#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
简化OCR测试脚本
"""

import cv2
import json
from pathlib import Path

def test_paddleocr():
    """测试PaddleOCR基本功能"""
    print("🔧 测试PaddleOCR...")
    
    try:
        from paddleocr import PaddleOCR
        
        # 初始化OCR
        ocr = PaddleOCR(
            use_textline_orientation=True,
            lang='ch'
        )
        print("✅ PaddleOCR初始化成功")
        
        # 测试视频文件
        video_path = "/root/autodl-tmp/hot_video_analyse/video2audio/sample_demo_6.mp4"
        
        # 读取一帧进行测试
        cap = cv2.VideoCapture(video_path)
        ret, frame = cap.read()
        cap.release()
        
        if not ret:
            print("❌ 无法读取视频帧")
            return False
        
        print("📸 成功读取视频帧")
        
        # 进行OCR识别 - 使用predict方法
        try:
            result = ocr.predict(frame)
        except:
            # 如果predict不工作，尝试ocr方法
            result = ocr.ocr(frame)
        
        print(f"🔍 OCR结果类型: {type(result)}")
        
        # 处理结果
        if result:
            if isinstance(result, list) and len(result) > 0:
                detections = result[0] if isinstance(result[0], list) else result
                print(f"🔍 OCR结果: {len(detections)} 个检测")
                
                for i, detection in enumerate(detections[:3]):  # 只显示前3个
                    if detection and len(detection) >= 2:
                        try:
                            bbox, text_info = detection
                            if isinstance(text_info, tuple) and len(text_info) >= 2:
                                text, confidence = text_info
                                print(f"  {i+1}. 文本: '{text}' 置信度: {confidence:.3f}")
                            else:
                                print(f"  {i+1}. 检测到文本但格式异常: {text_info}")
                        except Exception as e:
                            print(f"  {i+1}. 处理检测结果时出错: {e}")
            else:
                print("🔍 OCR结果为空或格式异常")
        else:
            print("🔍 OCR结果为None")
        
        return True
        
    except Exception as e:
        print(f"❌ PaddleOCR测试失败: {e}")
        import traceback
        print(f"详细错误: {traceback.format_exc()}")
        return False

def test_easyocr():
    """测试EasyOCR基本功能"""
    print("\n🔧 测试EasyOCR...")
    
    try:
        import easyocr
        
        # 初始化OCR
        reader = easyocr.Reader(['ch_sim', 'en'])
        print("✅ EasyOCR初始化成功")
        
        # 测试视频文件
        video_path = "/root/autodl-tmp/hot_video_analyse/video2audio/sample_demo_6.mp4"
        
        # 读取一帧进行测试
        cap = cv2.VideoCapture(video_path)
        ret, frame = cap.read()
        cap.release()
        
        if not ret:
            print("❌ 无法读取视频帧")
            return
        
        print("📸 成功读取视频帧")
        
        # 进行OCR识别
        result = reader.readtext(frame)
        
        print(f"🔍 OCR结果: {len(result)} 个检测")
        
        for i, detection in enumerate(result[:3]):  # 只显示前3个
            bbox, text, confidence = detection
            print(f"  {i+1}. 文本: '{text}' 置信度: {confidence:.3f}")
        
        return True
        
    except Exception as e:
        print(f"❌ EasyOCR测试失败: {e}")
        return False

def main():
    """主函数"""
    print("🧪 简化OCR功能测试")
    print("=" * 40)
    
    paddle_ok = test_paddleocr()
    easy_ok = test_easyocr()
    
    print("\n" + "=" * 40)
    print("📊 测试结果:")
    print(f"  PaddleOCR: {'✅ 成功' if paddle_ok else '❌ 失败'}")
    print(f"  EasyOCR:   {'✅ 成功' if easy_ok else '❌ 失败'}")
    
    if paddle_ok or easy_ok:
        print("\n🎉 至少有一个OCR引擎工作正常，可以继续使用！")
    else:
        print("\n⚠️ 所有OCR引擎都有问题，请检查安装")

if __name__ == "__main__":
    main()