bunsane 0.3.1 → 0.4.0

package/studio/tailwind.config.js

@@ -1,77 +0,0 @@
-/** @type {import('tailwindcss').Config} */
-export default {
-  darkMode: ["class"],
-  content: [
-    './pages/**/*.{ts,tsx}',
-    './components/**/*.{ts,tsx}',
-    './app/**/*.{ts,tsx}',
-    './src/**/*.{ts,tsx}',
-  ],
-  prefix: "",
-  theme: {
-    container: {
-      center: true,
-      padding: "2rem",
-      screens: {
-        "2xl": "1400px",
-      },
-    },
-    extend: {
-      colors: {
-        border: "hsl(var(--border))",
-        input: "hsl(var(--input))",
-        ring: "hsl(var(--ring))",
-        background: "hsl(var(--background))",
-        foreground: "hsl(var(--foreground))",
-        primary: {
-          DEFAULT: "hsl(var(--primary))",
-          foreground: "hsl(var(--primary-foreground))",
-        },
-        secondary: {
-          DEFAULT: "hsl(var(--secondary))",
-          foreground: "hsl(var(--secondary-foreground))",
-        },
-        destructive: {
-          DEFAULT: "hsl(var(--destructive))",
-          foreground: "hsl(var(--destructive-foreground))",
-        },
-        muted: {
-          DEFAULT: "hsl(var(--muted))",
-          foreground: "hsl(var(--muted-foreground))",
-        },
-        accent: {
-          DEFAULT: "hsl(var(--accent))",
-          foreground: "hsl(var(--accent-foreground))",
-        },
-        popover: {
-          DEFAULT: "hsl(var(--popover))",
-          foreground: "hsl(var(--popover-foreground))",
-        },
-        card: {
-          DEFAULT: "hsl(var(--card))",
-          foreground: "hsl(var(--card-foreground))",
-        },
-      },
-      borderRadius: {
-        lg: "var(--radius)",
-        md: "calc(var(--radius) - 2px)",
-        sm: "calc(var(--radius) - 4px)",
-      },
-      keyframes: {
-        "accordion-down": {
-          from: { height: "0" },
-          to: { height: "var(--radix-accordion-content-height)" },
-        },
-        "accordion-up": {
-          from: { height: "var(--radix-accordion-content-height)" },
-          to: { height: "0" },
-        },
-      },
-      animation: {
-        "accordion-down": "accordion-down 0.2s ease-out",
-        "accordion-up": "accordion-up 0.2s ease-out",
-      },
-    },
-  },
-  plugins: [require("tailwindcss-animate")],
-}

package/studio/utils.ts

@@ -1,54 +0,0 @@
-/**
- * Finds the indicator component name from an archetype's field list.
- * The indicator is used to identify entities of this archetype type.
- * 
- * Priority order:
- * 1. {ArcheTypeName}Tag (e.g., UserTag)
- * 2. {ArcheTypeName}Id (e.g., UserId)
- * 3. Any field starting with {ArcheTypeName}
- * 4. Any field containing {ArcheTypeName}
- * 5. Fallback to first component
- */
-export function findIndicatorComponentName(
-    archeTypeName: string,
-    fields: Array<{ componentName: string; fieldName: string }>
-): string | null {
-    if (fields.length === 0) {
-        return null;
-    }
-
-    const archeTypeNameLower = archeTypeName.toLowerCase();
-    const componentNames = fields.map(field => field.componentName);
-
-    const tagComponentName = `${archeTypeName}Tag`;
-    const tagMatch = componentNames.find(
-        name => name.toLowerCase() === tagComponentName.toLowerCase()
-    );
-    if (tagMatch) {
-        return tagMatch;
-    }
-
-    const idComponentName = `${archeTypeName}Id`;
-    const idMatch = componentNames.find(
-        name => name.toLowerCase() === idComponentName.toLowerCase()
-    );
-    if (idMatch) {
-        return idMatch;
-    }
-
-    const startsWithMatch = componentNames.find(
-        name => name.toLowerCase().startsWith(archeTypeNameLower)
-    );
-    if (startsWithMatch) {
-        return startsWithMatch;
-    }
-
-    const containsMatch = componentNames.find(
-        name => name.toLowerCase().includes(archeTypeNameLower)
-    );
-    if (containsMatch) {
-        return containsMatch;
-    }
-
-    return componentNames[0] ?? null;
-}

package/studio/vite.config.js

@@ -1,19 +0,0 @@
-import { defineConfig } from 'vite'
-import react from '@vitejs/plugin-react'
-
-export default defineConfig({
-  base: '/studio/',
-  plugins: [react()],
-  build: {
-    outDir: 'dist',
-    rollupOptions: {
-      output: {
-        manualChunks: undefined,
-      }
-    }
-  },
-  server: {
-    port: 5173,
-    host: true
-  }
-})

