@ak--47/dungeon-master 1.5.3 → 1.6.0

package/dungeons/vertical/food-delivery/food-delivery.sql

@@ -0,0 +1,183 @@
+-- ============================================================
+-- food-delivery.js — v1.6.0 hook verification queries (human eyeball)
+-- Machine contract lives in the `stories` export of food-delivery.js;
+-- run ./food-delivery.verify.mjs for verdicts. These queries mirror the
+-- story reads for interactive inspection.
+--
+-- Generate first:
+--   node scripts/verify-runner.mjs dungeons/vertical/food-delivery/food-delivery.js verify-food-delivery
+--
+-- Key derivation notes (full math in HOOK STORIES block of food-delivery.js):
+-- - H1/H7 read delivered-per-PLACED ratios: 'order placed' shares the soup
+--   HOD/hash distribution but is untouched by either drop, so it cancels
+--   the confound. Raw delivered HOD volume does NOT show a clean 0.70.
+-- - H6 zero-post share lands near 0.25 (not the 0.60 knob): the deletion
+--   removes late 'subscription started' events themselves, so churned
+--   users with a late trial start vanish from the visible cohort.
+-- - H8 buckets by HOOK-TIME order count; output counts are contaminated by
+--   H4 duplication and H6 deletion, so the clean read excludes rainy-window
+--   orderers and users with zero post-day-14 activity.
+-- - H10 reads reorders-per-DELIVERED: last funnel step fires only on
+--   conversion (determineConversion, lib/generators/funnels.js), so the
+--   ratio recovers the conversionRate scaling exactly (1.40 / 0.70).
+-- ============================================================
+
+
+-- H1: RUSH-HOUR KEEP RATE — delivered-per-placed off/rush ≈ 0.70
+WITH ev AS (
+  SELECT event, hour(time::TIMESTAMP) AS hr
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+  WHERE event IN ('order delivered','order placed')
+), agg AS (
+  SELECT (hr BETWEEN 11 AND 13) OR (hr BETWEEN 17 AND 20) AS rush,
+    count(*) FILTER (WHERE event='order delivered') AS del,
+    count(*) FILTER (WHERE event='order placed') AS placed
+  FROM ev GROUP BY 1
+)
+SELECT rush, del, placed, round(del::DOUBLE/placed, 3) AS del_per_placed FROM agg ORDER BY rush;
+
+
+-- H2: COUPON INJECTION — coupons-per-checkout diff ≈ +0.30 (Free − QB+); cpu ratio ≈ 2.2
+WITH pu AS (
+  SELECT user_id::VARCHAR AS uid, any_value(subscription_tier) AS tier,
+    count(*) FILTER (WHERE event='coupon applied') AS coupons,
+    count(*) FILTER (WHERE event='checkout started') AS checkouts
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+  WHERE event IN ('coupon applied','checkout started') GROUP BY 1
+)
+SELECT tier, count(*) AS users,
+  round(sum(coupons)::DOUBLE/count(*), 2) AS coupons_per_user,
+  round(sum(coupons)::DOUBLE/nullif(sum(checkouts),0), 3) AS coupons_per_checkout
+FROM pu GROUP BY tier ORDER BY tier;
+
+
+-- H3: LATE-NIGHT MUNCHIES — flip inversion (late−off)/(1−off) ≈ 0.70; price ratio ≈ 1.30
+WITH ev AS (
+  SELECT event, cuisine_type, TRY_CAST(item_price AS DOUBLE) AS price,
+    hour(time::TIMESTAMP) AS hr
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+  WHERE event IN ('restaurant viewed','restaurant browsed','item added to cart')
+)
+SELECT
+  round(avg((cuisine_type='American')::INT) FILTER (WHERE event='restaurant viewed' AND (hr>=22 OR hr<=2)), 3) AS amer_late,
+  round(avg((cuisine_type='American')::INT) FILTER (WHERE event='restaurant viewed' AND hr BETWEEN 6 AND 18), 3) AS amer_off,
