Compare commits

..
Author SHA1 Message Date
Tudor 3c69b65ff0 fix(places): phase-grouped tables, plain-English measures, styled links
Three presentation faults on the place pages, all found by looking at a
rendered page rather than at a test.

An unphased place page showed one primary-only measure for a list holding both
phases: 8 of 27 rows on /schools/brentwood were blank, because secondaries
have no reading-writing-maths score. Picking the other measure would only have
inverted which rows were empty, and putting both in one column would have
mixed a percentage with a 0-90 score. Each phase now gets its own table, so a
blank cell means the school genuinely has no published result — which is worth
saying, and now says "Not published" rather than a bare dash.

"RWM expected" was invented here. The site already names the measure in
METRIC_DEFINITIONS, surfaced at /api/metrics: "Reading, Writing & Maths
Combined %". The heading now reads "Reading, writing & maths" with the full
definition in the tooltip.

Links carried no class at all, so they rendered as default blue underlined
browser links beside a site that styles table links as body colour with a
brand hover. They now follow RankingsView's convention, and running-copy links
take the brand colour.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015mWQnpye9F299NVRCCSRvj
2026-08-21 21:49:39 +01:00
8 changed files with 30 additions and 404 deletions

No files matched your search

-7
View File
@@ -1232,13 +1232,6 @@ async def get_place(request: Request, kind: str, slug: str,
"place": {"kind": place.kind, "slug": place.slug, "name": place.name,
"count": len(place.urns),
"parent_authority": place.parent_authority,
# Every authority the place meaningfully sits in. SW19 is
# mostly Merton but partly Wandsworth; naming one asserts
# something false.
"authorities": [
{"name": name, "slug": _slugify(name), "count": n}
for name, n in place.authorities
],
# Only phases that clear the threshold, so the page links
# variants that exist rather than 404s.
"phases": [ph for ph in ("primary", "secondary")
+18 -89
View File
@@ -30,12 +30,6 @@ class Place:
name: str
urns: tuple[int, ...]
parent_authority: str | None # authority NAME, for the 301 target
# Every authority the place meaningfully sits in, largest first. A quarter
# of outcodes and a third of towns straddle a boundary — SW19 is mostly
# Merton but partly Wandsworth — so naming only one asserts something
# false. parent_authority stays single because a redirect needs one
# target; this is what the page shows.
authorities: tuple[tuple[str, int], ...] = ()
# URNs per phase, so the per-phase threshold can be applied without
# re-querying. A place with 30 primaries and 2 secondaries publishes a
# primary variant and no secondary one.
@@ -77,100 +71,39 @@ def _phase_urns(group, publishable: set[int]) -> dict[str, tuple[int, ...]]:
return out
# A place is described by an authority when it holds at least a tenth of the
# schools, and at least two. GIAS carries occasional postcode errors — EN6
# lists two Shropshire schools among fourteen in Hertfordshire — and a bare
# "any authority present" rule would print those as though they were real.
# There is deliberately no cap on how many are named. An earlier cut stopped
# at three, which silently dropped the fourth in exactly the case where the
# information matters most — a genuinely fragmented place. The share rule is
# the only limit, and it already bounds the list at ten.
_AUTHORITY_MIN_SHARE = 0.10
_AUTHORITY_MIN_SCHOOLS = 2
def _parent_authority(group) -> str | None:
"""The most common authority in a group — the useful 301 target.
def _authorities(group) -> tuple[tuple[str, int], ...]:
"""Authorities this place meaningfully sits in, largest first."""
from backend.app import EXCLUDED_FILTER_VALUES
if "local_authority" not in group.columns:
return ()
counts = group["local_authority"].dropna().value_counts()
total = int(counts.sum())
if not total:
return ()
kept = [
(str(name), int(n)) for name, n in counts.items()
if str(name) not in EXCLUDED_FILTER_VALUES
and n >= _AUTHORITY_MIN_SCHOOLS
and n / total >= _AUTHORITY_MIN_SHARE
]
# A place too small or too fragmented for the share rule still names its
# largest authority, or the page would say nothing about where it is.
if not kept:
for name, n in counts.items():
if str(name) not in EXCLUDED_FILTER_VALUES:
return ((str(name), int(n)),)
return ()
return tuple(kept)
def _parent_authority(authorities: tuple[tuple[str, int], ...]) -> str | None:
"""The 301 target: the largest authority a place sits in.
