Compare commits

...

4 Commits

Author SHA1 Message Date
xiaoju 404ee3e34f feat(daemon): add CAS blob store with sha256 content-addressable storage — closes #39
- createBlobStore(root) with write/read/exists API
- sha256 hex, first 2 chars as shard directory
- Atomic writes via temp file + rename
- CAS mismatch detection on read and write
- Inject blobStore into sense compute via options.blobs
- Export createBlobStore, normalizeBlobHash, BlobStore type
2026-04-23 00:19:35 +00:00
xiaomo cbc6db6b7d Merge pull request 'feat(daemon): log store archival — Meta table + JSONL cold archive (closes #38)' (#45) from feat/log-archive into main 2026-04-23 00:17:54 +00:00
xiaomo b1f6c775ce Merge pull request 'fix(init): auto-verify and retry better-sqlite3 native build — closes #44' (#50) from fix/init-sqlite-retry into main 2026-04-23 00:14:30 +00:00
xiaoju 978b1680a3 feat(daemon): add log store archival with meta watermark + JSONL cold archive — closes #38
- Add meta table with archived_up_to watermark in logs.db
- Archive logs older than 30 days to data/archive/logs/YYYY-MM-DD.jsonl
- Idempotent: same-day re-export overwrites file
- Single transaction: DELETE + UPDATE meta
- Optional VACUUM after archive loop
- CLI: nerve store archive [--vacuum]
- 15+ new tests for archive logic
2026-04-23 00:10:20 +00:00
19 changed files with 785 additions and 10 deletions
@@ -4,15 +4,21 @@
*/
import type {
ArchiveLogsDayResult as DaemonArchiveLogsDayResult,
ArchiveLogsOptions as DaemonArchiveLogsOptions,
ArchiveLogsResult as DaemonArchiveLogsResult,
LogEntry as DaemonLogEntry,
LogQuery as DaemonLogQuery,
LogStore as DaemonLogStore,
WorkflowRun as DaemonWorkflowRun,
WorkflowRunStatus as DaemonWorkflowRunStatus,
} from "@uncaged/nerve-daemon";
import { describe, it, expectTypeOf } from "vitest";
import { describe, expectTypeOf, it } from "vitest";
import type {
ArchiveLogsDayResult,
ArchiveLogsOptions,
ArchiveLogsResult,
LogEntry,
LogQuery,
LogStore,
@@ -42,6 +48,26 @@ describe("daemon-types drift guard", () => {
});
it("LogStore has all required methods", () => {
expectTypeOf<LogStore>().toMatchTypeOf<Pick<DaemonLogStore, "query" | "getWorkflowRun" | "getActiveWorkflowRuns" | "getAllWorkflowRuns" | "upsertWorkflowRun" | "close">>();
expectTypeOf<LogStore>().toMatchTypeOf<
Pick<
DaemonLogStore,
| "query"
| "getWorkflowRun"
| "getActiveWorkflowRuns"
| "getAllWorkflowRuns"
| "upsertWorkflowRun"
| "archiveLogs"
| "close"
>
>();
});
it("ArchiveLogs types match daemon", () => {
expectTypeOf<ArchiveLogsOptions>().toMatchTypeOf<DaemonArchiveLogsOptions>();
expectTypeOf<DaemonArchiveLogsOptions>().toMatchTypeOf<ArchiveLogsOptions>();
expectTypeOf<ArchiveLogsResult>().toMatchTypeOf<DaemonArchiveLogsResult>();
expectTypeOf<DaemonArchiveLogsResult>().toMatchTypeOf<ArchiveLogsResult>();
expectTypeOf<ArchiveLogsDayResult>().toMatchTypeOf<DaemonArchiveLogsDayResult>();
expectTypeOf<DaemonArchiveLogsDayResult>().toMatchTypeOf<ArchiveLogsDayResult>();
});
});
+2
View File
@@ -6,6 +6,7 @@ import { senseCommand } from "./commands/sense.js";
import { startCommand } from "./commands/start.js";
import { statusCommand } from "./commands/status.js";
import { stopCommand } from "./commands/stop.js";
import { storeCommand } from "./commands/store.js";
import { validateCommand } from "./commands/validate.js";
import { workflowCommand } from "./commands/workflow.js";
@@ -22,6 +23,7 @@ const main = defineCommand({
logs: logsCommand,
validate: validateCommand,
sense: senseCommand,
store: storeCommand,
workflow: workflowCommand,
},
});
+70
View File
@@ -0,0 +1,70 @@
import { existsSync } from "node:fs";
import { join } from "node:path";
import { defineCommand } from "citty";
import { loadDaemonModule } from "../workspace-daemon.js";
import { getNerveRoot } from "../workspace.js";
// ---------------------------------------------------------------------------
// nerve store archive
// ---------------------------------------------------------------------------
const storeArchiveCommand = defineCommand({
meta: {
name: "archive",
description:
"Export logs older than 30 days from logs.db to data/archive/logs/YYYY-MM-DD.jsonl and delete those rows (RFC-001 §5.4)",
},
args: {
vacuum: {
type: "boolean",
description: "Run SQLite VACUUM after archiving",
default: false,
