feat(memory): implement keyword search adapter — no vector dependency

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Jarvis
2026-04-02 20:55:00 -05:00
parent 25383ea645
commit 35fbd88a1d
6 changed files with 508 additions and 51 deletions

View File

@@ -18,6 +18,7 @@
},
"dependencies": {
"@mosaic/db": "workspace:*",
"@mosaic/storage": "workspace:*",
"@mosaic/types": "workspace:*",
"drizzle-orm": "^0.45.1"
},

View File

@@ -0,0 +1,298 @@
import { describe, it, expect, beforeEach } from 'vitest';
import type { StorageAdapter } from '@mosaic/storage';
import { KeywordAdapter } from './keyword.js';
/* ------------------------------------------------------------------ */
/* In-memory mock StorageAdapter */
/* ------------------------------------------------------------------ */
function createMockStorage(): StorageAdapter {
const collections = new Map<string, Map<string, Record<string, unknown>>>();
let idCounter = 0;
function getCollection(name: string): Map<string, Record<string, unknown>> {
if (!collections.has(name)) collections.set(name, new Map());
return collections.get(name)!;
}
const adapter: StorageAdapter = {
name: 'mock',
async create<T extends Record<string, unknown>>(
collection: string,
data: T,
): Promise<T & { id: string }> {
const id = String(++idCounter);
const record = { ...data, id };
getCollection(collection).set(id, record);
return record as T & { id: string };
},
async read<T extends Record<string, unknown>>(
collection: string,
id: string,
): Promise<T | null> {
const record = getCollection(collection).get(id);
return (record as T) ?? null;
},
async update(collection: string, id: string, data: Record<string, unknown>): Promise<boolean> {
const col = getCollection(collection);
const existing = col.get(id);
if (!existing) return false;
col.set(id, { ...existing, ...data });
return true;
},
async delete(collection: string, id: string): Promise<boolean> {
return getCollection(collection).delete(id);
},
async find<T extends Record<string, unknown>>(
collection: string,
filter?: Record<string, unknown>,
): Promise<T[]> {
const col = getCollection(collection);
const results: T[] = [];
for (const record of col.values()) {
if (filter && !matchesFilter(record, filter)) continue;
results.push(record as T);
}
return results;
},
async findOne<T extends Record<string, unknown>>(
collection: string,
filter: Record<string, unknown>,
): Promise<T | null> {
const col = getCollection(collection);
for (const record of col.values()) {
if (matchesFilter(record, filter)) return record as T;
}
return null;
},
async count(collection: string, filter?: Record<string, unknown>): Promise<number> {
const rows = await adapter.find(collection, filter);
return rows.length;
},
async transaction<T>(fn: (tx: StorageAdapter) => Promise<T>): Promise<T> {
return fn(adapter);
},
async migrate(): Promise<void> {},
async close(): Promise<void> {},
};
return adapter;
}
function matchesFilter(record: Record<string, unknown>, filter: Record<string, unknown>): boolean {
for (const [key, value] of Object.entries(filter)) {
if (record[key] !== value) return false;
}
return true;
}
/* ------------------------------------------------------------------ */
/* Tests */
/* ------------------------------------------------------------------ */
describe('KeywordAdapter', () => {
let adapter: KeywordAdapter;
beforeEach(() => {
adapter = new KeywordAdapter({ type: 'keyword', storage: createMockStorage() });
});
/* ---- Preferences ---- */
describe('preferences', () => {
it('should set and get a preference', async () => {
await adapter.setPreference('u1', 'theme', 'dark');
const value = await adapter.getPreference('u1', 'theme');
expect(value).toBe('dark');
});
it('should return null for missing preference', async () => {
const value = await adapter.getPreference('u1', 'nonexistent');
expect(value).toBeNull();
});
it('should upsert an existing preference', async () => {
await adapter.setPreference('u1', 'theme', 'dark');
await adapter.setPreference('u1', 'theme', 'light');
const value = await adapter.getPreference('u1', 'theme');
expect(value).toBe('light');
});
it('should delete a preference', async () => {
await adapter.setPreference('u1', 'theme', 'dark');
const deleted = await adapter.deletePreference('u1', 'theme');
expect(deleted).toBe(true);
const value = await adapter.getPreference('u1', 'theme');
expect(value).toBeNull();
});
it('should return false when deleting nonexistent preference', async () => {