Derived from `authorities` rather than computed separately. The first cut
used `mode()` here while `authorities` used `value_counts()`, and on an
exact tie pandas does not guarantee the two pick the same name — so the
redirect could have pointed somewhere other than the authority the page
named first. One computation, one answer.
Deriving it also inherits the sentinel filter, so a place can no longer
redirect to /schools/authority/does-not-apply.
A town spanning several authorities has no single parent, so the mode is
the honest answer rather than an arbitrary first row.
"""
return authorities[0][0] if authorities else None
if "local_authority" not in group.columns:
return None
top = group["local_authority"].dropna()
return str(top.mode().iloc[0]) if not top.empty else None
def _group(df, column: str, kind: str, publishable: set[int]) -> dict[str, Place]:
"""One Place per distinct SLUG in `column` that clears the threshold.
Grouped by slug, not by raw value, because GIAS spells the same place
several ways and they all resolve to one URL. Five town slugs come from
more than one spelling: "London" (1,819 schools) and "LONDON" (12) both
slugify to `london`; Weston-super-Mare is split 14/19 across two
spellings; Newcastle-under-Lyme across three.
Grouping by raw value meant the later group simply overwrote the earlier
one in this dict — so /schools/london could have shown twelve schools
instead of 1,819, silently and depending on row order.
The display name is the most common spelling, which is the one a reader
expects to see.
"""
"""One Place per distinct value of `column` that clears the threshold."""
from backend.app import _slugify
if column not in df.columns:
return {}
working = df.assign(_slug=df[column].map(
lambda v: _slugify(str(v).strip()) if isinstance(v, str) and v.strip() else None))
working = working[working["_slug"].notna() & (working["_slug"] != "")]
out: dict[str, Place] = {}
for slug, group in working.groupby("_slug"):
slug = str(slug)
for name, group in df.groupby(column, dropna=True):
name = str(name).strip()
if not name:
continue
urns = tuple(sorted({int(u) for u in group["urn"]} & publishable))
if len(urns) < MIN_SCHOOLS:
continue
spellings = group[column].dropna().value_counts()
if spellings.empty:
slug = _slugify(name)
if not slug:
continue
name = str(spellings.index[0]).strip()
authorities = () if kind == "authority" else _authorities(group)
place = Place(
kind=kind, slug=slug, name=name, urns=urns,
parent_authority=_parent_authority(authorities),
authorities=authorities,
parent_authority=_parent_authority(group) if kind == "town" else None,
phase_urns=_phase_urns(group, publishable),
)
out[place.key] = place
@@ -203,10 +136,8 @@ def _outcode_places(df, publishable: set[int]) -> dict[str, Place]:
urns = tuple(sorted({int(u) for u in group["urn"]} & publishable))
if len(urns) < MIN_SCHOOLS:
continue
authorities = _authorities(group)
place = Place(kind="outcode", slug=str(oc).lower(), name=str(oc),
urns=urns, parent_authority=_parent_authority(authorities),
authorities=authorities,
urns=urns, parent_authority=_parent_authority(group),
phase_urns=_phase_urns(group, publishable))
out[place.key] = place
return out
@@ -248,10 +179,8 @@ def _locality_places(df, publishable: set[int],
"threshold of %d - not published",
slug, ", ".join(outcodes), len(urns), MIN_SCHOOLS)
continue
authorities = _authorities(group)
place = Place(kind="locality", slug=slug, name=name, urns=urns,
parent_authority=_parent_authority(authorities),
authorities=authorities,
parent_authority=_parent_authority(group),
phase_urns=_phase_urns(group, publishable))
out[place.key] = place
return out
-114
View File
@@ -229,117 +229,3 @@ def test_no_curated_locality_names_a_london_borough():
f"these are boroughs, not districts: {sorted(named)} - they already "
"have an authority page covering every school"
)
def test_a_place_names_every_authority_it_straddles():
"""SW19 is mostly Merton but partly Wandsworth.
