seen.add(child)
+                stack.append(child)
+    return seen
+
+
+def expand(tokens, children):
+    out = set()
+    for token in tokens:
+        out |= downstream(token[:-1], children) if token.endswith('+') else {token}
+    return out
+
+
+def scheduled_builds():
+    """{dag_id: dbt selection arguments} for every dbt build in the DAG file."""
+    text = DAG_FILE.read_text()
+    starts = [(m.start(), m.group(1)) for m in DAG_ID.finditer(text)]
+    builds = {}
+    for i, (start, dag_id) in enumerate(starts):
+        end = starts[i + 1][0] if i + 1 < len(starts) else len(text)
+        found = DBT_BUILD.search(text, start, end)
+        if found:
+            builds[dag_id] = found.group(1)
+    return builds
+
+
+def selected_models(args, parents):
+    children = defaultdict(set)
+    for model, ps in parents.items():
+        for p in ps:
+            children[p].add(model)
+    select = re.search(r'--select (.+?)(?= --exclude|$)', args).group(1).split()
+    excluded = re.search(r'--exclude (.+)$', args)
+    exclude = excluded.group(1).split() if excluded else []
+    return (expand(select, children) - expand(exclude, children)) & set(parents)
+
+
+PARENTS = model_parents()
+BUILDS = scheduled_builds()
+DAILY_MODELS = selected_models(BUILDS[DAILY], PARENTS)
+
+
+def test_every_dag_with_a_dbt_build_is_parsed():
+    assert set(BUILDS) == {
+        'school_data_daily', 'school_data_monthly_ofsted', 'school_data_annual_ees',
+        'school_data_annual_idaci', 'school_data_annual_distance',
+    }
+
+
+@pytest.mark.parametrize('dag_id', sorted(BUILDS))
+def test_selected_models_only_read_models_that_exist(dag_id):
+    selected = selected_models(BUILDS[dag_id], PARENTS)
+    # The daily build is the base layer: other DAGs may rely on what it builds.
+    available = selected | OPTIONAL_PARENTS | (DAILY_MODELS if dag_id != DAILY else set())
+    missing = {model: sorted(PARENTS[model] - available) for model in sorted(selected)
+               if PARENTS[model] - available}
+    assert missing == {}, f'{dag_id} builds models whose parents it never builds: {missing}'

pipeline/tests/test_gias_encoding.py

@@ -0,0 +1,66 @@
+"""GIAS publishes its extracts in Windows-1252 and declares no charset.
+
+The tap used to hand pandas `resp.text`, so requests guessed the codec.
+On 3 Oct 2026 it guessed windows-1250, and "St Thomas à Becket" was stored
+as "St Thomas ŕ Becket". The `encoding=` passed to read_csv did nothing,
+because the text was already decoded.
+"""
+import importlib.util
+import logging
+from pathlib import Path
+
+import pytest
+
+MODULE = (Path(__file__).resolve().parents[1] / 'plugins' / 'extractors' / 'tap-uk-gias'
+          / 'tap_uk_gias' / 'gias_csv.py')
+
+
+@pytest.fixture
+def gias_csv():
+    spec = importlib.util.spec_from_file_location('gias_csv', MODULE)
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+    return module
+
+
+# Byte for byte as GIAS writes it: 0xE0 à, 0x92 ’, 0xE9 é, 0xB0 °, 0xE7 ç.
+EXTRACT = (
+    b'"URN","EstablishmentName","HeadLastName"\r\n'
+    b'"138950","St Thomas \xe0 Becket Catholic Secondary School","Smith"\r\n'
+    b'"100000","The Dean and Chapter of St Paul\x92s Cathedral","Pr\xe9vert"\r\n'
+    b'"140677","North Star 180\xb0","Fran\xe7ois"\r\n'
+    b'"100001","No head recorded",""\r\n'
+)
+
+
+def test_names_decode_as_windows_1252(gias_csv):
+    df = gias_csv.read_gias_csv(EXTRACT)
+    assert list(df['EstablishmentName']) == [
+        'St Thomas à Becket Catholic Secondary School',
+        'The Dean and Chapter of St Paul’s Cathedral',
+        'North Star 180°',
+        'No head recorded',
+    ]
+    assert list(df['HeadLastName']) == ['Smith', 'Prévert', 'François', '']
+
+
+def test_the_codec_requests_guessed_is_not_used(gias_csv):
+    # What the tap stored on 3 Oct: the same bytes read as windows-1250.
+    assert 'ŕ' in EXTRACT.decode('cp1250')
+    names = ' '.join(gias_csv.read_gias_csv(EXTRACT)['EstablishmentName'])
+    assert 'ŕ' not in names
+
+
+def test_values_stay_strings(gias_csv):
+    df = gias_csv.read_gias_csv(EXTRACT)
+    assert df.loc[0, 'URN'] == '138950'
+
+
+def test_a_byte_windows_1252_leaves_undefined_does_not_stop_the_load(gias_csv, caplog):
+    # 0x81 has no Windows-1252 character. One odd name must not block the daily
+    # refresh of every school, but it must be visible in the log.
+    extract = b'"URN","EstablishmentName"\r\n"100002","Odd \x81 Name"\r\n'
+    with caplog.at_level(logging.WARNING):
+        df = gias_csv.read_gias_csv(extract, logger=logging.getLogger('gias'))
+    assert df.loc[0, 'EstablishmentName'] == 'Odd � Name'
+    assert 'could not be decoded' in caplog.text