+  round(avg((cuisine_type='American')::INT) FILTER (WHERE event='restaurant browsed'), 3) AS amer_organic_ctrl,
+  round(avg(price) FILTER (WHERE event='item added to cart' AND (hr>=22 OR hr<=2)) /
+        avg(price) FILTER (WHERE event='item added to cart' AND hr BETWEEN 6 AND 18), 3) AS price_ratio
+FROM ev;
+
+
+-- H4: RAINY WEEK (day_idx 20-27) — dup share ≈ 0.40, fee ratio ≈ 2.0, vol RoR vs checkout ≈ 1.4
+WITH ev AS (
+  SELECT user_id::VARCHAR AS uid, event, order_id, TRY_CAST(delivery_fee AS DOUBLE) AS fee,
+    date_diff('day', TIMESTAMP '2026-01-01 00:00:00', time::TIMESTAMP) AS day_idx
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+  WHERE event IN ('order placed','checkout started')
+)
+SELECT
+  round((count(*) FILTER (WHERE event='order placed' AND day_idx BETWEEN 20 AND 27)
+       - count(DISTINCT uid || '|' || order_id) FILTER (WHERE event='order placed' AND day_idx BETWEEN 20 AND 27))::DOUBLE
+       / count(DISTINCT uid || '|' || order_id) FILTER (WHERE event='order placed' AND day_idx BETWEEN 20 AND 27), 3) AS dup_share,
+  round(avg(fee) FILTER (WHERE event='order placed' AND day_idx BETWEEN 20 AND 27) /
+        avg(fee) FILTER (WHERE event='order placed' AND (day_idx BETWEEN 10 AND 19 OR day_idx BETWEEN 28 AND 37)), 3) AS fee_ratio,
+  round((count(*) FILTER (WHERE event='order placed' AND day_idx BETWEEN 20 AND 27)::DOUBLE /
+         count(*) FILTER (WHERE event='order placed' AND (day_idx BETWEEN 10 AND 19 OR day_idx BETWEEN 28 AND 37))) /
+        (count(*) FILTER (WHERE event='checkout started' AND day_idx BETWEEN 20 AND 27)::DOUBLE /
+         count(*) FILTER (WHERE event='checkout started' AND (day_idx BETWEEN 10 AND 19 OR day_idx BETWEEN 28 AND 37))), 3) AS vol_ror
+FROM ev;
+
+
+-- H5: REFERRAL POWER USERS (born-in only) — reorders/user ratio ≈ 1.4; ratings 4.4 vs 2.8
+WITH ev AS (
+  SELECT user_id::VARCHAR AS uid, event, referral_code, TRY_CAST(food_rating AS DOUBLE) AS rating
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+  WHERE event IN ('account created','reorder initiated','order rated')
+), born AS (
+  SELECT uid, bool_or(referral_code = true) AS referred
+  FROM ev WHERE event='account created' GROUP BY 1
+), pu AS (
+  SELECT b.uid, b.referred,
+    count(*) FILTER (WHERE e.event='reorder initiated') AS reorders,
+    avg(e.rating) FILTER (WHERE e.event='order rated') AS user_rating
+  FROM born b LEFT JOIN ev e ON e.uid=b.uid GROUP BY 1,2
+)
+SELECT referred, count(*) AS users,
+  round(sum(reorders)::DOUBLE/count(*), 2) AS reorders_per_user,
+  round(avg(user_rating), 2) AS mean_user_rating
+FROM pu GROUP BY referred ORDER BY referred;
+
+
+-- H6: TRIAL CHURN (survivor-biased view) — zero-post share ~0.25 nonactivated vs ~0.04 activated
+WITH ev AS (
+  SELECT user_id::VARCHAR AS uid, time::TIMESTAMP AS t, event, trial
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+), fu AS (SELECT uid, min(t) AS first_t FROM ev GROUP BY 1),
+trial AS (SELECT uid FROM ev WHERE event='subscription started' AND trial = true GROUP BY 1),
+pux AS (
+  SELECT f.uid,
+    count(*) FILTER (WHERE e.event='order placed' AND e.t <= f.first_t + INTERVAL '14 days') AS early_orders,
+    count(*) FILTER (WHERE e.t > f.first_t + INTERVAL '14 days') AS post_n
+  FROM fu f JOIN trial tr ON tr.uid=f.uid JOIN ev e ON e.uid=f.uid GROUP BY 1
+)
+SELECT (early_orders >= 3) AS activated, count(*) AS users,