A quarter of viable outcodes and a third of viable towns cross an
authority boundary, so naming only the largest asserts something false.
"""
rows = (_town(26, "London", "Merton", start=300000)
+ _town(7, "London", "Wandsworth", start=400000))
for r in rows:
r["postcode"] = "SW19 1AA"
reg = build_place_registry(_df(rows))
names = [n for n, _ in reg["outcode:sw19"].authorities]
assert names == ["Merton", "Wandsworth"] # largest first
assert dict(reg["outcode:sw19"].authorities)["Wandsworth"] == 7
def test_the_redirect_target_stays_a_single_authority():
# parent_authority and authorities do different jobs: a 301 needs one
# target, the page needs the truth.
rows = (_town(26, "London", "Merton", start=300000)
+ _town(7, "London", "Wandsworth", start=400000))
for r in rows:
r["postcode"] = "SW19 1AA"
reg = build_place_registry(_df(rows))
assert reg["outcode:sw19"].parent_authority == "Merton"
def test_a_stray_authority_below_the_share_threshold_is_not_named():
# GIAS carries postcode errors — EN6 lists two Shropshire schools among
# fourteen in Hertfordshire. Printing those as though real would be worse
# than omitting them.
rows = (_town(30, "Barnet", "Hertfordshire", start=300000)
+ _town(1, "Barnet", "Shropshire", start=400000))
for r in rows:
r["postcode"] = "EN6 1AA"
reg = build_place_registry(_df(rows))
assert [n for n, _ in reg["outcode:en6"].authorities] == ["Hertfordshire"]
def test_a_sentinel_authority_is_never_named():
rows = (_town(20, "London", "Merton", start=300000)
+ _town(6, "London", "Does not apply", start=400000))
for r in rows:
r["postcode"] = "SW19 1AA"
reg = build_place_registry(_df(rows))
assert [n for n, _ in reg["outcode:sw19"].authorities] == ["Merton"]
def test_a_place_always_names_at_least_one_authority():
# Even when every authority is below the share threshold, the page has to
# say where the place is.
rows = []
for i, la in enumerate(["A", "B", "C", "D", "E", "F", "G"]):
rows += _town(1, "Fragmented", la, start=300000 + i * 100)
reg = build_place_registry(_df(rows))
place = reg.get("town:fragmented")
assert place is not None
assert len(place.authorities) == 1
def test_every_qualifying_authority_is_named_with_no_cap():
"""An earlier cut stopped at three, dropping the fourth silently.
That truncation bit exactly where the information matters most — a
genuinely fragmented place — and nothing recorded it.
"""
rows = []
for i, la in enumerate(["Hackney", "Lambeth", "Westminster", "Lewisham"]):
rows += _town(3, "Fourway", la, start=300000 + i * 100)
reg = build_place_registry(_df(rows))
assert len(reg["town:fourway"].authorities) == 4
def test_the_redirect_target_is_the_authority_named_first():
"""They were computed separately — mode() against value_counts() — and on
an exact tie pandas does not guarantee the two agree."""
rows = (_town(26, "London", "Merton", start=300000)
+ _town(7, "London", "Wandsworth", start=400000))
for r in rows:
r["postcode"] = "SW19 1AA"
place = build_place_registry(_df(rows))["outcode:sw19"]
assert place.parent_authority == place.authorities[0][0]
def test_a_place_never_redirects_to_a_sentinel_authority():
# Deriving the parent from `authorities` inherits its sentinel filter.
rows = (_town(6, "Someplace", "Does not apply", start=300000)
+ _town(5, "Someplace", "Essex", start=400000))
reg = build_place_registry(_df(rows))
assert reg["town:someplace"].parent_authority == "Essex"
def test_spellings_of_one_place_are_merged_not_overwritten():
"""GIAS spells the same place several ways, and they share a URL.
"London" (1,819 schools) and "LONDON" (12) both slugify to `london`.
Grouping by raw value let the later group overwrite the earlier one, so
the page could have shown twelve schools instead of 1,819 — silently, and
depending on row order.