},
},
async run({ args }) {
const nerveRoot = getNerveRoot();
const dbPath = join(nerveRoot, "data", "logs.db");
if (!existsSync(dbPath)) {
process.stderr.write("❌ No data/logs.db found — start the daemon at least once.\n");
process.exit(1);
}
const { createLogStore } = await loadDaemonModule(nerveRoot);
const store = createLogStore(dbPath);
try {
const result = store.archiveLogs({ vacuum: args.vacuum });
if (result.days.length === 0) {
process.stdout.write(
"✅ Nothing to archive (no eligible UTC days beyond the 30-day window).\n",
);
} else {
process.stdout.write(`✅ Archived ${result.days.length} day(s):\n`);
for (const d of result.days) {
process.stdout.write(` ${d.day} rows=${d.rowCount} ${d.filePath}\n`);
}
}
if (result.vacuumed) {
process.stdout.write(" VACUUM completed.\n");
}
} finally {
store.close();
}
},
});
// ---------------------------------------------------------------------------
// nerve store
// ---------------------------------------------------------------------------
export const storeCommand = defineCommand({
meta: {
name: "store",
description: "Maintain local Nerve SQLite stores (log cold-archive, …)",
},
subCommands: {
archive: storeArchiveCommand,
},
});
+19
View File
@@ -40,6 +40,24 @@ export type LogQuery = {
limit?: number;
};
export type ArchiveLogsOptions = {
now?: number;
vacuum?: boolean;
maxDays?: number;
retentionMs?: number;
};
export type ArchiveLogsDayResult = {
day: string;
rowCount: number;
filePath: string;
};
export type ArchiveLogsResult = {
days: ArchiveLogsDayResult[];
vacuumed: boolean;
};
/** Subset of daemon LogStore used by the CLI workflow commands. */
export type LogStore = {
query: (filter?: LogQuery) => LogEntry[];
@@ -47,5 +65,6 @@ export type LogStore = {
getActiveWorkflowRuns: (workflowName?: string) => WorkflowRun[];
getAllWorkflowRuns: (workflowName: string | null) => WorkflowRun[];
upsertWorkflowRun: (entry: Omit<LogEntry, "id">, run: WorkflowRun) => LogEntry;
archiveLogs: (options?: ArchiveLogsOptions) => ArchiveLogsResult;
close: () => void;
};
@@ -0,0 +1,105 @@
import { createHash } from "node:crypto";
import { existsSync, readdirSync, writeFileSync } from "node:fs";
import { tmpdir } from "node:os";
import { join } from "node:path";
import { describe, expect, it } from "vitest";
import { createBlobStore, normalizeBlobHash } from "../blob-store.js";
function makeRoot(): string {
return join(tmpdir(), `nerve-blob-${Date.now()}-${Math.random().toString(16).slice(2)}`);
}
describe("normalizeBlobHash", () => {
it("accepts 64-char lowercase hex", () => {
const h = "a".repeat(64);
expect(normalizeBlobHash(h)).toBe(h);
});
it("normalizes uppercase to lowercase", () => {
const h = "A".repeat(64);
expect(normalizeBlobHash(h)).toBe("a".repeat(64));
});
it("rejects wrong length and non-hex", () => {
expect(normalizeBlobHash("ab")).toBeNull();
expect(normalizeBlobHash("g".repeat(64))).toBeNull();
});
});
describe("createBlobStore", () => {
it("write returns sha256 hex and stores under 2-char shard", () => {
const root = makeRoot();
const store = createBlobStore(root);
const content = "hello cas";
const hash = store.write(content);
expect(hash).toMatch(/^[0-9a-f]{64}$/);
expect(createHash("sha256").update(content, "utf8").digest("hex")).toBe(hash);
const shard = hash.slice(0, 2);
const rel = hash.slice(2);
const filePath = join(root, shard, rel);
expect(existsSync(filePath)).toBe(true);
});
it("read returns stored bytes and exists is true", () => {
const root = makeRoot();
const store = createBlobStore(root);
const buf = Buffer.from([0, 255, 128]);
const hash = store.write(buf);
expect(store.exists(hash)).toBe(true);
const got = store.read(hash);
expect(got).not.toBeNull();
expect(Buffer.compare(got as Buffer, buf)).toBe(0);
});
it("write is idempotent for same content", () => {
const root = makeRoot();
const store = createBlobStore(root);
const h1 = store.write("same");
const h2 = store.write("same");
expect(h1).toBe(h2);
const shard = h1.slice(0, 2);
const names = readdirSync(join(root, shard));
expect(names.filter((n: string) => !n.startsWith("."))).toHaveLength(1);
});
it("read returns null for missing blob", () => {
const root = makeRoot();
const store = createBlobStore(root);
const missing = "0".repeat(64);
expect(store.read(missing)).toBeNull();
