MOB-08: Search list pagination is already implemented (page_size 50,
"Load more schools" button + count) — no change needed; ticket closed.
MOB-10: Rather than build a full bottom-sheet filter modal (large
change, modal/focus-trap/scroll-lock infra), promote the existing
"Advanced" toggle to a coral pill labeled "Filters (n)" whenever
dropdown filters are applied. Users now see at a glance that the list
is being narrowed; the inline accordion remains the disclosure
mechanism. Adds aria-expanded for screen readers.
MOB-23: Add MOBILE.md at the repo root with the 360 px design baseline,
acceptance checks for any UI PR (no horizontal overflow, ≥44px tap
targets, no <11px visible text, iOS Chrome parity, safe-area-inset,
dvh), and the established component patterns. Playwright regression
test deferred — adding the dep for one test is heavier than the
current value warrants; documented as a future option.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
MOB-19: Add a viewport Viewport export with viewportFit: 'cover' and
themeColor entries for light (#faf7f2) / dark (#1a1612), plus the
appleWebApp metadata for the home-screen status bar style and title.
Manifest's stale #3b82f6 theme_color updated to match brand cream.
MOB-20: Apply env(safe-area-inset-*) to the sticky chrome — the top
header gets max(padding, inset-left/right) so the logo and tab links
clear the notch in landscape; the bottom tab bar already had
inset-bottom and now also gets inset-left/right.
MOB-21: Replace 100vh with 100dvh in body min-height, modal max-heights,
the map view container, and the fullscreen map. Older engines fall
back via the duplicated vh declaration.
MOB-22: Set -webkit-tap-highlight-color: transparent on body to
suppress the iOS Safari grey flash; add a generic touch-pointer
:active rule (opacity 0.7) so taps still register visually on plain
anchors and bare buttons. Components with their own :active styling
are unaffected.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
MOB-14: PerformanceChart and ComparisonChart both already configure
legend position 'top' — no change needed.
MOB-15: Compare's detailedTable is 774px wide inside an overflow-x:
auto wrapper. Add a right-edge mask-image fade at ≤640px so phone
users see the table extends past the viewport.
MOB-16: ComparisonView's Share button previously did clipboard-only.
Prefer navigator.share when available (iOS/Android native sheet) so
users can send straight to Messages / WhatsApp / Mail / etc. Fall
back to clipboard with the existing "Copied!" toast otherwise. User
cancellations swallow silently; only real errors trigger fallback.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
MOB-07: Admissions deadlines strip becomes a horizontal snap-scroller
on ≤640px instead of a cramped 2×2 grid (which forced "Secondary ·
Deadline" track labels down to 9.6px). Cards stay readable, the right
edge fades to signal more content past the viewport, .chipTrack font
bumped to 0.7rem.
MOB-09: Result-card line3 (headline metric + secondary stats) was
crowding everything onto one row. Force the first .stat (Attainment
8 / RWM headline) to flex-basis 100% on mobile so delta-vs-LA and
pupil count wrap below it with a visible row-gap. Applied to both
SchoolRow (primary) and SecondarySchoolRow.
MOB-12: MetricTooltip ⓘ icons rendered at ~9px and relied on :hover
(which doesn't fire on touch). Hide the whole .wrapper at ≤640px —
metric labels themselves carry the meaning. Saves building a
tap-to-show layer for now.
MOB-13: The "Ofsted pending / No inspection on record" empty state
took a full hero card to communicate non-information. Add a
data-ofsted-state attribute on the hero chip; on ≤640px, the
"none" state collapses to a single muted line.
MOB-17: Already had Type+Action columns hidden on rankings mobile —
no change needed beyond marking complete.
MOB-18: Long metric headers ("Reading, Writing & Maths Combined %")
forced the value column wide. Drop .valueHeader to 0.625rem with
white-space: normal at ≤640px so labels wrap onto 2 short lines.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The /api/rankings endpoint returned each row keyed by the metric's
column name (e.g. rwm_high_pct) but never under a generic `value`
field. The frontend RankingItem type and RankingsView both read
ranking.value, so every row rendered "—" for every metric — the
default rwm_expected_pct included.