+  round(avg(CASE WHEN post_n = 0 THEN 1.0 ELSE 0 END), 3) AS zero_post_share
+FROM pux GROUP BY 1 ORDER BY 1;
+
+
+-- H7: HASH-BUCKET DROP — delivered-per-placed odd/even ≈ 0.70
+WITH pu AS (
+  SELECT user_id::VARCHAR AS uid, (ascii(substr(user_id::VARCHAR,1,1)) % 2 = 0) AS even_bucket,
+    count(*) FILTER (WHERE event='order placed') AS placed,
+    count(*) FILTER (WHERE event='order delivered') AS delivered
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+  WHERE event IN ('order placed','order delivered') GROUP BY 1,2
+)
+SELECT even_bucket, count(*) AS users,
+  round(sum(delivered)::DOUBLE/nullif(sum(placed),0), 3) AS del_per_placed
+FROM pu GROUP BY even_bucket ORDER BY even_bucket;
+
+
+-- H8: ORDER-COUNT MAGIC NUMBER (clean population) — sweet/base ≈ 1.40, over/sweet ≈ 0.46
+WITH raw AS (
+  SELECT user_id::VARCHAR AS uid, time::TIMESTAMP AS t, event,
+    TRY_CAST(order_total AS DOUBLE) AS ot,
+    date_diff('day', TIMESTAMP '2026-01-01 00:00:00', time::TIMESTAMP) AS day_idx
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+), fu AS (SELECT uid, min(t) AS first_t FROM raw GROUP BY 1),
+flags AS (
+  SELECT r.uid,
+    max(CASE WHEN r.event='order placed' AND r.day_idx BETWEEN 20 AND 27 THEN 1 ELSE 0 END) AS rainy,
+    max(CASE WHEN r.t > f.first_t + INTERVAL '14 days' THEN 1 ELSE 0 END) AS alive,
+    count(*) FILTER (WHERE r.event='order placed') AS orders,
+    sum(r.ot) FILTER (WHERE r.event='order placed') AS spend
+  FROM raw r JOIN fu f USING(uid) GROUP BY 1
+)
+SELECT CASE WHEN orders BETWEEN 4 AND 8 THEN 'sweet' WHEN orders >= 9 THEN 'over' ELSE 'base' END AS bucket,
+  count(*) AS users, round(sum(spend)/nullif(sum(orders),0), 1) AS avg_order_total
+FROM flags WHERE rainy = 0 AND alive = 1 AND orders > 0
+GROUP BY 1 ORDER BY 1;
+
+
+-- H9: TIER DELIVERY SPEED — QB+/Free property ratio ≈ 0.479 (= 0.67/1.4) on both props
+SELECT
+  round(avg(TRY_CAST(actual_delivery_mins AS DOUBLE)) FILTER (WHERE subscription_tier='QuickBite+' AND event='order delivered') /
+        avg(TRY_CAST(actual_delivery_mins AS DOUBLE)) FILTER (WHERE subscription_tier='Free' AND event='order delivered'), 3) AS adm_ratio,
+  round(avg(TRY_CAST(eta_mins AS DOUBLE)) FILTER (WHERE subscription_tier='QuickBite+' AND event='order tracked') /
+        avg(TRY_CAST(eta_mins AS DOUBLE)) FILTER (WHERE subscription_tier='Free' AND event='order tracked'), 3) AS eta_ratio
+FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+WHERE event IN ('order delivered','order tracked');
+
+
+-- H10: CITY DENSITY — reorders-per-delivered dense/base ≈ 1.40, sprawl/base ≈ 0.70
+WITH pu AS (
+  SELECT user_id::VARCHAR AS uid, any_value(city) AS city,
+    count(*) FILTER (WHERE event='order delivered') AS delivered,
+    count(*) FILTER (WHERE event='reorder initiated') AS reorders
+  FROM read_json_auto('data/verify-food-delivery-EVENTS*.json', sample_size=-1, union_by_name=true)
+  WHERE event IN ('order delivered','reorder initiated') GROUP BY 1
+)
+SELECT CASE WHEN city IN ('San Francisco','New York') THEN 'dense'
+  WHEN city IN ('Houston','Phoenix') THEN 'sprawl' ELSE 'base' END AS grp,
+  count(*) AS users,
+  round(sum(reorders)::DOUBLE/nullif(sum(delivered),0), 4) AS reorders_per_delivered
+FROM pu GROUP BY 1 ORDER BY 1;

package/dungeons/vertical/food-delivery/food-delivery.verify.mjs

@@ -0,0 +1,75 @@
+#!/usr/bin/env node
+/**
+ * food-delivery.verify.mjs — thin wrapper around the story runner.