expect(store.exists(missing)).toBe(false);
});
it("read and exists return null/false for invalid hash", () => {
const root = makeRoot();
const store = createBlobStore(root);
expect(store.read("not-a-hash")).toBeNull();
expect(store.exists("not-a-hash")).toBe(false);
});
it("throws when on-disk content does not match path hash", () => {
const root = makeRoot();
const store = createBlobStore(root);
const hash = store.write("ok");
const filePath = join(root, hash.slice(0, 2), hash.slice(2));
writeFileSync(filePath, "tampered");
expect(() => store.read(hash)).toThrow(/CAS mismatch/i);
});
it("write throws when an existing file at the digest path has wrong content", () => {
const root = makeRoot();
const store = createBlobStore(root);
const hash = store.write("truth");
const filePath = join(root, hash.slice(0, 2), hash.slice(2));
writeFileSync(filePath, "lies");
expect(() => store.write("truth")).toThrow(/CAS mismatch/i);
});
});
@@ -91,6 +91,7 @@ function makeLogStore(
}),
getTriggerPayload: vi.fn(() => ({ value: 42 })),
getThreadEvents: vi.fn(() => [{ type: "thread_start", triggerPayload: {} }]),
archiveLogs: vi.fn(() => ({ days: [], vacuumed: false })),
close: vi.fn(),
};
return store;
@@ -77,6 +77,7 @@ function makeLogStore() {
getActiveWorkflowRuns: vi.fn(() => []),
getTriggerPayload: vi.fn(() => null),
getThreadEvents: vi.fn(() => []),
archiveLogs: vi.fn(() => ({ days: [], vacuumed: false })),
close: vi.fn(),
};
}
@@ -74,6 +74,7 @@ function makeMockLogStore() {
getAllWorkflowRuns: vi.fn(() => []),
getTriggerPayload: vi.fn(() => null),
getThreadEvents: vi.fn(() => []),
archiveLogs: vi.fn(() => ({ days: [], vacuumed: false })),
close: vi.fn(),
};
}
@@ -80,6 +80,7 @@ function makeLogStore() {
getActiveWorkflowRuns: vi.fn(() => []),
getTriggerPayload: vi.fn(() => null),
getThreadEvents: vi.fn(() => []),
archiveLogs: vi.fn(() => ({ days: [], vacuumed: false })),
close: vi.fn(),
};
}
@@ -0,0 +1,40 @@
import { describe, expect, it } from "vitest";
import {
assertValidUtcDay,
compareIsoDays,
lastArchivableUtcDay,
nextUtcDay,
prevUtcDay,
utcDateStringFromMs,
utcDayEndExclusiveMs,
utcDayStartMs,
} from "../log-archive.js";
describe("log-archive UTC helpers", () => {
it("lastArchivableUtcDay matches RFC-style boundary (exclusive end of day ≤ boundary)", () => {
const boundary = Date.UTC(2026, 1, 2, 12, 0, 0); // 2026-02-02 12:00 UTC
expect(lastArchivableUtcDay(boundary)).toBe("2026-02-01");
});
it("round-trips UTC day bounds", () => {
expect(utcDayStartMs("2026-02-01")).toBe(Date.UTC(2026, 1, 1));
expect(utcDayEndExclusiveMs("2026-02-01")).toBe(Date.UTC(2026, 1, 2));
expect(utcDateStringFromMs(Date.UTC(2026, 1, 1, 23, 59))).toBe("2026-02-01");
});
it("nextUtcDay / prevUtcDay", () => {
expect(nextUtcDay("2026-02-01")).toBe("2026-02-02");
expect(prevUtcDay("2026-02-01")).toBe("2026-01-31");
});
it("compareIsoDays sorts lexicographically for YYYY-MM-DD", () => {
expect(compareIsoDays("2026-01-01", "2026-02-01")).toBeLessThan(0);
expect(compareIsoDays("2026-02-01", "2026-02-01")).toBe(0);
});
it("assertValidUtcDay rejects invalid calendars", () => {
expect(() => assertValidUtcDay("2026-02-31")).toThrow();
expect(() => assertValidUtcDay("bad")).toThrow();
});
});
@@ -0,0 +1,139 @@
import { mkdtempSync, readFileSync, rmSync } from "node:fs";
import { tmpdir } from "node:os";
import { join } from "node:path";
import { afterEach, beforeEach, describe, expect, it } from "vitest";
import { LOG_ARCHIVE_META_KEY, createLogStore } from "../log-store.js";
import type { LogStore } from "../log-store.js";
const DAY_MS = 86_400_000;
/** `now` such that 2026-02-01 is the last archivable UTC day under a 30-day window. */
function nowForLastArchivableFeb1(): number {
const boundary = Date.UTC(2026, 1, 2, 12, 0, 0);
return boundary + 30 * DAY_MS;
}
describe("LogStore — cold archive (RFC-001 §5.4)", () => {
let tmpDir: string;
let store: LogStore;
beforeEach(() => {
tmpDir = mkdtempSync(join(tmpdir(), "nerve-archive-"));
store = createLogStore(join(tmpDir, "data", "logs.db"));
});
afterEach(() => {
store.close();
rmSync(tmpDir, { recursive: true, force: true });
});
it("exports one UTC day to JSONL, deletes rows, advances archived_up_to", () => {
const ts = Date.UTC(2026, 1, 1, 10, 0, 0);
