ArchCoder/llm-excel-plotter-agent

1import pandas as pd2import os3import logging4 5class DataProcessor:6    def __init__(self, data_path=None):7        logging.info("Initializing DataProcessor")8        # Allow dynamic data path (for user uploads), fallback to default9        if data_path and os.path.exists(data_path):10            self.data_path = data_path11        else:12            self.data_path = os.path.join(os.path.dirname(__file__), 'data', 'sample_data.csv')13        self.data = self.load_data(self.data_path)14 15    def load_data(self, path):16        ext = os.path.splitext(path)[1].lower()17        try:18            if ext == '.csv':19                data = pd.read_csv(path)20            elif ext == '.xls':21                data = pd.read_excel(path, engine='xlrd')22            elif ext == '.xlsx':23                data = pd.read_excel(path, engine='openpyxl')24            else:25                raise ValueError(f"Unsupported file type: {ext}")26            logging.info(f"Loaded data from {path} with shape {data.shape}")27            return data28        except Exception as e:29            logging.error(f"Failed to load data: {e}")30            return pd.DataFrame()31 32    def validate_columns(self, required_columns):33        missing = [col for col in required_columns if col not in self.data.columns]34        if missing:35            logging.warning(f"Missing columns: {missing}")36            return False, missing37        return True, []38 39    def get_columns(self):40        return list(self.data.columns)41 42    def preview(self, n=5):43        return self.data.head(n).to_dict(orient='records')44 45    def get_dtypes(self) -> dict:46        result = {}47        for col, dtype in self.data.dtypes.items():48            if pd.api.types.is_integer_dtype(dtype):49                result[col] = "integer"50            elif pd.api.types.is_float_dtype(dtype):51                result[col] = "float"52            elif pd.api.types.is_datetime64_any_dtype(dtype):53                result[col] = "datetime"54            elif pd.api.types.is_bool_dtype(dtype):55                result[col] = "boolean"56            else:57                result[col] = "string"58        return result59 60    def get_stats(self) -> dict:61        numeric = self.data.select_dtypes(include='number')62        if numeric.empty:63            return {}64        desc = numeric.describe().to_dict()65        return {col: {k: round(v, 4) for k, v in stats.items()} for col, stats in desc.items()}66 67