package/tests/benchmark/BENCHMARK_DATABASES_PLAN.md

@@ -1,338 +0,0 @@
-# PGlite Persistent Benchmark Databases Plan
-
-## Overview
-
-Create pre-populated PGlite databases at various scale tiers for consistent, reproducible query benchmarks without re-seeding overhead.
-
-## Database Tiers
-
-| Tier | Entities | Products | Orders | Users | Reviews | Est. Size |
-|------|----------|----------|--------|-------|---------|-----------|
-| xs   | 10,000   | 5,000    | 3,000  | 1,000 | 1,000   | ~50 MB    |
-| sm   | 50,000   | 25,000   | 15,000 | 5,000 | 5,000   | ~250 MB   |
-| md   | 100,000  | 50,000   | 30,000 | 10,000| 10,000  | ~500 MB   |
-| lg   | 500,000  | 250,000  | 150,000| 50,000| 50,000  | ~2.5 GB   |
-| xl   | 1,000,000| 500,000  | 300,000| 100,000| 100,000| ~5 GB     |
-
-## Directory Structure
-
-```
-tests/benchmark/
-├── BENCHMARK_DATABASES_PLAN.md   # This plan
-├── databases/                     # Persistent PGlite databases (gitignored)
-│   ├── bench-xs/                 # 10k entities
-│   ├── bench-sm/                 # 50k entities
-│   ├── bench-md/                 # 100k entities
-│   ├── bench-lg/                 # 500k entities
-│   └── bench-xl/                 # 1M entities
-├── fixtures/
-│   └── BenchmarkComponents.ts    # Reusable components for benchmarks
-├── generators/
-│   ├── index.ts                  # Main generator entry point
-│   ├── DataGenerator.ts          # Realistic data generation utilities
-│   ├── EntitySeeder.ts           # Batch entity creation with relations
-│   └── ProgressReporter.ts       # CLI progress reporting
-├── runners/
-│   ├── QueryBenchmark.ts         # Query benchmark runner
-│   └── RelationBenchmark.ts      # BelongsTo/HasMany benchmark
-└── generate-databases.ts          # CLI script to generate all tiers
-```
-
-## Data Model (E-commerce Scenario)
-
-### Components (from RealisticComponents.ts)
-
-1. **Product** - Core product data (name, SKU, category, status, rating)
-2. **Inventory** - Stock tracking (quantity, warehouse, stock status)
-3. **Pricing** - Price info (base, sale, cost, discount, currency)
-4. **Vendor** - Supplier info (name, region, rating, tier)
-5. **ProductMetrics** - Analytics (views, purchases, conversion)
-
-### Additional Components for Relations
-
-6. **User** - Customer data (name, email, tier, region)
-7. **Order** - Order data (userId, status, total, itemCount)
-8. **OrderItem** - Line items (orderId, productId, quantity, price)
-9. **Review** - Product reviews (userId, productId, rating, text)
-10. **Wishlist** - User wishlists (userId, productIds[])
-
-### Relation Structure
-
-```
-User (1) ──HasMany──> Order (N)
-User (1) ──HasMany──> Review (N)
-User (1) ──HasMany──> Wishlist (1)
-
-Order (1) ──HasMany──> OrderItem (N)
-Order (N) ──BelongsTo──> User (1)
-
-OrderItem (N) ──BelongsTo──> Order (1)
-OrderItem (N) ──BelongsTo──> Product (1)
-
-Review (N) ──BelongsTo──> User (1)
-Review (N) ──BelongsTo──> Product (1)
-
-Product (1) ──HasMany──> Review (N)
-Product (1) ──HasMany──> OrderItem (N)
-Product (N) ──BelongsTo──> Vendor (1)
-```
-
-## Implementation Steps
-
-### Phase 1: Infrastructure Setup
-
-1. **Create directory structure**
-   - Create `tests/benchmark/` directories
-   - Add `databases/` to `.gitignore`
-
-2. **Create BenchmarkComponents.ts**
-   - Define User, Order, OrderItem, Review, Wishlist components
-   - Define BenchmarkArchetypes with proper relations
-
-3. **Create DataGenerator.ts**
-   - Faker-like utilities for realistic data (names, emails, addresses)
-   - Deterministic seeding with configurable seed for reproducibility
-   - Batch generation for memory efficiency
-
-### Phase 2: Seeding System
-
-4. **Create EntitySeeder.ts**
-   ```typescript
-   interface SeederConfig {
-     tier: 'xs' | 'sm' | 'md' | 'lg' | 'xl';
-     seed?: number;           // For reproducibility
-     batchSize?: number;      // Default 1000
-     showProgress?: boolean;  // CLI progress bar
-   }
-
-   class EntitySeeder {
-     async seedProducts(count: number): Promise<string[]>;