const deleted = await adapter.deletePreference('u1', 'nope');
expect(deleted).toBe(false);
});
it('should list preferences by userId', async () => {
await adapter.setPreference('u1', 'theme', 'dark', 'appearance');
await adapter.setPreference('u1', 'lang', 'en', 'locale');
await adapter.setPreference('u2', 'theme', 'light', 'appearance');
const prefs = await adapter.listPreferences('u1');
expect(prefs).toHaveLength(2);
expect(prefs.map((p) => p.key).sort()).toEqual(['lang', 'theme']);
});
it('should filter preferences by category', async () => {
await adapter.setPreference('u1', 'theme', 'dark', 'appearance');
await adapter.setPreference('u1', 'lang', 'en', 'locale');
const prefs = await adapter.listPreferences('u1', 'appearance');
expect(prefs).toHaveLength(1);
expect(prefs[0]!.key).toBe('theme');
});
});
/* ---- Insights ---- */
describe('insights', () => {
it('should store and retrieve an insight', async () => {
const insight = await adapter.storeInsight({
userId: 'u1',
content: 'TypeScript is great for type safety',
source: 'chat',
category: 'technical',
relevanceScore: 0.9,
});
expect(insight.id).toBeDefined();
expect(insight.content).toBe('TypeScript is great for type safety');
const fetched = await adapter.getInsight(insight.id);
expect(fetched).not.toBeNull();
expect(fetched!.content).toBe('TypeScript is great for type safety');
});
it('should return null for missing insight', async () => {
const result = await adapter.getInsight('nonexistent');
expect(result).toBeNull();
});
it('should delete an insight', async () => {
const insight = await adapter.storeInsight({
userId: 'u1',
content: 'test',
source: 'chat',
category: 'general',
relevanceScore: 0.5,
});
const deleted = await adapter.deleteInsight(insight.id);
expect(deleted).toBe(true);
const fetched = await adapter.getInsight(insight.id);
expect(fetched).toBeNull();
});
});
/* ---- Keyword Search ---- */
describe('searchInsights', () => {
beforeEach(async () => {
await adapter.storeInsight({
userId: 'u1',
content: 'TypeScript provides excellent type safety for JavaScript projects',
source: 'chat',
category: 'technical',
relevanceScore: 0.9,
});
await adapter.storeInsight({
userId: 'u1',
content: 'React hooks simplify state management in components',
source: 'chat',
category: 'technical',
relevanceScore: 0.8,
});
await adapter.storeInsight({
userId: 'u1',
content: 'TypeScript and React work great together for type safe components',
source: 'chat',
category: 'technical',
relevanceScore: 0.85,
});
await adapter.storeInsight({
userId: 'u2',
content: 'TypeScript is popular',
source: 'chat',
category: 'general',
relevanceScore: 0.5,
});
});
it('should find insights by exact keyword', async () => {
const results = await adapter.searchInsights('u1', 'hooks');
expect(results).toHaveLength(1);
expect(results[0]!.content).toContain('hooks');
});
it('should be case-insensitive', async () => {
const results = await adapter.searchInsights('u1', 'TYPESCRIPT');
expect(results.length).toBeGreaterThanOrEqual(1);
for (const r of results) {
expect(r.content.toLowerCase()).toContain('typescript');
}
});
it('should rank multi-word matches higher', async () => {
const results = await adapter.searchInsights('u1', 'TypeScript React');
// The insight mentioning both "TypeScript" and "React" should rank first (score=2)
expect(results[0]!.score).toBe(2);
expect(results[0]!.content).toContain('TypeScript');
expect(results[0]!.content).toContain('React');
});
it('should return empty for no matches', async () => {
const results = await adapter.searchInsights('u1', 'python django');
expect(results).toHaveLength(0);
});
it('should filter by userId', async () => {
const results = await adapter.searchInsights('u2', 'TypeScript');
expect(results).toHaveLength(1);
expect(results[0]!.content).toBe('TypeScript is popular');
});
it('should respect limit option', async () => {
const results = await adapter.searchInsights('u1', 'TypeScript', { limit: 1 });
expect(results).toHaveLength(1);
});
it('should return empty for empty query', async () => {
const results = await adapter.searchInsights('u1', ' ');
expect(results).toHaveLength(0);
});
});
/* ---- Lifecycle ---- */
describe('lifecycle', () => {
it('should have name "keyword"', () => {
expect(adapter.name).toBe('keyword');
});
it('should have null embedder', () => {
expect(adapter.embedder).toBeNull();
});
it('should close without error', async () => {
await expect(adapter.close()).resolves.toBeUndefined();
});
});
});