store.append({ source: "system", type: "x", refId: null, payload: '{"a":1}', ts });
store.append({ source: "reflex", type: "y", refId: "z", payload: null, ts: ts + 1 });
const now = nowForLastArchivableFeb1();
const result = store.archiveLogs({ now, retentionMs: 30 * DAY_MS });
expect(result.days).toHaveLength(1);
expect(result.days[0].day).toBe("2026-02-01");
expect(result.days[0].rowCount).toBe(2);
const jsonlPath = join(tmpDir, "data", "archive", "logs", "2026-02-01.jsonl");
expect(result.days[0].filePath).toBe(jsonlPath);
const lines = readFileSync(jsonlPath, "utf8").trim().split("\n");
expect(lines).toHaveLength(2);
const o = JSON.parse(lines[0] ?? "{}") as Record<string, unknown>;
expect(o.source).toBe("system");
expect(o.refId).toBeNull();
expect(store.query()).toHaveLength(0);
expect(store.getMeta(LOG_ARCHIVE_META_KEY)).toBe("2026-02-01");
});
it("returns nothing for an empty logs table", () => {
const r = store.archiveLogs({ now: nowForLastArchivableFeb1(), retentionMs: 30 * DAY_MS });
expect(r.days).toHaveLength(0);
expect(store.getMeta(LOG_ARCHIVE_META_KEY)).toBeNull();
});
it("does nothing when all logs are inside the hot window", () => {
const now = Date.UTC(2026, 3, 23, 12, 0, 0);
const ts = now - 5 * DAY_MS;
store.append({ source: "system", type: "warm", refId: null, payload: null, ts });
const r = store.archiveLogs({ now, retentionMs: 30 * DAY_MS });
expect(r.days).toHaveLength(0);
expect(store.query()).toHaveLength(1);
});
it("second archive with same clock is a no-op (watermark already caught up)", () => {
const ts = Date.UTC(2026, 1, 1, 10, 0, 0);
store.append({ source: "system", type: "x", refId: null, payload: null, ts });
const now = nowForLastArchivableFeb1();
store.archiveLogs({ now, retentionMs: 30 * DAY_MS });
const path = join(tmpDir, "data", "archive", "logs", "2026-02-01.jsonl");
const first = readFileSync(path, "utf8");
const second = store.archiveLogs({ now, retentionMs: 30 * DAY_MS });
expect(second.days).toHaveLength(0);
expect(readFileSync(path, "utf8")).toBe(first);
});
it("overwrites JSONL when the same UTC day is archived again after watermark rewind", () => {
const ts = Date.UTC(2026, 1, 1, 10, 0, 0);
store.append({ source: "a", type: "1", refId: null, payload: null, ts });
const now = nowForLastArchivableFeb1();
store.archiveLogs({ now, retentionMs: 30 * DAY_MS });
store.setMeta(LOG_ARCHIVE_META_KEY, "2026-01-31");
store.append({ source: "b", type: "2", refId: null, payload: null, ts: ts + 100 });
store.archiveLogs({ now, retentionMs: 30 * DAY_MS });
const path = join(tmpDir, "data", "archive", "logs", "2026-02-01.jsonl");
const lines = readFileSync(path, "utf8").trim().split("\n");
expect(lines).toHaveLength(1);
expect(JSON.parse(lines[0] ?? "{}").source).toBe("b");
});
it("respects maxDays across invocations", () => {
const t1 = Date.UTC(2026, 1, 1, 10, 0, 0);
const t2 = Date.UTC(2026, 1, 2, 10, 0, 0);
store.append({ source: "system", type: "a", refId: null, payload: null, ts: t1 });
store.append({ source: "system", type: "b", refId: null, payload: null, ts: t2 });
const now = Date.UTC(2027, 0, 1, 12, 0, 0);
const r1 = store.archiveLogs({ now, retentionMs: 30 * DAY_MS, maxDays: 1 });
expect(r1.days).toHaveLength(1);
expect(r1.days[0].day).toBe("2026-02-01");
const r2 = store.archiveLogs({ now, retentionMs: 30 * DAY_MS, maxDays: 1 });
expect(r2.days).toHaveLength(1);
expect(r2.days[0].day).toBe("2026-02-02");
expect(store.getMeta(LOG_ARCHIVE_META_KEY)).toBe("2026-02-02");
expect(store.query()).toHaveLength(0);
});
it("starts from earliest log day when it is before watermark+1", () => {
store.setMeta(LOG_ARCHIVE_META_KEY, "2026-01-10");
const ts = Date.UTC(2026, 1, 1, 10, 0, 0);
store.append({ source: "x", type: "p", refId: null, payload: null, ts });
const result = store.archiveLogs({ now: nowForLastArchivableFeb1(), retentionMs: 30 * DAY_MS });
expect(result.days.map((d) => d.day)).toContain("2026-02-01");
});
it("throws on invalid archived_up_to watermark", () => {
store.setMeta(LOG_ARCHIVE_META_KEY, "not-a-date");
expect(() => store.archiveLogs({ now: Date.now() })).toThrow();
});
it("runs VACUUM when vacuum: true", () => {
const ts = Date.UTC(2026, 1, 1, 10, 0, 0);
store.append({ source: "system", type: "x", refId: null, payload: null, ts });
const r = store.archiveLogs({
now: nowForLastArchivableFeb1(),
