#!/usr/bin/python3
# -*- coding: utf-8 -*-
# @Author: Hui
# @Desc: { Excel文件操作工具模塊 }
# @Date: 2022/04/03 19:34
import pandas as pd
from typing import List, Union, Dict, IO


class ExcelUtils(object):
    """ Excel文件操作工具類 """

    @classmethod
    def list_to_excel(
            cls,
            path_or_buffer: Union[str, IO],
            data_list: list,
            col_mapping: dict = None,
            sheet_name: str = 'Sheet1',
            **kwargs
    ):
        """
        列表轉(zhuǎn) excel文件
        Args:
            path_or_buffer: 文件路徑或者緩沖流
            data_list: 數(shù)據(jù)集 List[dict]
            col_mapping: 表頭列字段映射
            sheet_name: sheet名稱

        Returns:
        """
        with pd.ExcelWriter(path_or_buffer) as writer:
            _col_mapping = list(col_mapping) if col_mapping else None
            df = pd.DataFrame(data=data_list, columns=_col_mapping)
            if col_mapping:
                df.rename(columns=col_mapping, inplace=True)
            df.to_excel(writer, sheet_name=sheet_name, index=False, **kwargs)

# 示例
user_list = [
    dict(id=1, name='hui', age=20),
    dict(id=2, name='wang', age=22),
    dict(id=3, name='zack', age=25),
]
user_col_mapping = {
    'id': '用戶id',
    'name': '用戶名',
    'age': '年齡',
}

ExcelUtils.list_to_excel('user.xlsx', user_list, col_mapping=user_col_mapping)

# 導出為excel文件字節(jié)流處理 
excel_bio = BytesIO()
ExcelUtils.list_to_excel(
    excel_bio,
    data_list=user_list,
    col_mapping=user_col_mapping,
    sheet_name='demo'
) 
excel_bytes = excel_bio.getvalue()

print("excel_bytes type => ", type(excel_bytes))

>>>out
excel_bytes type =>  <class 'bytes'>

@classmethod
def multi_list_to_excel(
        cls,
        path_or_buffer: Union[str, IO],
        data_collects: List[tuple],
        **kwargs
):
    """
    多列表轉(zhuǎn)帶不同 sheet的excel文件
    Args:
        path_or_buffer: 文件路徑或者緩沖流
        data_collects: 大數(shù)據(jù)集 list[(data_collect, col_mapping, sheet_name)]
            data_collect: 數(shù)據(jù)集,
            col_mapping: 列字段映射,
            sheet_name: excel表sheet名稱

    Returns:
    """
    with pd.ExcelWriter(path_or_buffer) as writer:
        for data_collect, col_mapping, sheet_name in data_collects:
            df = pd.DataFrame(data=data_collect, columns=list(col_mapping))
            df.rename(columns=col_mapping, inplace=True)
            df.to_excel(writer, sheet_name=sheet_name, index=False, **kwargs)

user_list = [
    {'id': 1, 'name': 'hui', 'age': 18},
    {'id': 2, 'name': 'wang', 'age': 19},
    {'id': 3, 'name': 'zack', 'age': 20}
]

book_list = [
    {'id': 1, 'name': 'Python基礎教程', 'author': 'hui', 'price': 30},
    {'id': 2, 'name': 'Java高級編程', 'author': 'wang', 'price': 50},
    {'id': 3, 'name': '機器學習實戰(zhàn)', 'author': 'zack', 'price': 70},
]

user_col_mapping = {'id': '編號', 'name': '姓名', 'age': '年齡'}
book_col_mapping = {'id': '編號', 'name': '書名', 'author': '作者', 'price': '價格'}

data_collects = [
    (user_list, user_col_mapping, '用戶信息'),
    (book_list, book_col_mapping, '圖書信息')
]

ExcelUtils.multi_list_to_excel('multi_sheet_data.xlsx', data_collects)

@classmethod
def read_excel(
        cls,
        path_or_buffer: Union[str, IO],
        sheet_name: str = "Sheet1",
        col_mapping: dict = None,
        all_col: bool = True,
        header: int = 0,
        **kwargs
) -> List[dict]:
    """
    讀取excel表格數(shù)據(jù)，根據(jù)col_mapping替換列名
    Args:
        path_or_buffer: 文件路徑或者緩沖流
        sheet_name: 讀書excel表的sheet名稱
        col_mapping: 列字段映射
        all_col: True返回所有列信息，F(xiàn)alse則返回col_mapping對應的字段信息
        header: 默認0從第一行開啟讀取，用于指定從第幾行開始讀取

    Returns:
    """
    use_cols = None
    if not all_col:
        # 獲取excel表指定列數(shù)據(jù)
        use_cols = list(col_mapping) if col_mapping else None

    df = pd.read_excel(path_or_buffer, sheet_name=sheet_name, usecols=use_cols, header=header, **kwargs)
    if col_mapping:
        df.rename(columns=col_mapping, inplace=True)

    return df.to_dict("records")

data = [
    {"id": 1, "name": "hui", "age": 30},
    {"id": 2, "name": "zack", "age": 25},
    {"id": 3, "name": "wang", "age": 40},
]

# 將數(shù)據(jù)寫入Excel文件
ExcelUtils.list_to_excel("read_demo.xlsx", data, col_mapping={"id": "用戶ID", "name": "姓名", "age": "年齡"})

# 讀取Excel文件
result = ExcelUtils.read_excel("read_demo.xlsx", col_mapping={"用戶ID": "id", "姓名": "name"})

print(result)

>>>out
[{'id': 1, 'name': 'hui'}, {'id': 2, 'name': 'zack'}, {'id': 3, 'name': 'wang'}]

@classmethod
def merge_excel_files(
        cls,
        input_files: List[str],
        output_file: str,
        sheet_name_mapping: Dict[str, str] = None,
        **kwargs
):
    """
    合并多個Excel文件到一個文件中（每個文件對應一個工作表）
    如果Excel文件有多個作表，則默認取第一個工作表
    Args:
        input_files: 待合并的excel文件列表
        output_file: 輸出文件路徑
        sheet_name_mapping: 文件工作表映射，默認為文件名
            {"文件名1": "sheet1", "文件名2": "sheet2"}

    Returns:
    """
    sheet_name_mapping = sheet_name_mapping or {}
    with pd.ExcelWriter(output_file, **kwargs) as writer:
        for file in input_files:
            df = pd.read_excel(file)
            sheet_name = sheet_name_mapping.get(file, file)
            df.to_excel(writer, sheet_name=sheet_name, index=False)

def merge_excel_files_demo():
    # 合并多個Excel文件
    ExcelUtils.merge_excel_files(
        input_files=["user.xlsx", "multi_sheet_data.xlsx"],
        output_file="merged_data.xlsx",
        sheet_name_mapping={
            "user.xlsx": "user",
            "multi_sheet_data.xlsx": "multi_sheet_data"
        }
    )