"""
rows = (_town(6, "Weston-super-Mare", "North Somerset", start=300000)
+ _town(5, "Weston-Super-Mare", "North Somerset", start=400000))
reg = build_place_registry(_df(rows))
assert len(reg["town:weston-super-mare"].urns) == 11
def test_the_merged_place_takes_its_most_common_spelling():
rows = (_town(9, "Newcastle-under-Lyme", "Staffordshire", start=300000)
+ _town(5, "NEWCASTLE-UNDER-LYME", "Staffordshire", start=400000))
reg = build_place_registry(_df(rows))
assert reg["town:newcastle-under-lyme"].name == "Newcastle-under-Lyme"
+3 -69
View File
@@ -1649,29 +1649,13 @@ const CANONICAL_ROUTES: Array<[string, string]> = [
['/admissions', 'https://www.schoolcompare.co.uk/admissions'],
];
/**
* Next normalises canonical URLs against `trailingSlash: false`, so the root
* ships as `https://www.schoolcompare.co.uk` with no slash while every other
* route keeps its path. Both forms address the same document, and which one
* Next emits is its business, not something worth pinning a test to.
*
* The first cut hardcoded the slash and failed only on the homepage — the
* same gap as the doubled brand: it asserted the metadata object rather than
* what the page actually renders.
*/
function sameUrl(a: string | null, b: string): boolean {
const strip = (u: string) => u.replace(/\/+$/, '');
return strip(a ?? '') === strip(b);
}
for (const [path, expected] of CANONICAL_ROUTES) {
test(`${path} declares exactly one canonical, on the www host`, async ({ page }) => {
await page.goto(path);
const hrefs = await page.locator('link[rel="canonical"]').evaluateAll(
(els) => els.map((e) => e.getAttribute('href')));
expect(hrefs, `${path} should declare one canonical`).toHaveLength(1);
expect(sameUrl(hrefs[0], expected),
`${path} canonical was ${hrefs[0]}, expected ${expected}`).toBe(true);
expect(hrefs[0]).toBe(expected);
});
}
@@ -1679,8 +1663,7 @@ test('a filtered homepage still canonicalises to the bare root', async ({ page }
await page.goto('/?search=primary&phase=primary&sort=name&page=2');
const href = await page.locator('link[rel="canonical"]').first()
.getAttribute('href');
expect(sameUrl(href, 'https://www.schoolcompare.co.uk/'),
`filtered homepage canonical was ${href}`).toBe(true);
expect(href).toBe('https://www.schoolcompare.co.uk/');
});
test('a school page canonicalises to its own slug on the www host', async ({ page }) => {
@@ -1731,33 +1714,10 @@ test('staging answers noindex, and stays crawlable so the noindex is seen', asyn
// The other half, and the reason this is one test rather than two: a
// Disallow would stop Google fetching the page at all, so it would never
// see the noindex above. The two only work together.
//
// Scoped to the `*` group. The first cut matched `Disallow: /` anywhere in
// the file and tripped over the AI-crawler groups Cloudflare injects —
// ClaudeBot, GPTBot, Amazonbot and friends all carry a blanket disallow,
// deliberately, and none of them is Googlebot.
const robots = await (await page.request.get('/robots.txt')).text();
expect(blocksEverything(robots, '*'),
'the * group must not disallow the whole site, or the noindex is never seen')
.toBe(false);
expect(robots).not.toMatch(/^\s*Disallow:\s*\/\s*$/mi);
});
/** True when `agent`'s group in a robots.txt disallows the entire site. */
function blocksEverything(robots: string, agent: string): boolean {
let current: string | null = null;
let blocked = false;
for (const raw of robots.split('\n')) {
const line = raw.split('#')[0].trim();
if (!line) continue;
const [key, ...rest] = line.split(':');
const value = rest.join(':').trim();
const k = key.trim().toLowerCase();
if (k === 'user-agent') current = value;
else if (current === agent && k === 'disallow' && value === '/') blocked = true;
}
return blocked;
}
test('a school page on staging is noindexed too, not just the homepage', async ({ page }) => {
const list = await page.request.get('/api/schools?search=primary&per_page=1');
const [first] = (await list.json()).schools ?? [];
@@ -1925,29 +1885,3 @@ test('no page title repeats the brand', async ({ page }) => {
expect(brands, `${path} repeats the brand: ${title}`).toBeLessThanOrEqual(1);
}
});
test('a place straddling a boundary names every authority it sits in', async ({ page }) => {
// A quarter of outcodes and a third of towns cross an authority boundary —
// SW19 is mostly Merton but partly Wandsworth. Naming only the largest
// asserts something false about the place.