retentionMs: 30 * DAY_MS,
vacuum: true,
});
expect(r.vacuumed).toBe(true);
});
});
@@ -7,6 +7,7 @@ import { drizzle } from "drizzle-orm/better-sqlite3";
import { integer, real, sqliteTable } from "drizzle-orm/sqlite-core";
import { describe, expect, it } from "vitest";
import { createBlobStore } from "../blob-store.js";
import { parseParentMessage } from "../ipc.js";
import { executeCompute, openPeerDb, openSenseDb, runMigrations } from "../sense-runtime.js";
import type { DrizzleDB, PeerMap, SenseRuntime } from "../sense-runtime.js";
@@ -340,6 +341,20 @@ describe("executeCompute", () => {
expect(capturedSignal).toBeInstanceOf(AbortSignal);
sqlite.close();
});
it("passes BlobStore as options.blobs when blobStore argument is provided", async () => {
const blobsRoot = mkdtempSync(join(tmpdir(), "nerve-blobs-"));
const blobStore = createBlobStore(blobsRoot);
let seen: ReturnType<typeof createBlobStore> | undefined;
const { runtime, sqlite } = makeRuntime(async (_db, _peers, options) => {
seen = options?.blobs;
return null;
});
await executeCompute(runtime, emptyPeers, undefined, blobStore);
expect(seen).toBe(blobStore);
sqlite.close();
});
});
// ---------------------------------------------------------------------------
@@ -74,6 +74,7 @@ function makeLogStore() {
getActiveWorkflowRuns: vi.fn(() => []),
getTriggerPayload: vi.fn(() => null),
getThreadEvents: vi.fn(() => []),
archiveLogs: vi.fn(() => ({ days: [], vacuumed: false })),
close: vi.fn(),
};
}
+106
View File
@@ -0,0 +1,106 @@
/**
* CAS blob store sha256 content-addressable files under `data/blobs/`.
*
* Layout: `<root>/<2-hex-shard>/<62-hex-rest>` (RFC-001 §8).
*/
import { createHash, randomBytes } from "node:crypto";
import {
existsSync,
mkdirSync,
readFileSync,
renameSync,
unlinkSync,
writeFileSync,
} from "node:fs";
import { dirname, join } from "node:path";
const SHA256_HEX_LEN = 64;
const HEX_RE = /^[0-9a-f]+$/;
export type BlobStore = {
/** Persist UTF-8 or raw bytes; returns lowercase hex sha256. Idempotent for identical content. */
write: (content: string | Uint8Array | Buffer) => string;
/** Returns bytes or null if the hash is invalid or no blob exists. Verifies digest matches path. */
read: (hash: string) => Buffer | null;
/** True when hash is well-formed and the blob file is present. */
exists: (hash: string) => boolean;
};
function toBuffer(content: string | Uint8Array | Buffer): Buffer {
if (typeof content === "string") return Buffer.from(content, "utf8");
if (Buffer.isBuffer(content)) return content;
return Buffer.from(content);
}
function digestHex(buf: Buffer): string {
return createHash("sha256").update(buf).digest("hex");
}
/** @returns normalized lowercase hex or null if not a valid sha256 hex string */
export function normalizeBlobHash(hash: string): string | null {
const h = hash.trim().toLowerCase();
if (h.length !== SHA256_HEX_LEN) return null;
if (!HEX_RE.test(h)) return null;
return h;
}
function pathForHash(blobsRoot: string, hashLower: string): string {
return join(blobsRoot, hashLower.slice(0, 2), hashLower.slice(2));
}
function verifyPathMatchesContent(filePath: string, expectedHash: string): Buffer {
const data = readFileSync(filePath);
const actual = digestHex(data);
if (actual !== expectedHash) {
throw new Error(
`Blob CAS mismatch at "${filePath}": file digests to ${actual}, path expects ${expectedHash}`,
);
}
return data;
}
export function createBlobStore(blobsRoot: string): BlobStore {
function write(content: string | Uint8Array | Buffer): string {
const buf = toBuffer(content);
const hash = digestHex(buf);
const filePath = pathForHash(blobsRoot, hash);
if (existsSync(filePath)) {
verifyPathMatchesContent(filePath, hash);
return hash;
}
mkdirSync(dirname(filePath), { recursive: true });
const tmp = join(dirname(filePath), `.tmp.${randomBytes(16).toString("hex")}`);
try {
writeFileSync(tmp, buf);
renameSync(tmp, filePath);
} catch (e) {
try {
unlinkSync(tmp);
} catch {
// ignore cleanup errors
}
throw e;
}
return hash;
}
function read(hash: string): Buffer | null {
const h = normalizeBlobHash(hash);
if (h === null) return null;
const filePath = pathForHash(blobsRoot, h);
if (!existsSync(filePath)) return null;
return verifyPathMatchesContent(filePath, h);
}
function exists(hash: string): boolean {
