feat(iter-12): collapsible schema panel, funnel analysis & ROW_NUMBER dedup levels

[martinl5]

Summary

• Collapsible schema panel — right panel collapses to 40 px rail via › chevron; SQL editor expands to fill freed space; ‹ + vertical "SCHEMA" label restore it. Mobile unaffected.
• Level 72 — Loan Product Acquisition Funnel: UNION ALL multi-stage aggregation + CROSS JOIN scalar CTE yields 20→15→12→5 customers (100%→75%→60%→25%). Standard product-analytics funnel pattern.
• Level 73 — First Salary Credit Per Account: ROW_NUMBER() OVER (PARTITION BY account_id ORDER BY transaction_date ASC) + WHERE rn = 1 dedup. 4 rows. Universal ETL/SCD dedup pattern.

[UI/UX Improvements]

Collapsible schema panel (GameProvider.tsx): The right-hand Schema/Results column now collapses to a 40 px desktop rail by clicking the › chevron appended to the tab bar's right edge. When collapsed:

• Panel width transitions from lg:w-[420px] → lg:w-10 with transition-all duration-200 ease-in-out
• Collapsed rail shows a ‹ expand button (gold, bordered) and a vertical "SCHEMA" text label so the affordance is always visible
• SQL editor column changes from lg:w-[460px] flex-shrink-0 → lg:flex-1 min-w-0, growing to fill freed horizontal space — more room for long CTEs and window-function queries
• State is session-local (useState), not persisted — panel always opens on reload
• Both toggle buttons have aria-label + title; mobile stacked layout is unaffected

Directly implements the "collapsible schema panel" item from the UI/UX rotation.

[Game Design Tweaks]

Level 72 (Expert, difficulty 4) — "Loan Product Acquisition Funnel": Teaches the canonical SQL product-analytics funnel using UNION ALL of four scalar aggregate queries (tagged with ord) and a CROSS JOIN scalar CTE for the conversion-rate denominator. Four stages: All Customers (20, 100%), Loan Applicants (15, 75%), Active Loan Holders (12, 60%), Active Home Loan Holders (5, 25%). ORDER BY customer_count DESC gives natural funnel order (all counts distinct). Reinforces the CROSS JOIN scalar CTE idiom from Level 70 in a new product-analytics context. This exact UNION ALL + CROSS JOIN pattern is used in BigQuery/Redshift/Snowflake at FAANG and digital banks. No new data required.

Level 73 (Expert, difficulty 4) — "First Salary Credit Per Account (ROW_NUMBER Dedup)": Teaches the universal SQL deduplication pattern: ROW_NUMBER() OVER (PARTITION BY account_id ORDER BY transaction_date ASC) in a CTE, then WHERE rn = 1. Compliance framing (MAS Notice 632 income-verification) makes the business case concrete. Returns 4 rows — accounts 5/15/25/35 (customers Aisha Binte Yusof, Chen Mei Ling, Nur Hidayah, Fiona Tan) with their first Salary Credit amounts ($3 750–$8 250) and dates in Jan–Feb 2024. Pattern is identical across BigQuery, Redshift, Snowflake, Spark SQL, and PostgreSQL. No new data required.

[Database & Code Optimizations]

• LevelUpModal.tsx EPOCH_NEXT_HINT extended with entries for levels 72 (funnel) and 73 (ROW_NUMBER dedup); nextHintKey sentinel array extended to […, 71, 72, 73]; fallback updated ?? 71 → ?? 73
• MAX_LEVEL in progression.ts auto-derives from the last level array entry (now 73) — zero touch points
• Left panel layout: lg:w-[460px] flex-shrink-0 → lg:flex-1 min-w-0 so the editor grows when the panel collapses
• No new DB tables, seed rows, or indexes — both levels use existing loans, transactions, accounts, customers tables

Test results

Test Files  5 passed (5)
     Tests  405 passed (up from 395 in iter-11)
  Duration  786 ms
TypeScript  clean (tsc --noEmit)

🤖 Generated with Claude Code

https://claude.ai/code/session_011AFkb4EU6i5RTHn6sVFhGs

---

Generated by Claude Code

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
sqlwak	Ready	Preview, Comment	Jun 29, 2026 8:57pm