Как создать сводную таблицу в Excel с помощью python win32com

  • Учитывая существующий файл Excel с данными в длинном формате

    • enter image description here

  • Автоматизируйте создание следующей сводной таблицы в Excel с помощью Python win32com модуль

    • enter image description here

  • Ниже приведен код для настройки test.xlsx с данными и подключения для создания объекта Excel com.

Импорт

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

Функция для создания test.xlsx

  • Эта функция предназначена только для предоставления тестовых данных и файла
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')

Функция для создания объекта Excel com

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()

Главный

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)

python excel pivot-table win32com
person Trenton McKinney    schedule 22.06.2020    source источник

Ответы (1)


arrow_upward
0
arrow_downward

  • A helpful way to figure out the proper Excel methods to use, is record a step-by-step Macro in Excel, while creating a pivot table in the form you want.
    • This is useful for creating a pivot table that has to be run on a routine basis in a file with existing data.
  • Использует импорт и методы из вопроса
  • To modify this code for a new data file
    • Update def main
      • sheet_name
      • f_path
      • f_name
    • Update def run_excel
      • ws1
      • ws2_name
      • pt_name
      • pt_rows
      • pt_cols
      • pt_filters
      • pt_fields
    • Позвоните main(), чтобы запустить код

pivot_table функция

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

Обновите run_excel для вызова 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()

Ресурсы

person Trenton McKinney    schedule 22.06.2020