curo1305
0d51d023ce
- Replace POST /api/documents/upload with POST /api/documents/upload-url + /{id}/confirm
- upload-url: create pending Document row with user_id=None (Wave 2), return presigned PUT URL
- confirm: stat MinIO for authoritative size (T-03-05), atomic quota UPDATE (T-03-06, STORE-03)
- Confirm returns 413 with {used_bytes, limit_bytes, rejected_bytes} on quota exceeded (STORE-05)
- Wave 2 guard: skip quota UPDATE when doc.user_id is None (Plan 03-03 removes this)
- Add GET /api/auth/me/quota to api/auth.py (STORE-04)
- services/storage.py: remove save_upload (D-04); add GREATEST(0, used_bytes-delta) quota decrement to delete_document (STORE-06)
- tasks/document_tasks.py: add cleanup_abandoned_uploads Celery beat task (D-06)
- celery_app.py: add beat_schedule for cleanup-abandoned-uploads every 30 minutes
- tests/test_documents.py: replace legacy /upload tests with xfail; add real test logic for upload-url/confirm/get-quota
- tests/test_quota.py: implement real test logic with xfail for PostgreSQL-specific SQL
295 lines
11 KiB
Python
295 lines
11 KiB
Python
"""
|
|
Document API tests — async only (Plan 05 cutover).
|
|
|
|
Legacy sync tests (using the flat-file storage layer) were deleted in Plan 05.
|
|
All tests here use async_client (httpx.AsyncClient + ASGITransport + in-memory SQLite).
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import re
|
|
|
|
import pytest
|
|
|
|
|
|
@pytest.mark.xfail(strict=False, reason="POST /api/documents/upload removed in Plan 03-02 — replaced by upload-url + confirm flow")
|
|
async def test_upload_txt_no_classify(async_client, sample_txt):
|
|
"""Legacy multipart upload test — endpoint removed in Plan 03-02 (D-04).
|
|
|
|
Replaced by test_upload_url_endpoint + test_confirm_endpoint.
|
|
"""
|
|
with open(sample_txt, "rb") as f:
|
|
resp = await async_client.post(
|
|
"/api/documents/upload",
|
|
files={"file": ("sample.txt", f, "text/plain")},
|
|
data={"auto_classify": "false"},
|
|
)
|
|
assert resp.status_code == 200
|
|
data = resp.json()
|
|
assert data["original_name"] == "sample.txt"
|
|
|
|
|
|
@pytest.mark.xfail(strict=False, reason="POST /api/documents/upload removed in Plan 03-02 — replaced by upload-url + confirm flow")
|
|
async def test_upload_pdf_no_classify(async_client, sample_pdf):
|
|
"""Legacy multipart upload test — endpoint removed in Plan 03-02 (D-04)."""
|
|
with open(sample_pdf, "rb") as f:
|
|
resp = await async_client.post(
|
|
"/api/documents/upload",
|
|
files={"file": ("sample.pdf", f, "application/pdf")},
|
|
data={"auto_classify": "false"},
|
|
)
|
|
assert resp.status_code == 200
|
|
|
|
|
|
async def test_list_documents(async_client):
|
|
"""GET /api/documents returns an empty list when no documents exist."""
|
|
resp = await async_client.get("/api/documents")
|
|
assert resp.status_code == 200
|
|
data = resp.json()
|
|
assert data["total"] == 0
|
|
assert data["items"] == []
|
|
|
|
|
|
async def test_list_documents_filter_by_topic(async_client, db_session):
|
|
"""GET /api/documents?topic=finance returns only matching documents."""
|
|
import uuid as _uuid
|
|
from db.models import Document
|
|
from services import storage
|
|
|
|
# Create a document directly via ORM (bypasses the upload endpoint)
|
|
doc_id = _uuid.uuid4()
|
|
doc = Document(
|
|
id=doc_id,
|
|
user_id=None,
|
|
filename="test.txt",
|
|
content_type="text/plain",
|
|
size_bytes=100,
|
|
storage_backend="minio",
|
|
status="uploaded",
|
|
object_key=f"null-user/{doc_id}/{_uuid.uuid4()}.txt",
|
|
)
|
|
db_session.add(doc)
|
|
await db_session.commit()
|
|
|
|
await storage.update_document_topics(db_session, str(doc_id), ["finance"])
|
|
|
|
resp = await async_client.get("/api/documents?topic=finance")
|
|
assert resp.json()["total"] == 1
|
|
|
|
resp2 = await async_client.get("/api/documents?topic=legal")
|
|
assert resp2.json()["total"] == 0
|
|
|
|
|
|
async def test_get_document(async_client, db_session):
|
|
"""GET /api/documents/{id} returns metadata for an existing document."""