View File

@@ -0,0 +1,195 @@
import type { StorageAdapter } from '@mosaic/storage';
import type {
MemoryAdapter,
MemoryConfig,
NewInsight,
Insight,
InsightSearchResult,
} from '../types.js';
import type { EmbeddingProvider } from '../vector-store.js';
type KeywordConfig = Extract<MemoryConfig, { type: 'keyword' }>;
const PREFERENCES = 'preferences';
const INSIGHTS = 'insights';
type PreferenceRecord = Record<string, unknown> & {
id: string;
userId: string;
key: string;
value: unknown;
category: string;
};
type InsightRecord = Record<string, unknown> & {
id: string;
userId: string;
content: string;
source: string;
category: string;
relevanceScore: number;
metadata: Record<string, unknown>;
createdAt: string;
updatedAt?: string;
decayedAt?: string;
};
export class KeywordAdapter implements MemoryAdapter {
readonly name = 'keyword';
readonly embedder: EmbeddingProvider | null = null;
private storage: StorageAdapter;
constructor(config: KeywordConfig) {
this.storage = config.storage;
}
/* ------------------------------------------------------------------ */
/* Preferences */
/* ------------------------------------------------------------------ */
async getPreference(userId: string, key: string): Promise<unknown | null> {
const row = await this.storage.findOne<PreferenceRecord>(PREFERENCES, { userId, key });
return row?.value ?? null;
}
async setPreference(
userId: string,
key: string,
value: unknown,
category?: string,
): Promise<void> {
const existing = await this.storage.findOne<PreferenceRecord>(PREFERENCES, { userId, key });
if (existing) {
await this.storage.update(PREFERENCES, existing.id, {
value,
...(category !== undefined ? { category } : {}),
});
} else {
await this.storage.create(PREFERENCES, {
userId,
key,
value,
category: category ?? 'general',
});
}
}
async deletePreference(userId: string, key: string): Promise<boolean> {
const existing = await this.storage.findOne<PreferenceRecord>(PREFERENCES, { userId, key });
if (!existing) return false;
return this.storage.delete(PREFERENCES, existing.id);
}
async listPreferences(
userId: string,
category?: string,
): Promise<Array<{ key: string; value: unknown; category: string }>> {
const filter: Record<string, unknown> = { userId };
if (category !== undefined) filter.category = category;
const rows = await this.storage.find<PreferenceRecord>(PREFERENCES, filter);
return rows.map((r) => ({ key: r.key, value: r.value, category: r.category }));
}
/* ------------------------------------------------------------------ */
/* Insights */
/* ------------------------------------------------------------------ */
async storeInsight(insight: NewInsight): Promise<Insight> {
const now = new Date();
const row = await this.storage.create<Record<string, unknown>>(INSIGHTS, {
userId: insight.userId,
content: insight.content,
source: insight.source,
category: insight.category,
relevanceScore: insight.relevanceScore,
metadata: insight.metadata ?? {},
createdAt: now.toISOString(),
});
return {
id: row.id,
userId: insight.userId,
content: insight.content,
source: insight.source,
category: insight.category,
relevanceScore: insight.relevanceScore,
metadata: insight.metadata,
createdAt: now,
};
}
async getInsight(id: string): Promise<Insight | null> {
const row = await this.storage.read<InsightRecord>(INSIGHTS, id);
if (!row) return null;
return toInsight(row);
}
async searchInsights(
userId: string,
query: string,
opts?: { limit?: number; embedding?: number[] },
): Promise<InsightSearchResult[]> {
const limit = opts?.limit ?? 10;
const words = query
.toLowerCase()
.split(/\s+/)
.filter((w) => w.length > 0);
if (words.length === 0) return [];
const rows = await this.storage.find<InsightRecord>(INSIGHTS, { userId });
const scored: InsightSearchResult[] = [];
for (const row of rows) {
const content = row.content.toLowerCase();
let score = 0;
for (const word of words) {
if (content.includes(word)) score++;
}
if (score > 0) {
scored.push({
id: row.id,
content: row.content,
score,
metadata: row.metadata ?? undefined,
});
}
}
scored.sort((a, b) => b.score - a.score);
return scored.slice(0, limit);
}
async deleteInsight(id: string): Promise<boolean> {
return this.storage.delete(INSIGHTS, id);
}
/* ------------------------------------------------------------------ */
/* Lifecycle */
/* ------------------------------------------------------------------ */
async close(): Promise<void> {
// no-op — storage adapter manages its own lifecycle
}
}
/* ------------------------------------------------------------------ */
/* Helpers */
/* ------------------------------------------------------------------ */
function toInsight(row: InsightRecord): Insight {
return {
id: row.id,
userId: row.userId,
content: row.content,
source: row.source,
category: row.category,
relevanceScore: row.relevanceScore,
metadata: row.metadata ?? undefined,
createdAt: new Date(row.createdAt),
updatedAt: row.updatedAt ? new Date(row.updatedAt) : undefined,
decayedAt: row.decayedAt ? new Date(row.decayedAt) : undefined,
};
}