const h = normalizeBlobHash(hash);
if (h === null) return false;
return existsSync(pathForHash(blobsRoot, h));
}
return { write, read, exists };
}
+7 -1
View File
@@ -32,13 +32,19 @@ export type { Kernel, KernelOptions, KernelHealth } from "./kernel.js";
export { createFileWatcher } from "./file-watcher.js";
export type { FileWatcher, FileChange, FileChangeHandler } from "./file-watcher.js";
export { createLogStore } from "./log-store.js";
export { createBlobStore, normalizeBlobHash } from "./blob-store.js";
export type { BlobStore } from "./blob-store.js";
export { createLogStore, LOG_ARCHIVE_META_KEY } from "./log-store.js";
export type {
LogStore,
LogEntry,
LogQuery,
WorkflowRun,
WorkflowRunStatus,
ArchiveLogsDayResult,
ArchiveLogsOptions,
ArchiveLogsResult,
} from "./log-store.js";
export { createWorkflowManager } from "./workflow-manager.js";
+78
View File
@@ -0,0 +1,78 @@
/** Log cold-archive helpers (RFC-001 §5.4) — UTC calendar days, JSONL export. */
export const LOG_ARCHIVE_META_KEY = "archived_up_to";
export const DEFAULT_LOG_RETENTION_MS = 30 * 86_400_000;
export type ArchiveLogsOptions = {
/** Wall clock for retention boundary (default: `Date.now()`). */
now?: number;
/** Run `VACUUM` after archiving (outside the per-day transaction). */
vacuum?: boolean;
/** Max UTC days to process in one call (default: unlimited). */
maxDays?: number;
/** Override default 30-day retention (tests). */
retentionMs?: number;
};
export type ArchiveLogsDayResult = {
day: string;
rowCount: number;
filePath: string;
};
export type ArchiveLogsResult = {
days: ArchiveLogsDayResult[];
vacuumed: boolean;
};
export function utcDateStringFromMs(ms: number): string {
return new Date(ms).toISOString().slice(0, 10);
}
function parseUtcDayParts(day: string): [number, number, number] {
const m = /^(\d{4})-(\d{2})-(\d{2})$/.exec(day);
if (m === null) {
throw new Error(`Invalid UTC day (expected YYYY-MM-DD): ${day}`);
}
const y = Number(m[1]);
const mo = Number(m[2]);
const d = Number(m[3]);
const t = Date.UTC(y, mo - 1, d);
if (utcDateStringFromMs(t) !== day) {
throw new Error(`Invalid UTC calendar day: ${day}`);
}
return [y, mo, d];
}
export function assertValidUtcDay(day: string): void {
parseUtcDayParts(day);
}
export function utcDayStartMs(day: string): number {
const [y, mo, d] = parseUtcDayParts(day);
return Date.UTC(y, mo - 1, d);
}
export function utcDayEndExclusiveMs(day: string): number {
return utcDayStartMs(day) + 86_400_000;
}
export function prevUtcDay(day: string): string {
return utcDateStringFromMs(utcDayStartMs(day) - 86_400_000);
}
export function nextUtcDay(day: string): string {
return utcDateStringFromMs(utcDayEndExclusiveMs(day));
}
/** Last UTC calendar day D such that the exclusive end of D is ≤ boundaryMs. */
export function lastArchivableUtcDay(boundaryMs: number): string {
return prevUtcDay(utcDateStringFromMs(boundaryMs));
}
export function compareIsoDays(a: string, b: string): number {
if (a < b) return -1;
if (a > b) return 1;
return 0;
}
+151 -2
View File
@@ -7,11 +7,27 @@
* Also provides a `meta` key-value table for bookkeeping (e.g. archive watermarks).
*/
import { mkdirSync } from "node:fs";
import { dirname } from "node:path";
import { mkdirSync, writeFileSync } from "node:fs";
import { dirname, join } from "node:path";
import Database from "better-sqlite3";
import type BetterSqlite3 from "better-sqlite3";
import {
DEFAULT_LOG_RETENTION_MS,
LOG_ARCHIVE_META_KEY,
assertValidUtcDay,
compareIsoDays,
lastArchivableUtcDay,
nextUtcDay,
utcDateStringFromMs,
utcDayEndExclusiveMs,
utcDayStartMs,
} from "./log-archive.js";
import type { ArchiveLogsDayResult, ArchiveLogsOptions, ArchiveLogsResult } from "./log-archive.js";
export { LOG_ARCHIVE_META_KEY } from "./log-archive.js";
export type { ArchiveLogsDayResult, ArchiveLogsOptions, ArchiveLogsResult } from "./log-archive.js";
export type LogEntry = {
id?: number;
source: string;
@@ -105,6 +121,12 @@ export type LogStore = {
* Used for crash recovery to rebuild ThreadState.
*/
getThreadEvents: (runId: string) => Array<{ type: string; [key: string]: unknown }>;
/**
* Export logs older than the retention window to `data/archive/logs/YYYY-MM-DD.jsonl`,
* then delete those rows and advance `meta.archived_up_to` in one transaction per day
* (RFC-001 §5.4).
