app.py

"""
Gradio 网页版界面
提供浏览器端可视化的 OCR 自动录单演示
"""
import gradio as gr
import os
from ocr_engine import OCREngine
from data_extractor import DataExtractor
from excel_writer import ExcelWriter

# 初始化核心引擎
ocr_engine = OCREngine(lang='ch')
data_extractor = DataExtractor()
excel_writer = ExcelWriter(output_dir='output')

def process_single_image(image_path):
    """处理单张图片并返回提取结果"""
    if not image_path:
        return "请先上传一张图片", None
    
    try:
        # 1. OCR 识别
        ocr_result = ocr_engine.recognize(image_path)
        
        # 2. 提取数据
        data = data_extractor.extract(ocr_result)
        
        # 3. 写入单条 Excel
        data['image_path'] = os.path.basename(image_path)
        data['status'] = '成功'
        
        excel_path = excel_writer.create_report([data], filename="temp_single_result.xlsx")
        
        # 格式化展示文本
        result_text = "### 📥 提取结果：" + chr(10) + \
                      f"- **订单编号**: {data.get('order_no', '未提取到')}" + chr(10) + \
                      f"- **客户名称**: {data.get('customer', '未提取到')}" + chr(10) + \
                      f"- **产品名称**: {data.get('product', '未提取到')}" + chr(10) + \
                      f"- **数量**: {data.get('quantity', '未提取到')}" + chr(10) + \
                      f"- **金额**: {data.get('amount', '未提取到')}" + chr(10) + \
                      f"- **提取时间**: {data.get('extract_time', '')}"
        return result_text, excel_path
        
    except Exception as e:
        return f"❌ 处理失败: {str(e)}", None

def process_batch_images(image_files):
    """批量处理多张图片并生成汇总 Excel"""
    if not image_files:
        return "请先上传图片文件", None
        
    results = []
    summary_text = "### 📊 批量处理进度:" + chr(10)
    
    for i, file_obj in enumerate(image_files):
        # Gradio 批量上传的文件可能是 File 对象，通过 .name 获取路径
        path = file_obj.name if hasattr(file_obj, 'name') else file_obj
        filename = os.path.basename(path)
        try:
            ocr_result = ocr_engine.recognize(path)
            data = data_extractor.extract(ocr_result)
            data['image_path'] = filename
            data['status'] = '成功'
            results.append(data)
            summary_text += f"- ✅ {filename} 提取成功" + chr(10)
        except Exception as e:
            results.append({
                'image_path': filename,
                'status': f'失败: {str(e)}'
            })
            summary_text += f"- ❌ {filename} 处理失败: {str(e)}" + chr(10)
            
    # 生成 Excel 汇总报告
    output_path = excel_writer.create_report(results)
    summary_text += chr(10) + "**🎉 批量处理完成！已生成汇总 Excel 报告。**"
    
    return summary_text, output_path

# 构建 Gradio 界面
with gr.Blocks(title="OCR 自动录单系统") as demo:
    gr.Markdown("""
# 📄 OCR 自动录单系统 (网页版)
这是一个基于 **PaddleOCR** + **Python** 的轻量级高效率自动录单工具。适合制造业、电气、设备厂等办公室自动录入采购单、送货单和发票。
""")
    
    with gr.Tab("单张图片调试"):
        with gr.Row():
            with gr.Column():
                input_img = gr.Image(type="filepath", label="上传订单/采购单图片")
                btn_run = gr.Button("🚀 开始智能提取", variant="primary")
            with gr.Column():
                out_text = gr.Markdown("提取结果会显示在这里...")
                out_file = gr.File(label="下载当前生成的 Excel 录单")
                
        btn_run.click(
            fn=process_single_image,
            inputs=input_img,
            outputs=[out_text, out_file]
        )
        
    with gr.Tab("批量录单模式"):
        with gr.Row():
            with gr.Column():
                input_files = gr.File(file_count="multiple", label="上传多张订单图片", file_types=["image"])
                btn_batch_run = gr.Button("⚡ 开始批量全自动录入", variant="primary")
            with gr.Column():
                out_batch_text = gr.Markdown("批量处理进度会显示在这里...")
                out_batch_file = gr.File(label="下载汇总生成的 Excel 报表")
                
        btn_batch_run.click(
            fn=process_batch_images,
            inputs=input_files,
            outputs=[out_batch_text, out_batch_file]
        )

if __name__ == "__main__":
    demo.launch(server_name="0.0.0.0", server_port=7860)