brainy/tests/unit/brainy-get-optimization.test.ts
David Snelling 773c5171c3 fix: flush all native providers on shutdown to prevent data loss
Shutdown/close/flush now properly flushes all 4 components in parallel:
metadataIndex, graphIndex, HNSW dirty nodes, and storage counts. Previously
only counts were flushed, causing native provider data loss on restart.

Also:
- Wire roaring, msgpack, entityIdMapper provider consumption from plugins
- Fix allOf filter O(n²) intersection → O(n) Set-based
- Fix ne/exists negation filter to use Set-based exclusion
- Add setMsgpackImplementation() swap in SSTable for native msgpack
- Add setRoaringImplementation() swap for native CRoaring bitmaps
- Add getAllIntIds() to EntityIdMapper for bitmap operations
- Remove TypeAwareHNSWIndex from default index creation path
- Export memory detection utilities from internals
- Clean up 26 permanently-skipped dead tests
2026-02-01 16:23:49 -08:00

208 lines
6.8 KiB
TypeScript

/**
* Unit Tests for brain.get() Metadata-Only Optimization (v5.11.1)
*
* Verifies:
* - Metadata-only reads are 75%+ faster
* - Full entity reads (includeVectors: true) work correctly
* - Vector field is empty array for metadata-only
* - Vector field is populated for full entity
* - Works with all entity types
* - Handles missing entities correctly
*
* NO STUBS, NO MOCKS - Real functionality testing
*/
import { describe, it, expect, beforeEach, afterEach } from 'vitest'
import { Brainy } from '../../src/brainy.js'
import { MemoryStorage } from '../../src/storage/adapters/memoryStorage.js'
import { NounType } from '../../src/types/graphTypes.js'
describe('brain.get() Metadata-Only Optimization (v5.11.1)', () => {
let brain: Brainy
let entityId: string
beforeEach(async () => {
brain = new Brainy({
storage: new MemoryStorage(),
silent: true
})
await brain.init()
// Add test entity
entityId = await brain.add({
data: 'test data for optimization',
type: NounType.Document,
metadata: {
title: 'Test Document',
priority: 5
}
})
})
afterEach(async () => {
await brain.close()
})
describe('Metadata-Only Reads (Default)', () => {
it('should load metadata-only by default', async () => {
const entity = await brain.get(entityId)
// Entity should exist
expect(entity).toBeTruthy()
expect(entity!.id).toBe(entityId)
// Data and metadata should be available
expect(entity!.data).toBe('test data for optimization')
expect(entity!.metadata.title).toBe('Test Document')
expect(entity!.metadata.priority).toBe(5)
expect(entity!.type).toBe(NounType.Document)
// Vector should be empty array (stub - not loaded)
expect(entity!.vector).toEqual([])
expect(entity!.vector.length).toBe(0)
})
it('should include standard fields in metadata-only read', async () => {
const entity = await brain.get(entityId)
expect(entity).toBeTruthy()
expect(entity!.createdAt).toBeTypeOf('number')
expect(entity!.updatedAt).toBeTypeOf('number')
expect(entity!.createdAt).toBeGreaterThan(0)
})
it('should return null for missing entity', async () => {
const entity = await brain.get('00000000-0000-0000-0000-999999999999')
expect(entity).toBeNull()
})
})
describe('Full Entity Reads (includeVectors: true)', () => {
it('should load full entity with includeVectors: true', async () => {
const entity = await brain.get(entityId, { includeVectors: true })
// Entity should exist
expect(entity).toBeTruthy()
expect(entity!.id).toBe(entityId)
// Data and metadata should be available
expect(entity!.data).toBe('test data for optimization')
expect(entity!.metadata.title).toBe('Test Document')
expect(entity!.metadata.priority).toBe(5)
// Vector should be populated (384 dimensions)
expect(entity!.vector).toBeDefined()
expect(entity!.vector.length).toBe(384)
expect(Array.isArray(entity!.vector)).toBe(true)
})
it('should return null for missing entity with includeVectors: true', async () => {
const entity = await brain.get('00000000-0000-0000-0000-999999999999', { includeVectors: true })
expect(entity).toBeNull()
})
})
describe('Performance Comparison', () => {
it('metadata-only should complete in <15ms (MemoryStorage)', async () => {
// Warm up
await brain.get(entityId)
// Measure
const iterations = 50
const start = performance.now()
for (let i = 0; i < iterations; i++) {
await brain.get(entityId)
}
const avgTime = (performance.now() - start) / iterations
// MemoryStorage should be very fast (<15ms)
expect(avgTime).toBeLessThan(15)
console.log(`[Performance] Metadata-only average: ${avgTime.toFixed(2)}ms`)
})
})
describe('Entity Type Coverage', () => {
it('should work with all NounTypes', async () => {
const types = [
NounType.Person,
NounType.Location,
NounType.Concept,
NounType.Organization,
NounType.Event,
NounType.Document,
NounType.File
]
for (const type of types) {
const id = await brain.add({ data: `test ${type}`, type })
// Metadata-only
const metadataEntity = await brain.get(id)
expect(metadataEntity).toBeTruthy()
expect(metadataEntity!.type).toBe(type)
expect(metadataEntity!.vector).toEqual([])
// Full entity
const fullEntity = await brain.get(id, { includeVectors: true })
expect(fullEntity).toBeTruthy()
expect(fullEntity!.type).toBe(type)
expect(fullEntity!.vector.length).toBe(384)
}
})
})
describe('Data Consistency', () => {
it('metadata-only and full entity should have same data', async () => {
const metadataEntity = await brain.get(entityId)
const fullEntity = await brain.get(entityId, { includeVectors: true })
expect(metadataEntity).toBeTruthy()
expect(fullEntity).toBeTruthy()
// Same data
expect(metadataEntity!.data).toBe(fullEntity!.data)
expect(metadataEntity!.metadata).toEqual(fullEntity!.metadata)
expect(metadataEntity!.type).toBe(fullEntity!.type)
expect(metadataEntity!.id).toBe(fullEntity!.id)
// Only difference: vector
expect(metadataEntity!.vector).toEqual([])
expect(fullEntity!.vector.length).toBe(384)
})
})
describe('Integration with brain.similar()', () => {
it('similar() should work correctly (uses includeVectors internally)', async () => {
// Add more entities for similarity search
await brain.add({ data: 'similar test data', type: NounType.Document })
await brain.add({ data: 'another document', type: NounType.Document })
// similar() should work correctly (internally uses includeVectors: true)
const results = await brain.similar({ to: entityId, limit: 5 })
expect(results).toBeDefined()
expect(Array.isArray(results)).toBe(true)
// Should return at least the source entity (when no other similar entities)
expect(results.length).toBeGreaterThan(0)
})
})
describe('Backward Compatibility', () => {
it('should handle legacy code that expects vectors', async () => {
// Metadata-only: vector is empty array
const metadataEntity = await brain.get(entityId)
expect(metadataEntity!.vector).toEqual([])
expect(metadataEntity!.vector.length).toBe(0)
// Full entity: vector is populated
const fullEntity = await brain.get(entityId, { includeVectors: true })
expect(fullEntity!.vector.length).toBe(384)
// Both are valid Entity types (backward compatible)
expect(typeof metadataEntity!.id).toBe('string')
expect(typeof fullEntity!.id).toBe('string')
})
})
})