|
|
import uuid as _uuid
|
|
from db.models import Document
|
|
|
|
doc_id = _uuid.uuid4()
|
|
doc = Document(
|
|
id=doc_id,
|
|
user_id=None,
|
|
filename="test.txt",
|
|
content_type="text/plain",
|
|
size_bytes=100,
|
|
storage_backend="minio",
|
|
status="uploaded",
|
|
object_key=f"null-user/{doc_id}/{_uuid.uuid4()}.txt",
|
|
)
|
|
db_session.add(doc)
|
|
await db_session.commit()
|
|
|
|
resp = await async_client.get(f"/api/documents/{doc_id}")
|
|
assert resp.status_code == 200
|
|
assert resp.json()["id"] == str(doc_id)
|
|
|
|
|
|
async def test_get_document_not_found(async_client):
|
|
resp = await async_client.get("/api/documents/nonexistent")
|
|
assert resp.status_code == 404
|
|
|
|
|
|
async def test_delete_document(async_client, db_session, monkeypatch):
|
|
"""DELETE /api/documents/{id} removes the document."""
|
|
import uuid as _uuid
|
|
from db.models import Document
|
|
from unittest.mock import AsyncMock
|
|
|
|
# Mock MinIO delete so we don't need a live MinIO
|
|
monkeypatch.setattr("services.storage._backend", lambda: type("B", (), {"delete_object": AsyncMock()})())
|
|
|
|
doc_id = _uuid.uuid4()
|
|
doc = Document(
|
|
id=doc_id,
|
|
user_id=None,
|
|
filename="test.txt",
|
|
content_type="text/plain",
|
|
size_bytes=0,
|
|
storage_backend="minio",
|
|
status="uploaded",
|
|
object_key=f"null-user/{doc_id}/{_uuid.uuid4()}.txt",
|
|
)
|
|
db_session.add(doc)
|
|
await db_session.commit()
|
|
|
|
resp = await async_client.delete(f"/api/documents/{doc_id}")
|
|
assert resp.status_code == 200
|
|
assert resp.json()["success"] is True
|
|
|
|
resp2 = await async_client.get(f"/api/documents/{doc_id}")
|
|
assert resp2.status_code == 404
|
|
|
|
|
|
async def test_delete_document_not_found(async_client):
|
|
resp = await async_client.delete("/api/documents/nonexistent")
|
|
assert resp.status_code == 404
|
|
|
|
|
|
@pytest.mark.xfail(strict=False, reason="POST /api/documents/upload removed in Plan 03-02 — replaced by upload-url + confirm flow")
|
|
async def test_upload_empty_file(async_client):
|
|
"""Legacy empty file test — endpoint removed in Plan 03-02 (D-04)."""
|
|
resp = await async_client.post(
|
|
"/api/documents/upload",
|
|
files={"file": ("empty.txt", b"", "text/plain")},
|
|
data={"auto_classify": "false"},
|
|
)
|
|
assert resp.status_code == 400
|
|
|
|
|
|
@pytest.mark.xfail(strict=False, reason="POST /api/documents/upload removed in Plan 03-02 — replaced by upload-url + confirm flow")
|
|
async def test_upload_persists_to_postgres_and_minio(async_client, sample_txt):
|
|
"""Legacy upload+persist test — endpoint removed in Plan 03-02 (D-04).
|
|
|
|
Replaced by the upload-url + confirm flow tested in test_upload_url_endpoint
|
|
and test_confirm_endpoint.
|
|
"""
|
|
with open(sample_txt, "rb") as f:
|
|
resp = await async_client.post(
|
|
"/api/documents/upload",
|
|
files={"file": ("sample.txt", f, "text/plain")},
|
|
data={"auto_classify": "false"},
|
|
)
|
|
assert resp.status_code == 200
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Wave 0 xfail stubs for Phase 3 document endpoint tests — Plans 03-02 / 03-03
|
|
# ---------------------------------------------------------------------------
|
|
|
|
|
|
async def test_upload_url_endpoint(async_client, auth_user, mock_minio_presigned):
|
|
"""POST /api/documents/upload-url returns {upload_url, document_id} and creates
|
|
a Document row with status='pending'.
|
|
|
|
D-05: two-step upload flow — step 1 creates the pending Document row and
|
|
returns the presigned PUT URL (15-min TTL). Quota is NOT reserved here.
|
|
"""
|
|
resp = await async_client.post(
|
|
"/api/documents/upload-url",
|
|
json={"filename": "report.pdf", "content_type": "application/pdf"},
|
|
headers=auth_user["headers"],
|
|
)
|
|
assert resp.status_code == 200, resp.text
|
|
data = resp.json()
|
|
assert "upload_url" in data, f"Missing upload_url: {data}"
|
|
assert "document_id" in data, f"Missing document_id: {data}"
|
|
assert "presigned" in data["upload_url"] or "localhost" in data["upload_url"], (
|
|
f"Expected a presigned URL: {data['upload_url']}"
|
|
)
|
|
# Verify mock was called
|
|
assert mock_minio_presigned.called, "generate_presigned_put_url was not called"
|
|
|
|
|
|
async def test_confirm_endpoint(
|
|
async_client, auth_user, mock_minio_presigned, mock_minio_stat, monkeypatch
|
|
):
|
|
"""POST /api/documents/{id}/confirm calls stat_object once, updates Document.size_bytes
|
|
from the stat return value, and sets Document.status='uploaded'.
