import pytest
import os
import cv2
import numpy as np
from src.core.inference import InferenceEngine

class TestInference:
    @pytest.fixture
    def inference_engine(self):
        model_path = os.path.join('data', 'models', 'UAV-250321.onnx')
        return InferenceEngine(model_path)

    def test_single_image_inference(self, inference_engine):
        # 创建一个测试图像
        test_image = np.zeros((640, 640, 3), dtype=np.uint8)
        test_image[100:200, 100:200] = 255  # 添加一个白色方块

        # 执行推理
        results = inference_engine.infer(test_image)
        
        # 验证结果格式
        assert isinstance(results, list)
        for result in results:
            assert 'bbox' in result
            assert 'confidence' in result
            assert 'class_id' in result

    def test_batch_inference(self, inference_engine):
        # 创建测试图像列表
        test_images = [
            np.zeros((640, 640, 3), dtype=np.uint8) for _ in range(3)
        ]
        
        # 执行批量推理
        results = inference_engine.batch_infer(test_images)
        
        # 验证结果
        assert len(results) == len(test_images)
        assert all(isinstance(r, list) for r in results)

    def test_invalid_input(self, inference_engine):
        # 测试无效输入
        with pytest.raises(ValueError):
            inference_engine.infer(None)
        
        with pytest.raises(ValueError):
            inference_engine.infer(np.zeros((100, 100), dtype=np.uint8))  # 灰度图像