-     async seedUsers(count: number): Promise<string[]>;
-     async seedOrders(userIds: string[], productIds: string[]): Promise<void>;
-     async seedReviews(userIds: string[], productIds: string[]): Promise<void>;
-     async createIndexes(): Promise<void>;
-     async vacuum(): Promise<void>;
-   }
-   ```
-
-5. **Create ProgressReporter.ts**
-   - CLI progress bars for long-running operations
-   - ETA calculation
-   - Memory usage monitoring
-
-### Phase 3: Generator Script
-
-6. **Create generate-databases.ts**
-   ```bash
-   # Generate all tiers
-   bun tests/benchmark/generate-databases.ts --all
-
-   # Generate specific tier
-   bun tests/benchmark/generate-databases.ts --tier=md
-
-   # Regenerate (delete existing + recreate)
-   bun tests/benchmark/generate-databases.ts --tier=sm --force
-
-   # Custom seed for reproducibility
-   bun tests/benchmark/generate-databases.ts --tier=xs --seed=42
-   ```
-
-7. **Generation Process**
-   ```
-   For each tier:
-   1. Check if database directory exists (skip if exists, unless --force)
-   2. Create PGlite with dataDir
-   3. Run database migrations (PrepareDatabase)
-   4. Register components
-   5. Seed entities in batches with progress
-   6. Create relation foreign keys
-   7. Create indexes (CreateRelationIndexes)
-   8. Run VACUUM ANALYZE
-   9. Report final stats (entity count, db size, time taken)
-   ```
-
-### Phase 4: Benchmark Runner
-
-8. **Create benchmark loading utility**
-   ```typescript
-   // tests/benchmark/runners/loadBenchmarkDb.ts
-   export async function loadBenchmarkDatabase(
-     tier: 'xs' | 'sm' | 'md' | 'lg' | 'xl'
-   ): Promise<{ pg: PGlite; server: PGLiteSocketServer; cleanup: () => Promise<void> }>;
-   ```
-
-9. **Create QueryBenchmark.ts**
-   ```typescript
-   interface BenchmarkResult {
-     name: string;
-     tier: string;
-     entityCount: number;
-     queryTimeMs: number;
-     rowsReturned: number;
-     memoryUsedMB: number;
-   }
-
-   // Benchmark scenarios
-   - Simple filter (single component, indexed field)
-   - Multi-component filter (2-3 components with AND)
-   - Complex query (multi-component + sort + limit)
-   - BelongsTo resolution (N users with their orders)
-   - HasMany resolution (N products with their reviews)
-   - Deep nesting (Order -> User -> Reviews -> Products)
-   ```
-
-### Phase 5: npm Scripts
-
-10. **Add to package.json**
-    ```json
-    {
-      "scripts": {
-        "bench:generate": "bun tests/benchmark/generate-databases.ts --all",
-        "bench:generate:xs": "bun tests/benchmark/generate-databases.ts --tier=xs",
-        "bench:generate:sm": "bun tests/benchmark/generate-databases.ts --tier=sm",
-        "bench:generate:md": "bun tests/benchmark/generate-databases.ts --tier=md",
-        "bench:generate:lg": "bun tests/benchmark/generate-databases.ts --tier=lg",
-        "bench:generate:xl": "bun tests/benchmark/generate-databases.ts --tier=xl",
-        "bench:query": "bun tests/benchmark/runners/QueryBenchmark.ts",
-        "bench:relations": "bun tests/benchmark/runners/RelationBenchmark.ts"
-      }
-    }
-    ```
-
-## Data Generation Details
-
-### Realistic Distribution
-
-```typescript
-// Product categories follow power-law distribution
-// ~40% in top 3 categories, ~60% in remaining 7
-const categoryWeights = {
-  'Electronics': 0.20,
-  'Clothing': 0.12,
-  'Home & Garden': 0.08,
-  // ... rest evenly distributed
-};
-
-// Order counts per user follow exponential distribution
-// Most users: 1-5 orders, some power users: 50+ orders
-function getOrderCountForUser(): number {
-  const r = Math.random();
-  if (r < 0.7) return randomInt(1, 5);
-  if (r < 0.9) return randomInt(6, 20);
-  return randomInt(21, 100);
-}
-
-// Reviews are skewed toward extreme ratings (1, 4, 5 stars)
-function getReviewRating(): number {
-  const r = Math.random();
-  if (r < 0.05) return 1;      // 5% - 1 star
-  if (r < 0.10) return 2;      // 5% - 2 stars
-  if (r < 0.20) return 3;      // 10% - 3 stars
-  if (r < 0.50) return 4;      // 30% - 4 stars