const { places } = await (await page.request.get('/api/places')).json();
const outcode = places.find((p: { kind: string }) => p.kind === 'outcode');
expect(outcode).toBeTruthy();
// Find any place the registry reports as straddling.
let straddling: { kind: string; slug: string } | null = null;
for (const p of places.filter((p: { kind: string }) => p.kind === 'outcode').slice(0, 40)) {
const d = await (await page.request.get(`/api/places/outcode/${p.slug}`)).json();
if ((d.place.authorities ?? []).length > 1) { straddling = p; break; }
}
test.skip(!straddling, 'no straddling outcode found in the sample');
const detail = await (await page.request.get(
`/api/places/outcode/${straddling!.slug}`)).json();
await page.goto(`/schools/near/${straddling!.slug}`);
for (const a of detail.place.authorities) {
await expect(page.locator(`a[href="/schools/authority/${a.slug}"]`).first())
.toBeVisible();
}
});
@@ -168,78 +168,3 @@ describe('PlaceView presentation', () => {
.not.toBeInTheDocument();
});
});
describe('PlaceView table alignment', () => {
const aligned: PlaceDetail = {
place: { kind: 'town', slug: 'brentwood', name: 'Brentwood', count: 2,
parent_authority: 'Essex', phases: ['primary'] },
schools: [
{ urn: 1, school_name: 'Alpha Primary', phase: 'Primary',
rwm_expected_pct: 82, attainment_8_score: null } as never,
],
averages: { rwm_expected_pct: 63, attainment_8_score: null },
};
it('aligns the measure heading and its values with the same class', () => {
// They were aligned by two different selectors whose specificity did not
// match: `.table th:last-child` (0,2,1) won and went right, while `.num`
// (0,1,0) lost to `.table td` (0,1,1) and stayed left. Sharing one class
// is what makes them impossible to drift apart.
const { container } = render(<PlaceView detail={aligned} englandAverage={61}
neighbours={[]} />);
const th = container.querySelectorAll('th')[1];
const td = container.querySelectorAll('tbody td')[1];
expect(th.className).toBeTruthy();
expect(td.className).toBe(th.className);
});
it('leaves the school-name column unclassed so it takes the spare width', () => {
const { container } = render(<PlaceView detail={aligned} englandAverage={61}
neighbours={[]} />);
expect(container.querySelectorAll('th')[0].className).toBe('');
});
});
describe('PlaceView authorities', () => {
const straddling: PlaceDetail = {
place: { kind: 'outcode', slug: 'sw19', name: 'SW19', count: 33,
parent_authority: 'Merton', phases: ['primary'],
authorities: [
{ name: 'Merton', slug: 'merton', count: 26 },
{ name: 'Wandsworth', slug: 'wandsworth', count: 7 },
] },
schools: [
{ urn: 1, school_name: 'Alpha Primary', phase: 'Primary',
rwm_expected_pct: 82, attainment_8_score: null } as never,
],
averages: { rwm_expected_pct: 63, attainment_8_score: null },
};
it('names every authority the place straddles, not just the largest', () => {
// SW19 is mostly Merton but partly Wandsworth. Naming one asserts
// something false about a quarter of outcodes.
render(<PlaceView detail={straddling} englandAverage={61} neighbours={[]} />);
expect(screen.getByRole('link', { name: 'Merton' }))
.toHaveAttribute('href', '/schools/authority/merton');
expect(screen.getByRole('link', { name: 'Wandsworth' }))
.toHaveAttribute('href', '/schools/authority/wandsworth');
});
it('joins them readably rather than as a bare list', () => {
// Asserted on the summary line's whole text: a loose /and/ matcher also
// hits "Wandsworth".