Add `df["value"] = df[metric]` before JSON serialisation so the
frontend gets the value it has always expected. The raw metric
column is still in the row for any caller that wants it explicitly.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
MOB-06: On phones the home hero stacked eyebrow tag, h1, description
paragraph, search input, button, "Schools near me" and three explore
chips before the user could see the deadlines strip. Hide the eyebrow
and the descriptive paragraph at ≤640px (the h1 already names the
product; the search input is the primary action) and move the
"Start exploring" chips to render after the admissions deadlines —
time-sensitive info now leads, generic discovery follows. Result on
390×844: heading → search → Schools near me → first deadline chip
all fit above the fold.
MOB-11: The school-detail hero took ~2 viewports before the first real
metric. At ≤768px, switch .meta back to row+wrap so the short pills
("Manchester" / "Voluntary aided") flow 2-per-row instead of stacking
3 full rows, and hide the .headerDetails block (headteacher / website /
pupil count / trust) — secondary info that lives in the Pupils &
Inclusion section anyway. Reclaims ~70px of hero so the Ofsted card
and the headline metric surface within a single viewport.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
MOB-05: With the bottom tab bar's Compare badge now showing the
selected count and providing one-tap navigation to /compare, the
floating toast becomes redundant chrome on phones — it cost ~70px
of permanent vertical space and visually competed with the tab bar
right above it. Hide at ≤640px. Per-school removal still works on
the /compare page itself. Desktop unaffected.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
iOS Chrome (and some Android browsers) auto-hide their URL bar on
scroll. This grows the visual viewport without changing the layout
viewport, so a position:fixed bar pinned to bottom:0 — which is
relative to the layout viewport — appears to float mid-screen with
a gap beneath it. Safari masks the bug because its toolbar shrinks
rather than fully retracting.
Track the delta between the visual and layout viewports via the
VisualViewport API and write it to a --mobile-bar-offset CSS var.
The bar uses translate3d to apply that offset, which both fixes the
gap and enables hardware compositing so it tracks the toolbar
animation without flicker.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
MOB-03: The school-detail section nav (SATs / Admissions / Pupils /
Location / History) overflows on phones (scrollWidth 462 vs clientWidth
356) with no signal that more tabs exist past the right edge. Add a
right-edge mask-image fade at ≤640px, scroll-snap on each link, and
bigger tap targets (min-height 36px, padding bumped). A scroll/resize
listener toggles an .atEnd class that removes the fade once the user
has scrolled to the last tab.
MOB-04: The "What you'll see on every school" cards rendered preview
visuals (mini cascade chart, Ofsted badge, compare table) with text
scaled down to 6–10px — unreadable on a phone. Hide .hiwVisual at
≤640px and let the explanatory text carry each card; tiny visible
text count on the home dropped from 51 to 8.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
MOB-01: At ≤640px the top inline nav is hidden and replaced by a
fixed bottom tab bar (Search / Compare / Rankings / Admissions) with
icon + label, 56px tap targets, env(safe-area-inset-bottom) padding,
and the Compare count rendered as a badge on the icon. Eliminates
horizontal page overflow on every route (docW was 401–429 at vw 390;
now docW === vw).
MOB-02: Logo link gains a padded hit area so the touch target is
≥44×44 (was 36×36), without resizing the visual mark.
ComparisonToast lifted above the new bottom bar on mobile so the two
do not stack on top of each other. body gets a bottom padding equal
to the bar height + safe-area inset so page content is never hidden.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Header "Pupils" line reverts to plain inline text matching Headteacher
and website weight. Gender split lives in Pupils & Inclusion as a card
alongside pupil premium / EAL / SEN support — peers of similar weight.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Upgrades the existing "Pupils" stat to include a compact split bar and
percentage hint for mixed schools (single-sex schools already carry a
"Boys's/Girls's school" badge, so the split would be redundant).
Wires fact_pupil_characteristics into the API: new SQLAlchemy model and
a real census block in /api/schools/{urn} replacing the prior null stub.