+ *
+ * All verification logic lives in the `stories` export of
+ * ./food-delivery.js; this script just streams the shards and
+ * delegates. It is equivalent to:
+ *
+ *   node scripts/verify-stories.mjs dungeons/vertical/food-delivery/food-delivery.js --data-prefix verify-food-delivery
+ *
+ * Generate first:
+ *   node scripts/verify-runner.mjs dungeons/vertical/food-delivery/food-delivery.js verify-food-delivery
+ * Run:
+ *   node dungeons/vertical/food-delivery/food-delivery.verify.mjs [prefix]   # default verify-food-delivery
+ */
+import fs from 'node:fs';
+import path from 'node:path';
+import readline from 'node:readline';
+import { execFile } from 'node:child_process';
+import { promisify } from 'node:util';
+import { buildIdentityMap, evaluateStories, VERDICT_RANK } from '@ak--47/dungeon-master/verify';
+import config, { stories } from './food-delivery.js';
+
+const PREFIX = process.argv[2] || 'verify-food-delivery';
+const prefixPath = path.join('data', PREFIX);
+
+async function loadShards(suffix) {
+	// streaming load: full-fidelity EVENTS shards exceed the readFileSync cap
+	const dir = path.dirname(prefixPath), base = path.basename(prefixPath);
+	const out = [];
+	if (!fs.existsSync(dir)) return out;
+	for (const f of fs.readdirSync(dir).filter(f => f.startsWith(`${base}-${suffix}`) && f.endsWith('.json')).sort()) {
+		const rl = readline.createInterface({ input: fs.createReadStream(path.join(dir, f)), crlfDelay: Infinity });
+		for await (const line of rl) {
+			if (line.trim()) out.push(JSON.parse(line));
+		}
+	}
+	return out;
+}
+
+const events = await loadShards('EVENTS');
+const profiles = await loadShards('USERS');
+if (!events.length) {
+	console.error(`no shards at ${prefixPath}-EVENTS*.json — run: node scripts/verify-runner.mjs dungeons/vertical/food-delivery/food-delivery.js ${PREFIX}`);
+	process.exit(1);
+}
+console.log(`food-delivery — events=${events.length} users=${profiles.length} (${prefixPath})`);
+
+const execFileP = promisify(execFile);
+const runSql = async (sql) => {
+	const { stdout } = await execFileP('duckdb', ['-json', '-c', sql.replaceAll('{{PREFIX}}', prefixPath)], { maxBuffer: 64 * 1024 * 1024 });
+	return stdout.trim() ? JSON.parse(stdout) : [];
+};
+
+// All stories are DuckDB-backed. Every read here is a ratio-of-ratios
+// (delivered-per-placed, reorders-per-delivered, coupons-per-checkout) or a
+// behavioral-cohort aggregation (hash bucket, hook-time order-count bucket,
+// trial-activation split) — shapes emulateBreakdown's segmentation types
+// don't model, and the RoR normalization is exactly what cancels the soup
+// confounds a raw emulator read would inherit.
+const results = await evaluateStories(stories, events, {
+	profiles,
+	funnels: config.funnels,
+	identityMap: buildIdentityMap(profiles),
+	runSql,
+});
+
+let worst = 'NAILED';
+for (const r of results) {
+	console.log(`${r.verdict.padEnd(7)} ${r.id}`);
+	for (const a of r.assertions) console.log(`        ${a.verdict.padEnd(7)} ${a.detail}`);
+	if (VERDICT_RANK[r.verdict] < VERDICT_RANK[worst]) worst = r.verdict;
+}
+console.log(`\nworst verdict: ${worst}`);
+process.exit(VERDICT_RANK[worst] >= VERDICT_RANK.STRONG ? 0 : 1);