-  return 5;                     // 50% - 5 stars
-}
-```
-
-### Batch Insertion Strategy
-
-```typescript
-// Avoid memory issues with large datasets
-const BATCH_SIZE = 1000;
-
-async function seedInBatches<T>(
-  total: number,
-  generator: (batchIndex: number, batchSize: number) => Promise<T[]>,
-  inserter: (items: T[]) => Promise<void>
-) {
-  const batches = Math.ceil(total / BATCH_SIZE);
-  for (let i = 0; i < batches; i++) {
-    const items = await generator(i, Math.min(BATCH_SIZE, total - i * BATCH_SIZE));
-    await inserter(items);
-    // Allow GC between batches
-    if (i % 10 === 0) await Bun.sleep(1);
-  }
-}
-```
-
-## Benchmark Scenarios
-
-### Query Performance Matrix
-
-| Scenario | Description | Expected Behavior |
-|----------|-------------|-------------------|
-| Q1 | Filter by indexed field | O(log n), <50ms at 1M |
-| Q2 | Filter by non-indexed field | O(n), slower |
-| Q3 | Multi-component AND | Uses INTERSECT optimization |
-| Q4 | Multi-component + Sort | Uses scalar subquery sort |
-| Q5 | BelongsTo single | DataLoader batched |
-| Q6 | HasMany collection | Single filtered query |
-| Q7 | Nested relations 3-deep | DataLoader + batching |
-| Q8 | Pagination (offset) | Cursor recommended at scale |
-| Q9 | Count aggregation | Fast with proper indexes |
-| Q10 | Full table scan | Baseline comparison |
-
-### Expected Performance Targets
-
-| Tier | Q1 (indexed) | Q3 (multi) | Q5 (BelongsTo) | Q7 (nested) |
-|------|--------------|------------|----------------|-------------|
-| xs   | <10ms        | <20ms      | <30ms          | <50ms       |
-| sm   | <15ms        | <40ms      | <50ms          | <100ms      |
-| md   | <20ms        | <60ms      | <80ms          | <150ms      |
-| lg   | <30ms        | <100ms     | <150ms         | <300ms      |
-| xl   | <50ms        | <150ms     | <250ms         | <500ms      |
-
-## Files to Create
-
-1. `tests/benchmark/fixtures/BenchmarkComponents.ts`
-2. `tests/benchmark/fixtures/BenchmarkArchetypes.ts`
-3. `tests/benchmark/generators/DataGenerator.ts`
-4. `tests/benchmark/generators/EntitySeeder.ts`
-5. `tests/benchmark/generators/ProgressReporter.ts`
-6. `tests/benchmark/generators/index.ts`
-7. `tests/benchmark/generate-databases.ts`
-8. `tests/benchmark/runners/loadBenchmarkDb.ts`
-9. `tests/benchmark/runners/QueryBenchmark.ts`
-10. `tests/benchmark/runners/RelationBenchmark.ts`
-
-## Gitignore Addition
-
-```gitignore
-# Benchmark databases (large, generated locally)
-tests/benchmark/databases/
-```
-
-## Usage Example
-
-```bash
-# First time: Generate benchmark databases
-bun run bench:generate:xs   # Quick: ~30s
-bun run bench:generate:md   # Medium: ~5min
-bun run bench:generate:xl   # Large: ~30min
-
-# Run benchmarks (uses pre-generated databases)
-bun run bench:query --tier=md
-bun run bench:relations --tier=lg
-
-# Results output
-# ┌─────────────────────────────────────────────────────────┐
-# │ Query Benchmark Results - Tier: md (100,000 entities)  │
-# ├──────────────────────┬─────────┬──────────┬────────────┤
-# │ Scenario             │ Time    │ Rows     │ Memory     │
-# ├──────────────────────┼─────────┼──────────┼────────────┤
-# │ Q1: Indexed filter   │ 18ms    │ 1,234    │ 2.3 MB     │
-# │ Q3: Multi-component  │ 52ms    │ 456      │ 4.1 MB     │
-# │ Q5: BelongsTo batch  │ 71ms    │ 100      │ 5.8 MB     │
-# │ Q7: Nested 3-deep    │ 142ms   │ 50       │ 12.4 MB    │
-# └──────────────────────┴─────────┴──────────┴────────────┘
-```
-
-## Next Steps
-
-1. Review and approve this plan
-2. Create the directory structure
-3. Implement Phase 1-2 (components + seeder)
-4. Implement Phase 3 (generator script)
-5. Generate xs/sm tiers for initial testing
-6. Implement Phase 4 (benchmark runners)
-7. Generate remaining tiers
-8. Document results and establish baselines

package/tests/benchmark/bunfig.toml

@@ -1,9 +0,0 @@
-[test]
-# No preload for benchmark tests - we manage the database connection ourselves
-preload = []
-
-# Longer timeout for benchmarks
-timeout = 300000
-
-# Single-threaded to avoid connection conflicts
-smol = true