On the primary detail page the inline "Pupils: 241" text is replaced by
a richer block (display number + bar + "52% girls · 48% boys"). On the
secondary detail page the existing "Total pupils" hero stat card grows
the bar and hint beneath the number. Both fall back to the previous
text-only rendering when census gender data is missing.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Rail pins alongside the content with scroll-spy highlighting the current
section (Primary, Secondary, Tips). Collapses to a sticky top pill bar
on narrow screens. Primary section now precedes Secondary.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Chips now sort soonest-first at hydration time so the most urgent
deadline always appears first. The rail is hidden (opacity 0) until
the useEffect populates and sorts the chips, then fades in — avoiding
any visible layout shift from the reorder.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
int_ks4_with_lineage references stg_legacy_ks4 but the model was never
selected for build, causing a missing relation error.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
LegacyKS2Stream now auto-detects ZIP vs bare CSV — if the download is a ZIP
it extracts england_ks2final.csv; if it's a plain CSV file it reads directly.
This keeps backwards compatibility while allowing both streams to share the
same DfE annual archive URLs.
legacy_ks2_urls updated to point at the same 4 ZIPs as legacy_ks4_urls so
only one set of archives needs to be maintained going forward.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Mirrors the existing legacy KS2 pattern to fill the gap before EES hosted
KS4 data. Four files changed:
- tap-uk-ees: LegacyKS4Stream downloads each year's DfE Compare School
Performance ZIP, extracts england_ks4final.csv, maps 416 legacy columns
to Singer fields, strips % suffixes. Registered in discover_streams().
TapUKEES.config_jsonschema gains legacy_ks4_urls setting.
- stg_legacy_ks4.sql: safe_numeric casts + NULL placeholders for columns
not present in legacy format (ebacc_avg_score, gcse_grade_91_pct,
prior_attainment_avg, sen_pct).
- int_ks4_with_lineage.sql: adds all_ks4 CTE unioning stg_ees_ks4 and
stg_legacy_ks4, matching the int_ks2_with_lineage pattern.
- _stg_sources.yml + meltano.yml: source declaration and setting definition
for legacy_ks4. URLs configured per-year once provided.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Two bugs prevented historical secondary school data from loading:
1. stg_ees_ks4.sql filtered breakdown_topic = 'Total' only, but EES
releases prior to 2023/24 use breakdown_topic = 'All pupils' (matching
the KS2 convention). All older years were silently dropped to zero rows.
Fix: accept both values with an IN clause.
2. get_all_releases() in tap-uk-ees fetched only the first page of the
EES releases API. Now follows all pages via the paging.totalPages field
so no historical release is missed when more than 20 exist.
After re-running the annual EES pipeline, secondary school comparison
charts should show data across all available years (2018/19 onwards).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Once the user explicitly clicks a phase tab, suppress auto-phase detection
so switching to Secondary (or Primary) can't be snapped back by the effect
that fires when comparisonData re-fetches.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The useEffect that auto-selects the Primary/Secondary tab was calling
setComparePhase() without updating selectedMetric. When a URL carries a
primary metric (e.g. metric=rwm_expected_pct) but the shortlisted schools
are secondary, the tab would switch to Secondary while the metric stayed
at rwm_expected_pct — which is null for all secondary schools, causing
every card to show "–" and requiring a manual tab toggle to fix.
Fix: after determining the phase from school data, check whether the
current metric belongs to that phase's category list. If not, reset to
the phase default (attainment_8_score for secondary, rwm_expected_pct
for primary). A metric that already fits the phase (e.g. the URL already
carried attainment_8_score) is preserved unchanged.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
P1 (backend/data_loader.py): Add load_latest_school_data() which pre-computes
the one-row-per-school latest-year snapshot (groupby, prev-year trend merge)
once at startup instead of on every /api/schools request. get_schools route
now starts from the cached snapshot rather than rebuilding it.
S3 (backend/app.py): Wrap synchronous geocode_single_postcode() call in
asyncio.to_thread() so postcode lookups no longer block the uvicorn event
loop. Admin reload endpoint also uses to_thread for both cache primes.
