Files
openclaw/src/memory/index.test.ts
2026-02-15 00:26:41 +00:00

382 lines
12 KiB
TypeScript

import fs from "node:fs/promises";
import os from "node:os";
import path from "node:path";
import { afterAll, beforeAll, beforeEach, describe, expect, it, vi } from "vitest";
import { getMemorySearchManager, type MemoryIndexManager } from "./index.js";
let embedBatchCalls = 0;
// Unit tests: avoid importing the real chokidar implementation (native fsevents, etc.).
vi.mock("chokidar", () => ({
default: {
watch: () => ({ on: () => {}, close: async () => {} }),
},
watch: () => ({ on: () => {}, close: async () => {} }),
}));
vi.mock("./sqlite-vec.js", () => ({
loadSqliteVecExtension: async () => ({ ok: false, error: "sqlite-vec disabled in tests" }),
}));
vi.mock("./embeddings.js", () => {
const embedText = (text: string) => {
const lower = text.toLowerCase();
const alpha = lower.split("alpha").length - 1;
const beta = lower.split("beta").length - 1;
return [alpha, beta];
};
return {
createEmbeddingProvider: async (options: { model?: string }) => ({
requestedProvider: "openai",
provider: {
id: "mock",
model: options.model ?? "mock-embed",
embedQuery: async (text: string) => embedText(text),
embedBatch: async (texts: string[]) => {
embedBatchCalls += 1;
return texts.map(embedText);
},
},
}),
};
});
describe("memory index", () => {
let fixtureRoot = "";
let workspaceDir = "";
let memoryDir = "";
let extraDir = "";
let indexBasicPath = "";
let indexCachePath = "";
let indexHybridPath = "";
let indexVectorPath = "";
let indexExtraPath = "";
// Perf: keep managers open across tests, but only reset the one a test uses.
const managersByStorePath = new Map<string, MemoryIndexManager>();
const managersForCleanup = new Set<MemoryIndexManager>();
beforeAll(async () => {
fixtureRoot = await fs.mkdtemp(path.join(os.tmpdir(), "openclaw-mem-fixtures-"));
workspaceDir = path.join(fixtureRoot, "workspace");
memoryDir = path.join(workspaceDir, "memory");
extraDir = path.join(workspaceDir, "extra");
indexBasicPath = path.join(workspaceDir, "index-basic.sqlite");
indexCachePath = path.join(workspaceDir, "index-cache.sqlite");
indexHybridPath = path.join(workspaceDir, "index-hybrid.sqlite");
indexVectorPath = path.join(workspaceDir, "index-vector.sqlite");
indexExtraPath = path.join(workspaceDir, "index-extra.sqlite");
await fs.mkdir(memoryDir, { recursive: true });
});
afterAll(async () => {
await Promise.all(Array.from(managersForCleanup).map((manager) => manager.close()));
await fs.rm(fixtureRoot, { recursive: true, force: true });
});
beforeEach(async () => {
// Perf: most suites don't need atomic swap behavior for full reindexes.
// Keep atomic reindex tests on the safe path.
vi.stubEnv("OPENCLAW_TEST_MEMORY_UNSAFE_REINDEX", "1");
embedBatchCalls = 0;
// Keep the workspace stable to allow manager reuse across tests.
await fs.mkdir(memoryDir, { recursive: true });
await fs.writeFile(
path.join(memoryDir, "2026-01-12.md"),
"# Log\nAlpha memory line.\nZebra memory line.",
);
// Clean additional paths that may have been created by earlier cases.
await fs.rm(extraDir, { recursive: true, force: true });
});
function resetManagerForTest(manager: MemoryIndexManager) {
// These tests reuse managers for performance. Clear the index + embedding
// cache to keep each test fully isolated.
