AcidMap/app/utils/cd_prediction_wrapper.py

"""
Cd预测系统包装器
@description: 封装对原始Cd预测系统的调用，简化集成过程
@author: AcidMap Team
@version: 1.0.0
"""

import os
import sys
import logging
import subprocess
import time
from typing import Dict, Any, Optional
from datetime import datetime

class CdPredictionWrapper:
    """
    Cd预测系统包装器类
    
    @description: 简化对原始Cd预测系统的调用和集成
    @example
    >>> wrapper = CdPredictionWrapper()
    >>> result = wrapper.run_crop_cd_prediction()
    """
    
    def __init__(self, cd_system_path: str):
        """
        初始化包装器
        
        @param {str} cd_system_path - Cd预测系统的路径
        """
        self.cd_system_path = cd_system_path
        self.logger = logging.getLogger(__name__)
        
        # 验证Cd预测系统是否存在
        if not os.path.exists(cd_system_path):
            raise FileNotFoundError(f"Cd预测系统路径不存在: {cd_system_path}")
        
        # 检查关键文件是否存在
        self._validate_cd_system()
    
    def _validate_cd_system(self):
        """
        验证Cd预测系统的完整性
        
        @throws {FileNotFoundError} 当关键文件缺失时抛出
        """
        required_files = [
            "main.py",
            "config.py",
            "models",
            "analysis"
        ]
        
        for file_path in required_files:
            full_path = os.path.join(self.cd_system_path, file_path)
            if not os.path.exists(full_path):
                raise FileNotFoundError(f"Cd预测系统缺少必要文件: {file_path}")
        
        self.logger.info("Cd预测系统验证通过")
    
    def run_prediction_script(self, model_type: str = "both", raster_config_override: Optional[Dict[str, Any]] = None) -> Dict[str, Any]:
        """
        运行Cd预测脚本
        
        @param {str} model_type - 模型类型 ("crop", "effective", "both")
        @returns {Dict[str, Any]} 预测结果信息
        @throws {Exception} 当预测过程失败时抛出
        """
        try:
            self.logger.info(f"开始运行Cd预测脚本，模型类型: {model_type}")
            
            # 切换到Cd预测系统目录
            original_cwd = os.getcwd()
            os.chdir(self.cd_system_path)
            
            try:
                # 通过环境变量传递工作流配置，避免修改配置文件
                workflow_config = self._get_workflow_config(model_type)
                import json
                os.environ['CD_WORKFLOW_CONFIG'] = json.dumps(workflow_config)
                
                # 如果有栅格配置覆盖参数，也通过环境变量传递
                if raster_config_override:
                    os.environ['CD_RASTER_CONFIG_OVERRIDE'] = json.dumps(raster_config_override)
                
                # 运行主脚本
                result = subprocess.run(
                    [sys.executable, "main.py"],
                    capture_output=True,
                    text=True,
                    timeout=300  # 5分钟超时
                )
                
                if result.returncode != 0:
                    self.logger.error(f"Cd预测脚本执行失败: {result.stderr}")
                    raise Exception(f"预测脚本执行失败: {result.stderr}")
                
                self.logger.info("Cd预测脚本执行成功")
                
                # 获取输出文件信息
                output_info = self._get_output_files(model_type)
                
                return {
                    "success": True,
                    "model_type": model_type,
                    "output_files": output_info,
                    "stdout": result.stdout,
                    "timestamp": datetime.now().isoformat()
                }
                
            finally:
                # 恢复原始工作目录
                os.chdir(original_cwd)
                
                # 清理环境变量
                if 'CD_WORKFLOW_CONFIG' in os.environ:
                    del os.environ['CD_WORKFLOW_CONFIG']
                if 'CD_RASTER_CONFIG_OVERRIDE' in os.environ:
                    del os.environ['CD_RASTER_CONFIG_OVERRIDE']
                
        except subprocess.TimeoutExpired:
            self.logger.error("Cd预测脚本执行超时")
            raise Exception("预测脚本执行超时")
        except Exception as e:
            self.logger.error(f"运行Cd预测脚本失败: {str(e)}")
            raise
    
    def _get_workflow_config(self, model_type: str) -> dict:
        """
        获取工作流配置
        
