""" API Integration Tests Tests FastAPI endpoints with mocked services. These tests verify the API layer works correctly with the service layer. """ import io import tempfile from dataclasses import dataclass, field from datetime import datetime, timezone from pathlib import Path from typing import Any from unittest.mock import MagicMock, patch from uuid import uuid4 import pytest from fastapi import FastAPI from fastapi.testclient import TestClient @dataclass class MockServiceResult: """Mock result from inference service.""" document_id: str = "test-doc-123" success: bool = True document_type: str = "invoice" fields: dict[str, str] = field(default_factory=lambda: { "InvoiceNumber": "INV-2024-001", "Amount": "1500.00", "InvoiceDate": "2024-01-15", "OCR": "12345678901234", "Bankgiro": "1234-5678", }) confidence: dict[str, float] = field(default_factory=lambda: { "InvoiceNumber": 0.95, "Amount": 0.92, "InvoiceDate": 0.88, "OCR": 0.95, "Bankgiro": 0.90, }) detections: list[dict[str, Any]] = field(default_factory=list) processing_time_ms: float = 150.5 visualization_path: Path | None = None errors: list[str] = field(default_factory=list) @pytest.fixture def temp_storage_dir(): """Create temporary storage directories.""" with tempfile.TemporaryDirectory() as tmpdir: base = Path(tmpdir) uploads_dir = base / "uploads" / "inference" results_dir = base / "results" uploads_dir.mkdir(parents=True, exist_ok=True) results_dir.mkdir(parents=True, exist_ok=True) yield { "base": base, "uploads": uploads_dir, "results": results_dir, } @pytest.fixture def mock_inference_service(): """Create a mock inference service.""" service = MagicMock() service.is_initialized = True service.gpu_available = False # Create a realistic mock result mock_result = MockServiceResult() service.process_pdf.return_value = mock_result service.process_image.return_value = mock_result service.initialize.return_value = None return service @pytest.fixture def mock_storage_config(temp_storage_dir): """Create mock storage configuration.""" from backend.web.config import StorageConfig return StorageConfig( upload_dir=temp_storage_dir["uploads"], result_dir=temp_storage_dir["results"], max_file_size_mb=50, ) @pytest.fixture def mock_storage_helper(temp_storage_dir): """Create a mock storage helper.""" helper = MagicMock() helper.get_uploads_base_path.return_value = temp_storage_dir["uploads"] helper.get_result_local_path.return_value = None helper.result_exists.return_value = False return helper @pytest.fixture def test_app(mock_inference_service, mock_storage_config, mock_storage_helper): """Create a test FastAPI application with mocked storage.""" from backend.web.api.v1.public.inference import create_inference_router app = FastAPI() # Patch get_storage_helper to return our mock with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): inference_router = create_inference_router(mock_inference_service, mock_storage_config) app.include_router(inference_router) return app @pytest.fixture def client(test_app, mock_storage_helper): """Create a test client with storage helper patched.""" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): yield TestClient(test_app) class TestHealthEndpoint: """Tests for health check endpoint.""" def test_health_check(self, client, mock_inference_service): """Test health check returns status.""" response = client.get("/api/v1/health") assert response.status_code == 200 data = response.json() assert "status" in data assert "model_loaded" in data class TestInferenceEndpoint: """Tests for inference endpoint.""" def test_infer_pdf(self, client, mock_inference_service, mock_storage_helper, temp_storage_dir): """Test PDF inference endpoint.""" # Create a minimal PDF content pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) assert response.status_code == 200 data = response.json() assert "result" in data assert data["result"]["success"] is True assert "InvoiceNumber" in data["result"]["fields"] def test_infer_image(self, client, mock_inference_service, mock_storage_helper): """Test image inference endpoint.""" # Create minimal PNG header png_header = b"\x89PNG\r\n\x1a\n" + b"\x00" * 100 with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.png", io.BytesIO(png_header), "image/png")}, ) assert response.status_code == 200 data = response.json() assert "result" in data def test_infer_invalid_file_type(self, client, mock_storage_helper): """Test rejection of invalid file types.""" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.txt", io.BytesIO(b"hello"), "text/plain")}, ) assert response.status_code == 400 def test_infer_no_file(self, client, mock_storage_helper): """Test rejection when no file provided.""" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post("/api/v1/infer") assert response.status_code == 422 # Validation error def test_infer_result_structure(self, client, mock_inference_service, mock_storage_helper): """Test that result has expected structure.""" pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) data = response.json() result = data["result"] # Check required fields assert "document_id" in result assert "success" in result assert "fields" in result assert "confidence" in result assert "processing_time_ms" in result class TestInferenceResultFormat: """Tests for inference result formatting.""" def test_result_fields_mapped_correctly(self, client, mock_inference_service, mock_storage_helper): """Test that fields are mapped to API response format.""" pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) data = response.json() fields = data["result"]["fields"] assert fields["InvoiceNumber"] == "INV-2024-001" assert fields["Amount"] == "1500.00" assert fields["InvoiceDate"] == "2024-01-15" def test_confidence_values_included(self, client, mock_inference_service, mock_storage_helper): """Test that confidence values are included.""" pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) data = response.json() confidence = data["result"]["confidence"] assert "InvoiceNumber" in confidence assert confidence["InvoiceNumber"] == 0.95 class TestErrorHandling: """Tests for error handling in API.""" def test_service_error_handling(self, client, mock_inference_service, mock_storage_helper): """Test handling of service errors.""" mock_inference_service.process_pdf.side_effect = Exception("Processing failed") pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) # Should return error response assert response.status_code >= 400 def test_empty_file_handling(self, client, mock_storage_helper): """Test handling of empty files.""" # Empty file still has valid content type with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(b""), "application/pdf")}, ) # Empty file may be processed or rejected depending on implementation # Just verify we get a response assert response.status_code in [200, 400, 422, 500] class TestResponseFormat: """Tests for API response format consistency.""" def test_success_response_format(self, client, mock_inference_service, mock_storage_helper): """Test successful response format.""" pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) assert response.status_code == 200 assert response.headers["content-type"] == "application/json" data = response.json() assert isinstance(data, dict) assert "result" in data def test_json_serialization(self, client, mock_inference_service, mock_storage_helper): """Test that all result fields are JSON serializable.""" pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) # If this doesn't raise, JSON is valid data = response.json() assert data is not None class TestDocumentIdGeneration: """Tests for document ID handling.""" def test_document_id_generated(self, client, mock_inference_service, mock_storage_helper): """Test that document ID is generated.""" pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("test.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) data = response.json() assert "document_id" in data["result"] assert data["result"]["document_id"] is not None def test_document_id_from_filename(self, client, mock_inference_service, mock_storage_helper): """Test document ID derived from filename.""" pdf_content = b"%PDF-1.4\n%test\n" with patch( "backend.web.api.v1.public.inference.get_storage_helper", return_value=mock_storage_helper, ): response = client.post( "/api/v1/infer", files={"file": ("my_invoice_123.pdf", io.BytesIO(pdf_content), "application/pdf")}, ) data = response.json() # Document ID should be set (either from filename or generated) assert data["result"]["document_id"] is not None