(manager as unknown as { resetIndex: () => void }).resetIndex();
(manager as unknown as { db: { exec: (sql: string) => void } }).db.exec(
"DELETE FROM embedding_cache",
);
(manager as unknown as { dirty: boolean }).dirty = true;
(manager as unknown as { sessionsDirty: boolean }).sessionsDirty = false;
}
type TestCfg = Parameters<typeof getMemorySearchManager>[0]["cfg"];
async function getPersistentManager(cfg: TestCfg): Promise<MemoryIndexManager> {
const storePath = cfg.agents?.defaults?.memorySearch?.store?.path;
if (!storePath) {
throw new Error("store path missing");
}
const cached = managersByStorePath.get(storePath);
if (cached) {
resetManagerForTest(cached);
return cached;
}
const result = await getMemorySearchManager({ cfg, agentId: "main" });
expect(result.manager).not.toBeNull();
if (!result.manager) {
throw new Error("manager missing");
}
const manager = result.manager as MemoryIndexManager;
managersByStorePath.set(storePath, manager);
managersForCleanup.add(manager);
resetManagerForTest(manager);
return manager;
}
it("indexes memory files and searches by vector", async () => {
const cfg = {
agents: {
defaults: {
workspace: workspaceDir,
memorySearch: {
provider: "openai",
model: "mock-embed",
store: { path: indexBasicPath, vector: { enabled: false } },
sync: { watch: false, onSessionStart: false, onSearch: true },
query: { minScore: 0, hybrid: { enabled: false } },
},
},
list: [{ id: "main", default: true }],
},
};
const manager = await getPersistentManager(cfg);
await manager.sync({ reason: "test" });
const results = await manager.search("alpha");
expect(results.length).toBeGreaterThan(0);
expect(results[0]?.path).toContain("memory/2026-01-12.md");
const status = manager.status();
expect(status.sourceCounts).toEqual(
expect.arrayContaining([
expect.objectContaining({
source: "memory",
files: status.files,
chunks: status.chunks,
}),
]),
);
});
it("reindexes when the embedding model changes", async () => {
const indexModelPath = path.join(workspaceDir, "index-model-change.sqlite");
await fs.rm(indexModelPath, { force: true });
await fs.rm(`${indexModelPath}-shm`, { force: true });
await fs.rm(`${indexModelPath}-wal`, { force: true });
const base = {
agents: {
defaults: {
workspace: workspaceDir,
memorySearch: {
provider: "openai",
store: { path: indexModelPath },
sync: { watch: false, onSessionStart: false, onSearch: true },
query: { minScore: 0, hybrid: { enabled: false } },
},
},
list: [{ id: "main", default: true }],
},
};
const first = await getMemorySearchManager({
cfg: {
...base,
agents: {
...base.agents,
defaults: {
...base.agents.defaults,
memorySearch: {
...base.agents.defaults.memorySearch,
model: "mock-embed-v1",
},
},
},
},
agentId: "main",
});
expect(first.manager).not.toBeNull();
if (!first.manager) {
throw new Error("manager missing");
}
await first.manager.sync({ reason: "test" });
const callsAfterFirstSync = embedBatchCalls;
await first.manager.close();
const second = await getMemorySearchManager({
cfg: {
...base,
agents: {
...base.agents,
defaults: {
...base.agents.defaults,
memorySearch: {
...base.agents.defaults.memorySearch,
model: "mock-embed-v2",
},
},
},
},
agentId: "main",
});
expect(second.manager).not.toBeNull();
if (!second.manager) {
throw new Error("manager missing");
}
await second.manager.sync({ reason: "test" });
expect(embedBatchCalls).toBeGreaterThan(callsAfterFirstSync);
const status = second.manager.status();
expect(status.files).toBeGreaterThan(0);
await second.manager.close();
});
it("reuses cached embeddings on forced reindex", async () => {
const cfg = {
agents: {
defaults: {
workspace: workspaceDir,
memorySearch: {
provider: "openai",