        @param {str} model_type - 模型类型
        @returns {dict} 工作流配置字典
        """
        # 根据模型类型设置配置
        if model_type == "crop":
            workflow_config = {
                "run_crop_model": True,
                "run_effective_model": False,
                "combine_predictions": True,
                "generate_raster": True,
                "create_visualization": True,
                "create_histogram": True
            }
        elif model_type == "effective":
            workflow_config = {
                "run_crop_model": False,
                "run_effective_model": True,
                "combine_predictions": True,
                "generate_raster": True,
                "create_visualization": True,
                "create_histogram": True
            }
        else:  # both
            workflow_config = {
                "run_crop_model": True,
                "run_effective_model": True,
                "combine_predictions": True,
                "generate_raster": True,
                "create_visualization": True,
                "create_histogram": True
            }
        
        self.logger.info(f"生成工作流配置，模型类型: {model_type}")
        return workflow_config
    
    def _get_output_files(self, model_type: str) -> Dict[str, Any]:
        """
        获取输出文件信息
        
        @param {str} model_type - 模型类型
        @returns {Dict[str, Any]} 输出文件信息
        """
        output_dir = os.path.join(self.cd_system_path, "output")
        figures_dir = os.path.join(output_dir, "figures")
        raster_dir = os.path.join(output_dir, "raster")
        
        output_files = {
            "figures_dir": figures_dir,
            "raster_dir": raster_dir,
            "maps": [],
            "histograms": [],
            "rasters": []
        }
        
        # 查找相关文件
        if os.path.exists(figures_dir):
            for file in os.listdir(figures_dir):
                if file.endswith(('.jpg', '.png')):
                    file_path = os.path.join(figures_dir, file)
                    if "Prediction" in file and "results" in file:
                        output_files["maps"].append(file_path)
                    elif "frequency" in file.lower() or "histogram" in file.lower():
                        output_files["histograms"].append(file_path)
        
        if os.path.exists(raster_dir):
            for file in os.listdir(raster_dir):
                if file.endswith('.tif') and file.startswith('output'):
                    output_files["rasters"].append(os.path.join(raster_dir, file))
        
        return output_files
    
    def get_latest_outputs(self, output_type: str = "all", model_type: str = None) -> Dict[str, Optional[str]]:
        """
        获取最新的输出文件
        
        @param {str} output_type - 输出类型 ("maps", "histograms", "rasters", "all")
        @param {str} model_type - 模型类型 ("crop", "effective", None为获取所有)
        @returns {Dict[str, Optional[str]]} 最新输出文件路径
        """
        try:
            output_dir = os.path.join(self.cd_system_path, "output")
            figures_dir = os.path.join(output_dir, "figures")
            raster_dir = os.path.join(output_dir, "raster")
            
            latest_files = {}
            
            # 获取最新地图文件
            if output_type in ["maps", "all"]:
                map_files = []
                if os.path.exists(figures_dir):
                    for file in os.listdir(figures_dir):
                        if "Prediction" in file and "results" in file and file.endswith(('.jpg', '.png')):
                            file_path = os.path.join(figures_dir, file)
                            map_files.append(file_path)
                
                latest_files["latest_map"] = max(map_files, key=os.path.getctime) if map_files else None
            
            # 获取最新直方图文件
            if output_type in ["histograms", "all"]:
                histogram_files = []
                if os.path.exists(figures_dir):
                    for file in os.listdir(figures_dir):
                        if ("frequency" in file.lower() or "histogram" in file.lower()) and file.endswith(('.jpg', '.png')):
                            file_path = os.path.join(figures_dir, file)
                            histogram_files.append(file_path)
                
                latest_files["latest_histogram"] = max(histogram_files, key=os.path.getctime) if histogram_files else None
            
            # 获取最新栅格文件
            if output_type in ["rasters", "all"]:
                raster_files = []
                if os.path.exists(raster_dir):
                    for file in os.listdir(raster_dir):
                        if file.startswith('output') and file.endswith('.tif'):
                            file_path = os.path.join(raster_dir, file)
                            raster_files.append(file_path)
                
                latest_files["latest_raster"] = max(raster_files, key=os.path.getctime) if raster_files else None
            
            # 添加调试信息
            self.logger.info(f"获取最新输出文件 - 模型类型: {model_type}, 输出类型: {output_type}")
            for key, value in latest_files.items():
                if value:
                    self.logger.info(f"  {key}: {os.path.basename(value)}")
                else:
                    self.logger.warning(f"  {key}: 未找到文件")
            
            return latest_files
            
        except Exception as e:
            self.logger.error(f"获取最新输出文件失败: {str(e)}")
            return {}