*/
archiveLogs: (options?: ArchiveLogsOptions) => ArchiveLogsResult;
close: () => void;
};
@@ -138,6 +160,78 @@ CREATE INDEX IF NOT EXISTS idx_workflow_runs_status ON workflow_runs(status);
CREATE INDEX IF NOT EXISTS idx_workflow_runs_workflow ON workflow_runs(workflow);
`;
type SqlLogRow = {
id: number;
source: string;
type: string;
ref_id: string | null;
payload: string | null;
ts: number;
};
function buildJsonlBody(rows: SqlLogRow[]): string {
if (rows.length === 0) return "";
const lines = rows.map((r) =>
JSON.stringify({
id: r.id,
source: r.source,
type: r.type,
refId: r.ref_id,
payload: r.payload,
ts: r.ts,
}),
);
return `${lines.join("\n")}\n`;
}
function runOptionalVacuum(sqlite: BetterSqlite3.Database, vacuum?: boolean): boolean {
if (vacuum !== true) return false;
sqlite.exec("VACUUM");
return true;
}
function resolveArchiveStartDay(watermark: string | null, minDay: string): string {
if (watermark === null) return minDay;
const afterWatermark = nextUtcDay(watermark);
return compareIsoDays(minDay, afterWatermark) < 0 ? minDay : afterWatermark;
}
function runArchiveDayLoop(
dbPath: string,
options: ArchiveLogsOptions,
selectLogsForDayStmt: BetterSqlite3.Statement,
archiveDayTx: (day: string, start: number, endExclusive: number) => void,
startDay: string,
lastDay: string,
): ArchiveLogsDayResult[] {
const archiveDir = join(dirname(dbPath), "archive", "logs");
mkdirSync(archiveDir, { recursive: true });
const days: ArchiveLogsDayResult[] = [];
let d = startDay;
let processed = 0;
while (compareIsoDays(d, lastDay) <= 0) {
if (options.maxDays !== undefined && processed >= options.maxDays) {
break;
}
const start = utcDayStartMs(d);
const endExclusive = utcDayEndExclusiveMs(d);
const rows = selectLogsForDayStmt.all({ start, endExclusive }) as SqlLogRow[];
const filePath = join(archiveDir, `${d}.jsonl`);
writeFileSync(filePath, buildJsonlBody(rows), "utf8");
archiveDayTx(d, start, endExclusive);
days.push({ day: d, rowCount: rows.length, filePath });
processed += 1;
d = nextUtcDay(d);
}
return days;
}
export function createLogStore(dbPath: string): LogStore {
mkdirSync(dirname(dbPath), { recursive: true });
@@ -186,6 +280,14 @@ export function createLogStore(dbPath: string): LogStore {
"SELECT run_id, workflow, status, ts FROM workflow_runs WHERE workflow = ? ORDER BY ts DESC",
);
const minLogTsStmt = sqlite.prepare("SELECT MIN(ts) AS m FROM logs");
const selectLogsForDayStmt = sqlite.prepare(
"SELECT id, source, type, ref_id, payload, ts FROM logs WHERE ts >= @start AND ts < @endExclusive ORDER BY id ASC",
);
const deleteLogsForDayStmt = sqlite.prepare(
"DELETE FROM logs WHERE ts >= @start AND ts < @endExclusive",
);
const upsertWorkflowRunTx = sqlite.transaction(
(entry: Omit<LogEntry, "id">, run: WorkflowRun) => {
const info = insertStmt.run({
@@ -358,6 +460,52 @@ export function createLogStore(dbPath: string): LogStore {
return result;
}
const archiveDayTx = sqlite.transaction((day: string, start: number, endExclusive: number) => {
deleteLogsForDayStmt.run({ start, endExclusive });
setMetaStmt.run({ key: LOG_ARCHIVE_META_KEY, value: day });
});
function readWatermark(): string | null {
const raw = getMeta(LOG_ARCHIVE_META_KEY);
if (raw === null) return null;
assertValidUtcDay(raw);
return raw;
}
function firstLogUtcDay(): string | null {
const row = minLogTsStmt.get() as { m: number | null } | undefined;
const m = row?.m;
if (m === null || m === undefined) return null;
return utcDateStringFromMs(m);
}
function archiveLogs(options: ArchiveLogsOptions = {}): ArchiveLogsResult {
const now = options.now ?? Date.now();
const retentionMs = options.retentionMs ?? DEFAULT_LOG_RETENTION_MS;
const lastDay = lastArchivableUtcDay(now - retentionMs);
const watermark = readWatermark();
const minDay = firstLogUtcDay();
if (minDay === null) {
return { days: [], vacuumed: runOptionalVacuum(sqlite, options.vacuum) };
}
const startDay = resolveArchiveStartDay(watermark, minDay);
if (compareIsoDays(startDay, lastDay) > 0) {
return { days: [], vacuumed: runOptionalVacuum(sqlite, options.vacuum) };
}
const days = runArchiveDayLoop(
dbPath,
options,
selectLogsForDayStmt,
archiveDayTx,
startDay,
lastDay,
);
return { days, vacuumed: runOptionalVacuum(sqlite, options.vacuum) };
}
function close(): void {
sqlite.close();
}
@@ -374,6 +522,7 @@ export function createLogStore(dbPath: string): LogStore {
getAllWorkflowRuns,
getTriggerPayload,
getThreadEvents,
archiveLogs,
close,
};
}
+12 -2
View File
@@ -8,6 +8,8 @@ import type { BetterSQLite3Database } from "drizzle-orm/better-sqlite3";
import type { Result } from "@uncaged/nerve-core";
import { err, ok } from "@uncaged/nerve-core";
import type { BlobStore } from "./blob-store.js";
/** A Drizzle DB instance (schema-generic) */
export type DrizzleDB = BetterSQLite3Database<Record<string, never>>;
@@ -17,11 +19,14 @@ export type PeerMap = Readonly<Record<string, DrizzleDB>>;
/** Options passed to a compute function */
export type ComputeOptions = {
signal: AbortSignal;
/** CAS under `data/blobs/`; injected by the sense worker when available. */
blobs?: BlobStore;
};
/**
* The shape every sense's index.ts must export.