View File

@@ -22,12 +22,18 @@ export type {
} from './types.js';
export { createMemoryAdapter, registerMemoryAdapter } from './factory.js';
export { PgVectorAdapter } from './adapters/pgvector.js';
export { KeywordAdapter } from './adapters/keyword.js';
// Auto-register pgvector adapter at module load time
// Auto-register adapters at module load time
import { registerMemoryAdapter } from './factory.js';
import { PgVectorAdapter } from './adapters/pgvector.js';
import { KeywordAdapter } from './adapters/keyword.js';
import type { MemoryConfig } from './types.js';
registerMemoryAdapter('pgvector', (config: MemoryConfig) => {
return new PgVectorAdapter(config as Extract<MemoryConfig, { type: 'pgvector' }>);
});
registerMemoryAdapter('keyword', (config: MemoryConfig) => {
return new KeywordAdapter(config as Extract<MemoryConfig, { type: 'keyword' }>);
});

View File

@@ -1,5 +1,6 @@
export type { EmbeddingProvider, VectorSearchResult } from './vector-store.js';
import type { EmbeddingProvider } from './vector-store.js';
import type { StorageAdapter } from '@mosaic/storage';
/* ------------------------------------------------------------------ */
/* Insight types (adapter-level, decoupled from Drizzle schema) */
@@ -69,4 +70,4 @@ export interface MemoryAdapter {
export type MemoryConfig =
| { type: 'pgvector'; embedder?: EmbeddingProvider }
| { type: 'sqlite-vec'; embedder?: EmbeddingProvider }
| { type: 'keyword' };
| { type: 'keyword'; storage: StorageAdapter };

54
pnpm-lock.yaml generated
View File

@@ -456,6 +456,9 @@ importers:
'@mosaic/db':
specifier: workspace:*
version: link:../db
'@mosaic/storage':
specifier: workspace:*
version: link:../storage
'@mosaic/types':
specifier: workspace:*
version: link:../types
@@ -634,10 +637,10 @@ importers:
dependencies:
'@mariozechner/pi-agent-core':
specifier: ^0.63.1
version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)
version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)
'@mariozechner/pi-ai':
specifier: ^0.63.1
version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)
version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)
'@sinclair/typebox':
specifier: ^0.34.41
version: 0.34.48
@@ -6999,12 +7002,6 @@ snapshots:
'@jridgewell/gen-mapping': 0.3.13
'@jridgewell/trace-mapping': 0.3.31
'@anthropic-ai/sdk@0.73.0(zod@3.25.76)':
dependencies:
json-schema-to-ts: 3.1.1
optionalDependencies:
zod: 3.25.76
'@anthropic-ai/sdk@0.73.0(zod@4.3.6)':
dependencies:
json-schema-to-ts: 3.1.1
@@ -8589,18 +8586,6 @@ snapshots:
- ws
- zod
'@mariozechner/pi-agent-core@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)':
dependencies:
'@mariozechner/pi-ai': 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)
transitivePeerDependencies:
- '@modelcontextprotocol/sdk'
- aws-crt
- bufferutil
- supports-color
- utf-8-validate
- ws
- zod
'@mariozechner/pi-agent-core@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)':
dependencies:
'@mariozechner/pi-ai': 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)
@@ -8661,30 +8646,6 @@ snapshots:
- ws
- zod
'@mariozechner/pi-ai@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)':
dependencies:
'@anthropic-ai/sdk': 0.73.0(zod@3.25.76)
'@aws-sdk/client-bedrock-runtime': 3.1008.0
'@google/genai': 1.45.0(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))
'@mistralai/mistralai': 1.14.1
'@sinclair/typebox': 0.34.48
ajv: 8.18.0
ajv-formats: 3.0.1(ajv@8.18.0)
chalk: 5.6.2
openai: 6.26.0(ws@8.20.0)(zod@3.25.76)
partial-json: 0.1.7
proxy-agent: 6.5.0
undici: 7.24.3
zod-to-json-schema: 3.25.1(zod@3.25.76)
transitivePeerDependencies:
- '@modelcontextprotocol/sdk'
- aws-crt
- bufferutil
- supports-color
- utf-8-validate
- ws
- zod
'@mariozechner/pi-ai@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)':
dependencies:
'@anthropic-ai/sdk': 0.73.0(zod@4.3.6)
@@ -12903,11 +12864,6 @@ snapshots:
dependencies:
mimic-function: 5.0.1
openai@6.26.0(ws@8.20.0)(zod@3.25.76):
optionalDependencies:
ws: 8.20.0
zod: 3.25.76
openai@6.26.0(ws@8.20.0)(zod@4.3.6):
optionalDependencies:
ws: 8.20.0