python - 如何使用 python win32com 在 Excel 中创建数据透视表

Question

• 给定一个现有的 Excel 文件，其中数据采用长格式

  • 

• 使用 Python 在 Excel 中自动创建以下数据透视表 win32com模块

  • 

• 以下代码用于使用数据设置 test.xlsx 并连接以创建 Excel com 对象

导入

import win32com.client as win32
from pathlib import Path
import sys
import pandas as pd
import numpy as np
import random
from datetime import datetime

win32c = win32.constants

创建test.xlsx的函数

• 此功能仅提供测试数据和文件

def create_test_excel_file(f_path: Path, f_name: str, sheet_name: str):
    
    filename = f_path / f_name
    random.seed(365)
    np.random.seed(365)
    number_of_data_rows = 1000
    
    # create list of 31 dates
    dates = pd.bdate_range(datetime(2020, 7, 1), freq='1d', periods=31).tolist()

    data = {'date': [random.choice(dates) for _ in range(number_of_data_rows)],
            'expense': [random.choice(['business', 'personal']) for _ in range(number_of_data_rows)],
            'products': [random.choice(['book', 'ribeye', 'coffee', 'salmon', 'alcohol', 'pie']) for _ in range(number_of_data_rows)],
            'price': np.random.normal(15, 5, size=(1, number_of_data_rows))[0]}

    pd.DataFrame(data).to_excel(filename, index=False, sheet_name=sheet_name, float_format='%.2f')

创建 Excel com 对象的函数

def run_excel(f_path: Path, f_name: str, sheet_name: str):

    filename = f_path / f_name

    # create excel object
    excel = win32.gencache.EnsureDispatch('Excel.Application')

    # excel can be visible or not
    excel.Visible = True  # False
    
    # try except for file / path
    try:
        wb = excel.Workbooks.Open(filename)
    except com_error as e:
        if e.excepinfo[5] == -2146827284:
            print(f'Failed to open spreadsheet.  Invalid filename or location: {filename}')
        else:
            raise e
        sys.exit(1)

    # set worksheet
    ws1 = wb.Sheets('data')
    
#     wb.Close(True)
#     excel.Quit()

主要

def main():
    # sheet name for data
    sheet_name = 'data'  # update with sheet name from your file
    # file path
    f_path = Path.cwd()  # file in current working directory
#   f_path = Path(r'c:\...\Documents')  # file located somewhere else
    # excel file
    f_name = 'test.xlsx'
    
    # function calls
    create_test_excel_file(f_path, f_name, sheet_name)  # remove when running your own file
    run_excel(f_path, f_name, sheet_name)

Answer 1

• 确定要使用的正确 Excel 方法的一个有用方法是在 Excel 中记录分步宏，同时以所需的形式创建数据透视表。
  • 这对于创建必须在包含现有数据的文件中定期运行的数据透视表非常有用。
• 使用问题中的导入和方法
• 修改此代码以创建新的数据文件
  • 更新def main
    • sheet_name
    • f_path
    • f_name
  • 更新def run_excel
    • ws1
    • ws2_name
    • pt_name
    • pt_rows
    • pt_cols
    • pt_filters
    • pt_fields
  • 调用main()来运行代码

pivot_table 函数

def pivot_table(wb: object, ws1: object, pt_ws: object, ws_name: str, pt_name: str, pt_rows: list, pt_cols: list, pt_filters: list, pt_fields: list):
    """
    wb = workbook1 reference
    ws1 = worksheet1
    pt_ws = pivot table worksheet number
    ws_name = pivot table worksheet name
    pt_name = name given to pivot table
    pt_rows, pt_cols, pt_filters, pt_fields: values selected for filling the pivot tables
    """

    # pivot table location
    pt_loc = len(pt_filters) + 2
    
    # grab the pivot table source data
    pc = wb.PivotCaches().Create(SourceType=win32c.xlDatabase, SourceData=ws1.UsedRange)
    
    # create the pivot table object
    pc.CreatePivotTable(TableDestination=f'{ws_name}!R{pt_loc}C1', TableName=pt_name)

    # selecte the pivot table work sheet and location to create the pivot table
    pt_ws.Select()
    pt_ws.Cells(pt_loc, 1).Select()

    # Sets the rows, columns and filters of the pivot table

    for field_list, field_r in ((pt_filters, win32c.xlPageField), (pt_rows, win32c.xlRowField), (pt_cols, win32c.xlColumnField)):
        for i, value in enumerate(field_list):
            pt_ws.PivotTables(pt_name).PivotFields(value).Orientation = field_r
            pt_ws.PivotTables(pt_name).PivotFields(value).Position = i + 1

    # Sets the Values of the pivot table
    for field in pt_fields:
        pt_ws.PivotTables(pt_name).AddDataField(pt_ws.PivotTables(pt_name).PivotFields(field[0]), field[1], field[2]).NumberFormat = field[3]

    # Visiblity True or Valse
    pt_ws.PivotTables(pt_name).ShowValuesRow = True
    pt_ws.PivotTables(pt_name).ColumnGrand = True

更新run_excel以调用pivot_table

def run_excel(f_path: Path, f_name: str, sheet_name: str):

    filename = f_path / f_name

    # create excel object
    excel = win32.gencache.EnsureDispatch('Excel.Application')

    # excel can be visible or not
    excel.Visible = True  # False
    
    # try except for file / path
    try:
        wb = excel.Workbooks.Open(filename)
    except com_error as e:
        if e.excepinfo[5] == -2146827284:
            print(f'Failed to open spreadsheet.  Invalid filename or location: {filename}')
        else:
            raise e
        sys.exit(1)

    # set worksheet
    ws1 = wb.Sheets('data')
    
    # Setup and call pivot_table
    ws2_name = 'pivot_table'
    wb.Sheets.Add().Name = ws2_name
    ws2 = wb.Sheets(ws2_name)
    
    pt_name = 'example'
    pt_rows = ['expense']
    pt_cols = ['products']
    pt_filters = ['date']
    # [0]: field name [1]: pivot table column name [3]: calulation method [4]: number format
    pt_fields = [['price', 'price: mean', win32c.xlAverage, '$#,##0.00'],  
                 ['price', 'price: sum', win32c.xlSum, '$#,##0.00'],
                 ['price', 'price: count', win32c.xlCount, '0']]
    
    pivot_table(wb, ws1, ws2, ws2_name, pt_name, pt_rows, pt_cols, pt_filters, pt_fields)
    
#     wb.Close(True)
#     excel.Quit()

资源

• Jupyter Notebook: How to Create a Pivot Table in Excel with the Python win32com Module
• Automate Excel with Python
  • 数据透视表示例
• Using Python win32com to get list of Excel worksheets
• Excel VBA reference
• Workbook object (Excel)
• Worksheet object (Excel)