|
|
|
|
D-05: step 3 of the presigned upload flow. stat_object provides the authoritative
|
|
file size (D-07). The atomic quota UPDATE runs here (STORE-03).
|
|
"""
|
|
from unittest.mock import MagicMock
|
|
|
|
# Patch out the Celery delay call — no Redis in unit test environment
|
|
mock_delay = MagicMock()
|
|
monkeypatch.setattr("api.documents.extract_and_classify.delay", mock_delay)
|
|
|
|
mock_minio_stat.return_value = 2048
|
|
|
|
# Step 1: get upload URL
|
|
resp = await async_client.post(
|
|
"/api/documents/upload-url",
|
|
json={"filename": "doc.txt", "content_type": "text/plain"},
|
|
headers=auth_user["headers"],
|
|
)
|
|
assert resp.status_code == 200, resp.text
|
|
doc_id = resp.json()["document_id"]
|
|
|
|
# Step 2: confirm (Wave 2 — user_id is None so quota skipped, but stat is called)
|
|
conf_resp = await async_client.post(
|
|
f"/api/documents/{doc_id}/confirm",
|
|
headers=auth_user["headers"],
|
|
)
|
|
assert conf_resp.status_code == 200, conf_resp.text
|
|
conf_data = conf_resp.json()
|
|
assert conf_data["id"] == doc_id
|
|
assert conf_data["size_bytes"] == 2048
|
|
assert conf_data["status"] == "uploaded"
|
|
# stat_object was called once
|
|
assert mock_minio_stat.called, "stat_object was not called"
|
|
# Celery task was dispatched
|
|
assert mock_delay.called, "extract_and_classify.delay was not called"
|
|
|
|
|
|
async def test_get_quota(async_client, auth_user):
|
|
"""GET /api/auth/me/quota returns {used_bytes: 0, limit_bytes: 104857600}.
|
|
|
|
STORE-04: quota usage bar endpoint. Returns current usage and limit for the
|
|
authenticated user. Newly created users start at used_bytes=0.
|
|
"""
|
|
resp = await async_client.get(
|
|
"/api/auth/me/quota",
|
|
headers=auth_user["headers"],
|
|
)
|
|
assert resp.status_code == 200, resp.text
|
|
data = resp.json()
|
|
assert "used_bytes" in data, f"Missing used_bytes: {data}"
|
|
assert "limit_bytes" in data, f"Missing limit_bytes: {data}"
|
|
assert data["used_bytes"] == 0, f"Expected 0 used_bytes for new user: {data}"
|
|
assert data["limit_bytes"] == 104_857_600, f"Expected 100 MB limit: {data}"
|
|
|
|
|
|
@pytest.mark.xfail(strict=False, reason="implemented in plan 03-03")
|
|
async def test_cross_user_access_404(async_client, auth_user, db_session):
|
|
"""User B's request for GET /api/documents/{A_doc_id} returns 404.
|
|
|
|
SEC-04: cross-user access returns 404 (not 403) to avoid information leakage
|
|
(CONTEXT.md D-16). An attacker cannot distinguish between 'document does not
|
|
exist' and 'document belongs to someone else'.
|
|
"""
|
|
assert True # scaffold
|
|
|
|
|
|
@pytest.mark.xfail(strict=False, reason="implemented in plan 03-03")
|
|
async def test_admin_cannot_access_documents(async_client, admin_user):
|
|
"""GET /api/documents using admin_user.headers returns 403.
|
|
|
|
SEC-04 SC4: admin accounts cannot access document content (CLAUDE.md +
|
|
CONTEXT.md D-16). The get_regular_user dependency enforces this for all
|
|
/api/documents/* handlers.
|
|
"""
|
|
assert True # scaffold
|
|
|
|
|
|
@pytest.mark.xfail(strict=False, reason="implemented in plan 03-03: auth guard not yet added")
|
|
async def test_documents_require_auth(async_client):
|
|
"""Anonymous GET /api/documents (no Authorization header) returns 401 or 403.
|
|
|
|
D-16: all /api/documents/* endpoints require authentication via
|
|
get_current_user (Phase 2 D-07 fulfilled in Phase 3).
|
|
Note: auth guard is added in Plan 03-03 — this remains xfail until then.
|
|
"""
|
|
resp = await async_client.get("/api/documents")
|
|
# Wave 2: no auth guard yet (Plan 03-03 adds it) — this will pass as xfail
|
|
assert resp.status_code in (401, 403), f"Expected 401 or 403, got {resp.status_code}"
|