+ An article body with enough text for readability to choose it as the main content.
+ Another paragraph to satisfy the readability heuristic.
+ `;
+
+beforeEach(async () => {
+ ({ app, ownerHeaders } = await setup());
+ sp = await spaces.create({ slug: 'cap', name: 'Cap' }, { kind: 'user', id: null });
+ process.env.BLOB_ROOT = await fs.mkdtemp(path.join(os.tmpdir(), 'void-blobs-'));
+ await queue.start(); await registerWorkers();
+ global.fetch = vi.fn(async () => new Response(HTML, {
+ status: 200, headers: { 'content-type': 'text/html' }
+ }));
+});
+afterEach(async () => { await stopBoss(); vi.restoreAllMocks(); });
+
+describe('capture api', () => {
+ it('POST /api/capture enqueues ingest.url and returns 202', async () => {
+ const res = await request(app).post('/api/capture').set(ownerHeaders)
+ .send({ space_id: sp.id, url: 'https://example.com/a' });
+ expect(res.status).toBe(202);
+ expect(res.body.job_id).toBeTruthy();
+ expect(res.body.idempotency_key).toMatch(/^[0-9a-f]{64}$/);
+ });
+
+ it('POST /api/capture returns existing ref_id on duplicate', async () => {
+ const r1 = await request(app).post('/api/capture').set(ownerHeaders)
+ .send({ space_id: sp.id, url: 'https://example.com/dup' });
+ await waitForJob('ingest.url', r1.body.job_id, { timeoutMs: 10_000 });
+ const r2 = await request(app).post('/api/capture').set(ownerHeaders)
+ .send({ space_id: sp.id, url: 'https://example.com/dup' });
+ expect(r2.status).toBe(202);
+ expect(r2.body.job_id).toBeNull();
+ expect(r2.body.ref_id).toBeTruthy();
+ });
+
+ it('POST /api/capture/upload enqueues ingest.blob', async () => {
+ const res = await request(app).post('/api/capture/upload').set(ownerHeaders)
+ .field('space_id', sp.id)
+ .attach('file', Buffer.from('hi'), { filename: 'a.txt', contentType: 'text/plain' });
+ expect(res.status).toBe(202);
+ expect(res.body.job_id).toBeTruthy();
+ await waitForJob('ingest.blob', res.body.job_id, { timeoutMs: 10_000 });
+ const rows = await refs.list({ space_id: sp.id });
+ expect(rows[0].kind).toBe('file');
+ });
+
+ it('POST /api/capture rejects missing url', async () => {
+ const res = await request(app).post('/api/capture').set(ownerHeaders)
+ .send({ space_id: sp.id });
+ expect(res.status).toBe(400);
+ });
+
+ it('unauthenticated → 401', async () => {
+ const res = await request(app).post('/api/capture')
+ .send({ space_id: sp.id, url: 'https://example.com/a' });
+ expect(res.status).toBe(401);
+ });
+});
diff --git a/tests/ingest/safe_fetch.test.js b/tests/ingest/safe_fetch.test.js
new file mode 100644
index 0000000..0dec24f
--- /dev/null
+++ b/tests/ingest/safe_fetch.test.js
@@ -0,0 +1,34 @@
+import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
+import { safeFetch, SafeFetchError } from '../../lib/ingest/safe_fetch.js';
+
+beforeEach(() => { delete process.env.VOID_INGEST_ALLOW_PRIVATE; });
+afterEach(() => { vi.restoreAllMocks(); });
+
+describe('safeFetch', () => {
+ it('rejects file:// scheme', async () => {
+ await expect(safeFetch('file:///etc/passwd')).rejects.toThrow(SafeFetchError);
+ });
+
+ it('rejects literal loopback IP', async () => {
+ await expect(safeFetch('http://127.0.0.1/x')).rejects.toThrow(/blocked/);
+ });
+
+ it('rejects literal RFC1918 IP', async () => {
+ await expect(safeFetch('http://192.168.1.1/x')).rejects.toThrow(/blocked/);
+ });
+
+ it('rejects literal CGNAT IP', async () => {
+ await expect(safeFetch('http://100.64.0.1/x')).rejects.toThrow(/blocked/);
+ });
+
+ it('rejects AWS metadata literal IP', async () => {
+ await expect(safeFetch('http://169.254.169.254/latest/meta-data/')).rejects.toThrow(/blocked/);
+ });
+
+ it('allows literal IPs when VOID_INGEST_ALLOW_PRIVATE=true (test fixtures)', async () => {
+ process.env.VOID_INGEST_ALLOW_PRIVATE = 'true';
+ global.fetch = vi.fn(async () => new Response('ok', { status: 200 }));
+ const res = await safeFetch('http://127.0.0.1:65535/');
+ expect(res.status).toBe(200);
+ });
+});