P2 (nextjs-app/components/HomeView.tsx): Add mapParamsRef guard so switching
back to map view does not re-fetch 500 schools when search params haven't
changed. Reset ref on new searches so fresh data is always fetched.
P3 (nextjs-app/lib/chartSetup.ts): Extract Chart.js registration into a
shared side-effect module. ComparisonChart and PerformanceChart now import
it instead of each calling ChartJS.register() independently.
P4 (backend/database.py): Remove unnecessary db.commit() from the read-only
get_db_session() context manager — saves a DB round-trip on every request.
P5 (backend/database.py): Add pool_recycle=1800 to SQLAlchemy engine to
prevent stale TCP connections from accumulating in long-running processes.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- New /admissions route with AdmissionsView client component
- Live countdowns (days until) to Primary/Secondary deadlines and Offer Days
- Step-by-step timelines for both tracks with highlighted milestone rows
- Tips section covering equal preference rule, late applications, waiting lists
- Homepage countdown strip (4 cards) between discovery chips and how-it-works
- Admissions nav link and footer link added
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- GCSE top metrics (Att8, P8, Eng+Maths 4+/5+) promoted to heroStatCard
teal-tinted cards with Playfair serif values and DeltaChip vs national
- Attainment 8 visual bar: 0–80 scale with coral national-average marker
and pill label, mirrors the SatsChart concept for a score metric
- Progress 8 number line: −3 to +3 axis showing CI band, zero baseline,
and a teal/coral dot for the school's score (hidden when P8 suspended)
- SEN section upgraded from plain metricCard to heroStatCard grid
- History table moved into a details/summary accordion (collapsed by
default); PerformanceChart now lives at the top of the History section
always visible above the disclosure
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Hero: Playfair heading with coral italic accent, teal eyebrow pill,
richer sub-copy describing both primary and secondary coverage
- Discovery: geolocation "Schools near me" button (reverse-geocodes via
postcodes.io → /?postcode=…&radius=1), plus Start exploring chips
linking to /rankings and /compare
- How it works: 3-card grid showing miniature real-UI previews for
Performance (primary SATs cascade + secondary Att8 bar), Ofsted
inspection card, and side-by-side Compare table
- Editorial: text column + factbox (totalSchools, LA count, coverage
dates) rendered inside a white card below the how-it-works section
- Footer: expanded to 3 columns (brand blurb, Product, Resources);
links updated to / /rankings /compare and real gov.uk/ofsted URLs
- All new sections visible only on landing (no search active)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Renames the RWM-specific .rwmHero* classes to generic .heroStat*
and reuses them on the three Pupils & Inclusion headline cards
(disadvantaged, EAL, SEN support). Grid switches to auto-fit to
accommodate both 2- and 3-card layouts. SEN primary-need breakdown
stays on the dense .metricCard style.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Top two combined cards now use a teal-tinted background with a
2.1rem Playfair serif teal value and a compact uppercase label —
consistent with the hero treatment shown in the design mockup.
Scoped via new .rwmHero* classes so other metric cards on the page
keep their existing style.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Bars drop from 22px to 14px with softer 4px radius, percentages render
in Playfair serif to match the rest of the detail page typography,
national-average line opacity bumped from 25% to 35% for legibility.
Subject name and ruler proportions also aligned with the mockup.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Explains the intersection semantics of RWM combined — a pupil is only
counted if they met the bar in all three subjects — with a math line
showing the per-subject percentages collapsing to the combined figure.
Only renders when all four values are present; national-average pill
markers on the cascade are untouched.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The Oversubscribed badge + "Demand exceeds capacity" text lived at
the bottom of the tile as an afterthought. It's the headline finding —
a parent should read it first, then let the Q&A supply the detail.
