- Health panel relocated below Backtest (last section) per owner — it's
operational detail, not an investment view.
- scheduler._load_source_health now returns only the LATEST entry per source
key (was returning the full refresh history, so siamchart_vintages and
price_snapshot appeared as 2 rows). Verified: 11 sources / 11 unique rows.
- frontend v-for drops slice(0,20) and keys by source key.
Verified in-browser: #health is the last section; 11 unique source rows.
Two root causes of tables and panels spilling past the window frame:
1. Global 'table { min-width: 850px }' forced EVERY table (source/score/modal/
backtest) to 850px regardless of container -> now min-width:0 so tables
shrink to 100%; only .factor-table (stock board, inside .table-wrap)
keeps its 850px and scrolls. Small tables get table-layout:fixed +
overflow-wrap on mobile so long content wraps instead of pushing out.
2. Grid/flex tracks lacked minmax(0,1fr) / min-width:0, so wide grid items
(KPI numbers, theme cards) pushed the whole row past the viewport at
768-1040px (doc overflow 208px). Fixed: all grid-template-columns use
minmax(0,1fr); .content is now flex:1 1 auto + min-width:0; KPI/theme/bt
cards get min-width:0 + overflow-wrap.
Verified programmatically in-browser: document overflow = 0 at 320/360/500/
600/720/760/768/820/900/1040/1200/1440px; score-table + source-table over=0
at all widths (incl inside the per-source modal). Board scrolls in .table-wrap
as designed. Backend suite 369 green (CSS-only change).
After reviewer suggestion: per theme, display sum(contribution)/sum(|weight|) =
theme surprise so the owner can confirm the source audit matches the real score.
Q2 flexible scoring: _fetch_with_cache now degrades instead of raising
DashboardError — a source that fails with no cached value returns {} so the
theme scorer drops that source's factors; a previously-good value is kept as
stale by the daily cache. Verified: all-sources-down still builds 13 themes.
Q3 per-source audit: new themes.factor_source_breakdown(fetched, theme) shows
per factor source/raw/normalized/weight/contribution; dashboard exposes
fetch_data + factor_sources; per-symbol modal renders symbolDetail.factor_sources
(e.g. retail: te_thailand ยอดขายปลีก -14.5 -> -1.0 x 0.7 = -0.7).
Suite 369 green; independent review passed: true.
Q1 (HAR for deferred sources) spike recorded: method works, REIC needs deeper
interaction; NBTC 403 likely unbpassable without a session.
- define missing design tokens (--card/--foreground/--accent/--font/--text-2)
that theme/modal components referenced but :root never declared (they rendered
transparent/wrong color)
- zero section-panel padding so .signal-header is the single top-spacing source
(was double 22px+22px on theme/lineage/health/sim/backtest panels)
- theme cards now surface the new source reads (banks rate, retail sales YoY,
nonbank household debt, property prices, telecom business confidence)
Add self-contained Docker deployment for EasyPanel / docker-compose:
- Dockerfile: python:3.11-slim + nginx. Serves the PREBUILT Vue SPA
(frontend/dist, committed) via nginx and reverse-proxies /api to the Flask
backend on 127.0.0.1:5000. No node/npm in the image (avoids Vite/npm
flakiness in Docker builds). VOLUME /app/backend/data for persistence
(factor vintages, dividend ledger, forward runs, prices). HEALTHCHECK hits
/api/v1/dashboard/summary. Exposes :80.
- deploy/nginx.conf: SPA fallback + /api proxy to 127.0.0.1:5000 + gzip.
- deploy/entrypoint.sh: starts Flask (HOST=0.0.0.0:5000) + nginx foreground,
forwards signals.
- docker-compose.yml: single service 'set50', port 8080:80, mounts
./backend/data for persistence, healthcheck.
- .dockerignore excludes .git/.venv/backend/data/node_modules/docs.
- .gitignore now tracks frontend/dist/ (prebuilt bundle required by the
image); backend/data stays untracked.