const { container } = render(<PlaceView detail={straddling}
englandAverage={61} neighbours={[]} />);
const summary = container.querySelector('header p');
expect(summary?.textContent).toContain('Merton and Wandsworth');
});
it('falls back to the single parent when the field is absent', () => {
// A cached API response predating the authorities field must not blank
// the line entirely.
const legacy = { ...straddling,
place: { ...straddling.place, authorities: undefined } };
render(<PlaceView detail={legacy} englandAverage={61} neighbours={[]} />);
expect(screen.getByRole('link', { name: 'Merton' })).toBeInTheDocument();
});
});
@@ -144,24 +144,10 @@
border-bottom: none;
}
/*
* Header and value share one class and one rule, so they cannot drift apart.
*
* The first cut aligned them with two different selectors: `.table th:last-child`
* at (0,2,1) beat the element rule and went right, while `.num` at (0,1,0) lost
* to `.table td` at (0,1,1) and stayed left. The heading and its numbers sat on
* opposite edges of the column.
*
* width:1% with nowrap makes the measure column hug its content so the school
* name takes the remaining width — without it the two columns split evenly and
* the gap between heading and value reads as misalignment on a wide screen.
*/
.table th.num,
.table td.num {
.table th:last-child,
.num {
text-align: right;
font-variant-numeric: tabular-nums;
width: 1%;
white-space: nowrap;
}
/* The measure is spelled out; the tooltip carries the definition. */
+6 -22
View File
@@ -71,9 +71,7 @@ function SchoolTable({ schools, phase }: { schools: School[]; phase: PhaseKey })
<thead>
<tr>
<th scope="col">School</th>
{/* Same class as the value cell below: one rule aligns both, so
they cannot drift apart. */}
<th scope="col" className={styles.num}>
<th scope="col">
<abbr className={styles.metricHead} title={metric.hint}>
{metric.heading}
</abbr>
@@ -106,13 +104,6 @@ function SchoolTable({ schools, phase }: { schools: School[]; phase: PhaseKey })
export function PlaceView({ detail, phase, englandAverage, neighbours }: Props) {
const { place, schools, averages } = detail;
// Fall back to the single parent when the API predates the authorities
// field, so a stale cache never blanks the line entirely.
const authorities = place.authorities?.length
? place.authorities
: place.parent_authority
? [{ name: place.parent_authority, slug: authoritySlug(place.parent_authority), count: 0 }]
: [];
const local = averages[METRICS[phase ?? 'primary'].key];
const phaseWord = phase === 'secondary' ? 'Secondary schools'
: phase === 'primary' ? 'Primary schools' : 'Schools';
@@ -170,20 +161,13 @@ export function PlaceView({ detail, phase, englandAverage, neighbours }: Props)
<h1>{phaseWord} in {place.name}</h1>
<p className={styles.summary}>
{place.count} schools
{authorities.length > 0 && (
{place.parent_authority && (
<>
{' · '}
{/* Every authority, not just the largest. A quarter of outcodes
and a third of towns cross a boundary: SW19 is mostly Merton
but partly Wandsworth, and naming one asserts otherwise. */}
{authorities.map((a, i) => (
<span key={a.slug}>
{i > 0 && (i === authorities.length - 1 ? ' and ' : ', ')}
<Link href={`/schools/authority/${a.slug}`} className={styles.inlineLink}>
{a.name}
</Link>
</span>
))}
<Link href={`/schools/authority/${authoritySlug(place.parent_authority)}`}
className={styles.inlineLink}>
{place.parent_authority}
</Link>
</>
)}
</p>
+1 -12
View File
@@ -18,19 +18,8 @@ export interface PlaceSummary {
phases?: string[];
}
export interface PlaceAuthority {
name: string;
slug: string;
count: number;
}
export interface PlaceDetail {
place: PlaceSummary & {
parent_authority: string | null;
/** Every authority the place meaningfully sits in, largest first. SW19 is
* mostly Merton but partly Wandsworth. */
authorities?: PlaceAuthority[];
};
place: PlaceSummary & { parent_authority: string | null };
schools: School[];
averages: {
rwm_expected_pct: number | null;