model: "mock-embed",
store: { path: indexCachePath, vector: { enabled: false } },
sync: { watch: false, onSessionStart: false, onSearch: false },
query: { minScore: 0, hybrid: { enabled: false } },
cache: { enabled: true },
},
},
list: [{ id: "main", default: true }],
},
};
const manager = await getPersistentManager(cfg);
await manager.sync({ force: true });
const afterFirst = embedBatchCalls;
expect(afterFirst).toBeGreaterThan(0);
await manager.sync({ force: true });
expect(embedBatchCalls).toBe(afterFirst);
});
it("finds keyword matches via hybrid search when query embedding is zero", async () => {
const cfg = {
agents: {
defaults: {
workspace: workspaceDir,
memorySearch: {
provider: "openai",
model: "mock-embed",
store: { path: indexHybridPath, vector: { enabled: false } },
sync: { watch: false, onSessionStart: false, onSearch: true },
query: {
minScore: 0,
hybrid: { enabled: true, vectorWeight: 0, textWeight: 1 },
},
},
},
list: [{ id: "main", default: true }],
},
};
const manager = await getPersistentManager(cfg);
const status = manager.status();
if (!status.fts?.available) {
return;
}
await manager.sync({ reason: "test" });
const results = await manager.search("zebra");
expect(results.length).toBeGreaterThan(0);
expect(results[0]?.path).toContain("memory/2026-01-12.md");
});
it("reports vector availability after probe", async () => {
const cfg = {
agents: {
defaults: {
workspace: workspaceDir,
memorySearch: {
provider: "openai",
model: "mock-embed",
store: { path: indexVectorPath, vector: { enabled: true } },
sync: { watch: false, onSessionStart: false, onSearch: false },
},
},
list: [{ id: "main", default: true }],
},
};
const manager = await getPersistentManager(cfg);
const available = await manager.probeVectorAvailability();
const status = manager.status();
expect(status.vector?.enabled).toBe(true);
expect(typeof status.vector?.available).toBe("boolean");
expect(status.vector?.available).toBe(available);
});
it("rejects reading non-memory paths", async () => {
const cfg = {
agents: {
defaults: {
workspace: workspaceDir,
memorySearch: {
provider: "openai",
model: "mock-embed",
store: { path: indexBasicPath, vector: { enabled: false } },
sync: { watch: false, onSessionStart: false, onSearch: true },
query: { minScore: 0, hybrid: { enabled: false } },
},
},
list: [{ id: "main", default: true }],
},
};
const manager = await getPersistentManager(cfg);
await expect(manager.readFile({ relPath: "NOTES.md" })).rejects.toThrow("path required");
});
it("allows reading from additional memory paths and blocks symlinks", async () => {
await fs.mkdir(extraDir, { recursive: true });
await fs.writeFile(path.join(extraDir, "extra.md"), "Extra content.");
const cfg = {
agents: {
defaults: {
workspace: workspaceDir,
memorySearch: {
provider: "openai",
model: "mock-embed",
store: { path: indexExtraPath, vector: { enabled: false } },
sync: { watch: false, onSessionStart: false, onSearch: true },
query: { minScore: 0, hybrid: { enabled: false } },
extraPaths: [extraDir],
},
},
list: [{ id: "main", default: true }],
},
};
const manager = await getPersistentManager(cfg);
await expect(manager.readFile({ relPath: "extra/extra.md" })).resolves.toEqual({
path: "extra/extra.md",
text: "Extra content.",
});
const linkPath = path.join(extraDir, "linked.md");
let symlinkOk = true;
try {
await fs.symlink(path.join(extraDir, "extra.md"), linkPath, "file");
} catch (err) {
const code = (err as NodeJS.ErrnoException).code;
if (code === "EPERM" || code === "EACCES") {
symlinkOk = false;
} else {
throw err;
}
}
if (symlinkOk) {
await expect(manager.readFile({ relPath: "extra/linked.md" })).rejects.toThrow(
"path required",
);
}
});
});