- Backend runtime verified: test_client GET /api/v1/dashboard/summary=200
(the HEALTHCHECK path). entrypoint sh -n, compose YAML parse, and all
Dockerfile-referenced files exist. NOTE: no local docker here, so the
image itself was not built — that happens on EasyPanel.
Backtest result card and saved-run history now show an explicit dividend
badge per run: green 'ตามวันจริง' for dated_ledger (real ex-date x qty cash
flow), amber 'Proxy (ต่อหุ้น)'/'Proxy' for estimates, plus a footnote that
matches the actual dividend_method. Frontend build passes (bundle
index-DDItjwYy.js); served bundle contains the new strings.
Forward mode in the simulation tab now calls the durable /api/v1/forward
lifecycle instead of the single-pass /api/v1/simulation (which only differed
by a mode string). runSimulation branches to /api/v1/forward for forward mode,
loads runs on mount, and renders a forward panel listing each paper run with
its status (frozen/executed/marked/matured), non-PIT badge, holdings and a
Mark / Mature action per run. Frontend build passes (bundle index-INuvPOCF.js).
Map dividend_method (dated_ledger / dps_annual_proxy / final_holdings_yield
_proxy) to an honest Thai label. Show a Proxy badge whenever the dividend is
an estimate (everything but dated_ledger) on both the result card and the
saved-run history rows, and set the footnote to match the actual method.
Frontend build passes.
- P1: /api/v1/simulation now uses the canonical board score (default_scores)
instead of a divergent 3-theme recompute -> 'จำลอง' can't disagree with board
(live check: sim top pick PTT == top board combined 1.600). Removes binary
auto/en signs, restores quality+momentum+dividend screen consistency.
- P2: dashboard emits source_summary{factor_keys, rows}; frontend shows
'N ปัจจัย · M แหล่ง' so the 7-vs-5 count confusion is impossible.
- P5: removed dead themes.list_themes()/Theme/build_theme_scores/_map_index and
the tests that locked them; added tests/conftest.py so pytest needs no PYTHONPATH.
- docs: audit-and-plan-2026-08-26.md (full P0-P5 plan) + engineering-log entry.
- 203 backend tests pass; Vite build passes. Independent reviewer: no security or
logic blockers (minor error-leak suggestion applied: 503 message no longer leaks
exception detail).
- backtest.py: buy-and-hold backtest over [start,end] — allocates 50/20/30 at first available rebalance date, marks to market to end, accrues dividend, reports {final_value, price_pnl, dividend_income, net_return, trades, holdings}
- Fixed double-spend bug (was allocating full capital every rebalance -> negative cash)
- dashboard.default_scores(): per-symbol combined/dividend/yield baseline for backtest
- POST /api/v1/backtest + GET /api/v1/backtest/runs (results persisted in app state -> survive refresh)
- Frontend: backtest section w/ start/end/capital/freq inputs + P&L KPIs + run history table
- Honest note: uses current combined scores as static baseline (non-PIT); PIT score_fn pluggable
- Verified: 1M -> 1.088M (+8.80%) over 2024-06..2026-06; history persists across refresh
- THEME_SYMBOLS expanded: added banks, retail, telecom_it, property, healthcare, petrochem_materials, consumer_staples, utilities, nonbank_finance, exploration -> all 49 SET50 names now in a theme
- THEME_LABELS_TH Thai labels; THEME_FREQUENCY per theme
- symbol_breakdown now lists EVERY theme the symbol belongs to (label_th + surprise, or 'ยังไม่มีข้อมูล'), so theme_score is transparent per source
- frontend: theme column maps all 49 symbols (mirrors backend); modal shows per-theme score detail
- Fixed test for BANPU multi-theme; full suite 199 OK
- Verified: 49/49 rows have theme chip; AOT modal shows ท่องเที่ยว 0.57σ + full calc
- Theme panel (ธีม): shows tourism/auto_credit/refining_energy with Thai labels, frequency chips (monthly/quarterly), and real factor reads (tourism surprise, car sales YoY, TOP net profit) from /api/v1/themes
- Combined count badge; responsive 1-col on mobile
- Loads /api/v1/themes in loadDashboard (already wired backend)
- Verified: npm build passes, browser shows 3 Thai cards with live values
