mirror of
https://github.com/PR0M3TH3AN/unredact.git
synced 2026-11-01 11:10:29 +00:00
Alex-GilbertandClaude Opus 4.6
c95b6cb9d7
Adds POST /api/upload (runs full pipeline: rasterize, OCR, font detect, overlay), GET endpoints for original/overlay images and OCR data JSON. Also adds python-multipart, anyio, and pytest-anyio dependencies. Co-Authored-By: Claude Opus 4.6 <[email protected]>
80 lines
2.5 KiB
Python
80 lines
2.5 KiB
Python
import io
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
from httpx import ASGITransport, AsyncClient
|
|
|
|
from unredact.app import app
|
|
|
|
|
|
@pytest.fixture
|
|
def pdf_bytes(sample_pdf: Path) -> bytes:
|
|
return sample_pdf.read_bytes()
|
|
|
|
|
|
@pytest.mark.anyio
|
|
async def test_upload_pdf(pdf_bytes: bytes):
|
|
transport = ASGITransport(app=app)
|
|
async with AsyncClient(transport=transport, base_url="http://test") as client:
|
|
resp = await client.post(
|
|
"/api/upload",
|
|
files={"file": ("test.pdf", pdf_bytes, "application/pdf")},
|
|
)
|
|
assert resp.status_code == 200
|
|
data = resp.json()
|
|
assert "doc_id" in data
|
|
assert data["page_count"] > 0
|
|
|
|
|
|
@pytest.mark.anyio
|
|
async def test_get_page_overlay(pdf_bytes: bytes):
|
|
transport = ASGITransport(app=app)
|
|
async with AsyncClient(transport=transport, base_url="http://test") as client:
|
|
# Upload first
|
|
resp = await client.post(
|
|
"/api/upload",
|
|
files={"file": ("test.pdf", pdf_bytes, "application/pdf")},
|
|
)
|
|
doc_id = resp.json()["doc_id"]
|
|
|
|
# Get page 1 overlay
|
|
resp = await client.get(f"/api/doc/{doc_id}/page/1/overlay")
|
|
assert resp.status_code == 200
|
|
assert resp.headers["content-type"] == "image/png"
|
|
assert len(resp.content) > 1000 # Should be a real image
|
|
|
|
|
|
@pytest.mark.anyio
|
|
async def test_get_page_original(pdf_bytes: bytes):
|
|
transport = ASGITransport(app=app)
|
|
async with AsyncClient(transport=transport, base_url="http://test") as client:
|
|
resp = await client.post(
|
|
"/api/upload",
|
|
files={"file": ("test.pdf", pdf_bytes, "application/pdf")},
|
|
)
|
|
doc_id = resp.json()["doc_id"]
|
|
|
|
resp = await client.get(f"/api/doc/{doc_id}/page/1/original")
|
|
assert resp.status_code == 200
|
|
assert resp.headers["content-type"] == "image/png"
|
|
|
|
|
|
@pytest.mark.anyio
|
|
async def test_get_page_data(pdf_bytes: bytes):
|
|
"""Should return OCR data and font info as JSON."""
|
|
transport = ASGITransport(app=app)
|
|
async with AsyncClient(transport=transport, base_url="http://test") as client:
|
|
resp = await client.post(
|
|
"/api/upload",
|
|
files={"file": ("test.pdf", pdf_bytes, "application/pdf")},
|
|
)
|
|
doc_id = resp.json()["doc_id"]
|
|
|
|
resp = await client.get(f"/api/doc/{doc_id}/page/1/data")
|
|
assert resp.status_code == 200
|
|
data = resp.json()
|
|
assert "font" in data
|
|
assert "lines" in data
|
|
assert data["font"]["name"]
|
|
assert data["font"]["size"] > 0
|