* Engine injects `db` (read-write), `peers` (read-only), and `options`.
* Engine injects `db` (read-write), `peers` (read-only), and `options`
* (`signal`, and `blobs` when running in the sense worker RFC-001 §8 CAS).
* Returns T when a signal should be emitted, null for silence.
*/
export type ComputeFn<T = unknown> = (
@@ -192,14 +197,19 @@ export async function loadComputeFn(senseIndexPath: string): Promise<Result<Comp
* Execute a sense's compute function with an optional soft timeout.
* If timeoutMs is provided and compute takes longer, the AbortSignal is
* triggered and an error Result is returned.
* When `blobStore` is set, it is exposed as `options.blobs` (see RFC-001 §8).
*/
export async function executeCompute(
runtime: SenseRuntime,
peers: PeerMap,
timeoutMs?: number,
blobStore?: BlobStore,
): Promise<Result<unknown | null>> {
const controller = new AbortController();
const options: ComputeOptions = { signal: controller.signal };
const options: ComputeOptions =
blobStore !== undefined
? { signal: controller.signal, blobs: blobStore }
: { signal: controller.signal };
let timer: ReturnType<typeof setTimeout> | undefined;
const timeoutPromise =
+8 -3
View File
@@ -10,6 +10,7 @@
* senses/<name>/index.js compiled compute
* senses/<name>/migrations/ SQL migration files
* data/senses/<name>.db SQLite data file
* data/blobs/<aa>/<hashrest> CAS (sha256), via options.blobs in compute
* nerve.yaml config
*/
@@ -19,6 +20,7 @@ import { join, resolve } from "node:path";
import { parseNerveConfig } from "@uncaged/nerve-core";
import type { NerveConfig } from "@uncaged/nerve-core";
import { createBlobStore } from "./blob-store.js";
import type { WorkerToParentMessage } from "./ipc.js";
import { parseParentMessage } from "./ipc.js";
import { executeCompute, loadComputeFn, openPeerDb, openSenseDb } from "./sense-runtime.js";
@@ -162,9 +164,10 @@ async function runCompute(
peers: PeerMap,
timeoutMs: number,
gracePeriodMs: number | null,
blobStore: ReturnType<typeof createBlobStore>,
): Promise<void> {
try {
const result = await executeCompute(runtime, peers, timeoutMs);
const result = await executeCompute(runtime, peers, timeoutMs, blobStore);
if (!result.ok) {
sendError(senseName, result.error.message);
if (gracePeriodMs !== null && result.error.message.includes("timed out")) {
@@ -193,6 +196,7 @@ function handleMessage(
group: string,
senseConfigs: Map<string, { timeout: number | null; gracePeriod: number | null }>,
inFlight: Map<string, Promise<void>>,
blobStore: ReturnType<typeof createBlobStore>,
): void {
const parseResult = parseParentMessage(raw);
if (!parseResult.ok) {
@@ -230,7 +234,7 @@ function handleMessage(
const previous = inFlight.get(msg.sense) ?? Promise.resolve();
const next = previous
.then(() => runCompute(msg.sense, runtime, peers, timeoutMs, gracePeriodMs))
.then(() => runCompute(msg.sense, runtime, peers, timeoutMs, gracePeriodMs, blobStore))
.catch((e: unknown) => {
const errMsg = e instanceof Error ? e.message : String(e);
sendError(msg.sense, errMsg);
@@ -294,11 +298,12 @@ async function bootstrap(nerveRoot: string, group: string): Promise<void> {
}
const inFlight = new Map<string, Promise<void>>();
const blobStore = createBlobStore(join(nerveRoot, "data", "blobs"));
sendReady();
process.on("message", (raw: unknown) => {
handleMessage(raw, runtimes, peers, group, senseConfigs, inFlight);
handleMessage(raw, runtimes, peers, group, senseConfigs, inFlight, blobStore);
});
}