Replace the footer badge with a Playfair Display sentence directly
under the section title: "This school is oversubscribed." (state word
coloured coral for oversubscribed, teal for undersubscribed). The
"Demand exceeds capacity" / "Supply meets demand" line sits below as
quiet muted text — present for context but not competing with the
headline.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The "How Hard to Get Into This School" tile mixed a progress bar
(places vs first-choice) with three text metric cards, making the
data feel fragmented and hiding the real narrative. The progress bar
also broke visually when undersubscribed and didn't scale to different
school sizes.
Replace with a typographic Q&A list that answers the questions
parents actually ask — "How many places were offered?", "How many
families wanted this school first?", "How many got their first
choice?", "How many applied in total?" — with a verdict footer
(Oversubscribed / Not oversubscribed + one-sentence explanation).
The third row now uses first_preference_offers (already in the API
response) to show "27 of 42 (64.3%)" instead of just the percentage,
giving the raw count parents actually want.
Each row is independently null-gated; rows stack vertically under
480px so the Playfair numeral stays legible.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The staging model aliased EES's total_number_places_offered column as
published_admission_number, but PAN is the school's published capacity
(not exposed by EES at school level) — what we actually have is the
count of places offered in a given admissions round. The misnomer
propagated to the mart, SQLAlchemy model, API response, TS types, and
UI copy ("places per year", "(PAN)").
Rename end-to-end and fix the UI labels:
- "29 places for 42 first-choice applications"
→ "29 places offered for 42 first-choice applications"
- "Reception/Year 7 places per year"
→ "Reception/Year 7 places offered"
- drop the misleading "(PAN)" suffix in the secondary view
Also add a comment in stg_ees_admissions clarifying this is the number
of places offered, not PAN. Requires dbt to rebuild fact_admissions
(marts are materialized as tables) before the backend can start.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Labels were sharing a flex row with the bar, so the bar's width %
was computed against the flex remainder after the label rather than
the full chart area. A 96% bar rendered around the 75% ruler mark,
and mobile was worse. Move labels to a header row above each bar and
give the bar a full-width track, so X% now aligns with X% on the ruler.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Redesign the School Details page for better parent comprehension:
- New SatsChart component: horizontal cascade bars with ruler scale and
national average marker (teal/coral palette matching site theme)
- Admissions section: visual progress bar showing 1st-preference demand
vs available places, colour-coded by oversubscription status
- Historical data: collapse raw year-by-year table behind a disclosure
element while keeping the performance line chart always visible
- EAL metric: add national average comparison via DeltaChip (backend now
includes eal_pct in national averages endpoint)
- New formatWithSuppression utility for null/suppressed data handling
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
dim_school.sql was checking for int_ofsted_latest in target.schema (wrong schema)
due to the custom generate_schema_name macro using literal schema names. The
model lives in 'intermediate', so ofsted_grade/date/framework were always NULL
in dim_school, causing all list cards to show 'Not yet inspected'.
Fix 1: data_loader.py joins marts.fact_ofsted_inspection with DISTINCT ON to
get latest inspection per school — no pipeline re-run needed.
Fix 2: dim_school.sql uses schema='intermediate' so future dbt runs correctly
denormalise the Ofsted summary into dim_school.
Replaces the bare Leaflet popup with a mini card showing school name,
3-state Ofsted badge (OEIF grade / ReportCard / pending), phase tag,
headline metric (Att8 for secondary, RWM% for primary) with delta vs
LA/national average, and a styled View Details button. Threads
nationalAvgRwm and laAverages from HomeView → SchoolMap → LeafletMapInner.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add nationalAvgRwm state fetched from /api/national-averages on mount
- Pass nationalAvgRwm to SchoolRow (vs-national delta now active in list view)
- Pass nationalAvgRwm to SchoolMap (prop accepted, threaded to Task 7)
- Redesign CompactSchoolItem: Ofsted badge + single headline metric + delta
- Fix stray backslash in SchoolRow.module.css .vsNationalFlat selector
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add ofsted_framework field to School type
- Add OfstedListBadge interface and buildOfstedListBadge pure function to utils.ts
- Add fetchNationalAverages API function that calls GET /api/national-averages
- Add test suite for buildOfstedListBadge (all 6 new tests pass)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Tudor Sitaru