python处理Excel Pandas xlwings numpy

# 批量创建Excel
import  xlwings
#   xw.App(visible=True,add_book=True)  会打开Excel，且不会自动关闭
#	xw.App(visible=True,add_book=True)	会打开Excel，但一晃就自动关闭了
app = xlwings.App(visible=True, add_book=False)
for language in ['Java', 'Python', 'C#', 'Vue', "JavaScript"]:
    workbook = app.books.add()
    workbook.save(f"./畅销开发语言--{
      
      language}.xlsx")

# 批量打开Excel
import os
import  xlwings as xw
app = xw.App(visible=True,add_book=False)
# os.listdir(path)  列出指定目录下的内容
for file in  os.listdir("."):
    if file.endswith('.xlsx') or file.endswith('.xlsx'):
        app.books.open(file)
        

# 批量重命名工作表
import xlwings as xw
app = xw.App(visible=False, add_book=False)
workbook = app.books.open("畅销开发语言--Python.xlsx")
for sheet in workbook.sheets:
    sheet.name = sheet.name.replace('Sheet', 'ZEN')
workbook.save()
app.quit()

#  合并Excel文件
import pandas as pd
import os

data_list = []
for filename in os.listdir('.'):
    if filename.startswith('畅销开发语言--') and filename.endswith('.xlsx'):
    	# pd.read_excel("xx.xlsx", sheet_name=None)  
    	# sheet_name 默认值0 ，也就是默认打开Excel表中第一个工作簿
        data_list.append(pd.read_excel(filename))
   
data_all = pd.concat(data_list)
data_all.to_excel("合并表.xlsx", index=False)

# 把一个Excel的所有工作表合并，且结果插入第一个位置
# 只是把其它sheet表复制到首个，并没有汇总。
import pandas as pd
import os
import  xlwings as xw

# pd.read_excel("xx.xlsx", sheet_name=None)
# sheet_name 默认值0 ，也就是默认打开Excel表中第一个工作簿
# sheet_name=None 打开所有sheet 工作簿
data_list = pd.read_excel("xxx.xlsx", sheet_name=None)
print(data_list)
data_all = pd.concat(data_list.values())

app = xw.App(visible=False, add_book=False)

workbook = app.books.open("xxx.xlsx")
workbook.sheets.add("汇总表", before=workbook.sheets[0])
workbook.sheets["汇总表"].range("A1").options(index=False).value = data_all

workbook.save()
workbook.close()
app.quit()

#  把Excel 工作表 拆分多个Excel文件  按course 列拆分
#	Excel 列 course 	Total
import pandas as pd
data_list = pd.read_excel("xxx.xlsx", sheet_name=0)
courses = data_list["course"].unique()
for course in courses:
    data_single = data_list[data_list["course"] == course]
    data_single.to_excel(f"拆分数据-{
      
      course}.xlsx")

# 批量合并拆分Excel
import pandas as pd
#  读取excel所有工作表
#  此处的Excel工作表有 采购日期    采购物品    采购数量    采购金额
data_list = pd.read_excel("A.xlsx", sheet_name=None, parse_dates=False)
# print(data_list.keys())
#  把多个工作表合并在一起
#   注意是 values()，只合并sheet里面信息， 如果不加，也会有sheet的名称
data_all = pd.concat(data_list.values())

excel_writer = pd.ExcelWriter('采购表-按采购物品.xlsx', date_format="YYYY_MM_DD")

for product, data_all in data_all.groupby("采购物品"):
    data_all.to_excel(excel_writer, product, index=False)

excel_writer.close()

顺序不变，是可以的
如果李四和王五两行换一下，就不对了，

import xlwings as xw
app = xw.App(visible=True, add_book=False)
data = app.books.open("A.xlsx")
data_back = app.books.open("A - 副本.xlsx")

for cell in data.sheets[0].range("A1").expand():
    # cell.address  就是Excel  A1，B1   单元格地址
    back_cell = data_back.sheets[0].range(cell.address)
    if cell.value != back_cell.value:
        cell.color = back_cell.color = (255, 0, 0)
data.save()
data.close()
data_back.save()
data_back.close()
app.quit()

#  把文件AA下面所有Excel文件，规格表中规格列拆分三列，同时删除规格列
import xlwings as xw
import pandas as pd
import os

app = xw.App(visible=False, add_book=False)

for fname in os.listdir('AA'):
    if fname.endswith('.xlsx'):

        workbook = app.books.open(os.path.join('AA', fname))
        worksheet = workbook.sheets["规格表"]
        df = worksheet.range("A1").options(pd.DataFrame, expand='table').value
        worksheet.range("A1").options(pd.DataFrame)       
        split_columns = df["规格"].str.split("*", expand=True)
        df["长"] = split_columns[0]
        df["宽"] = split_columns[1]
        df["高"] = split_columns[2]
        # inplace  原地
        df.drop("规格", inplace=True)
        worksheet.range("A1").value = df
        workbook.save()
app.quit()

# 把A.xlsx的所有sheet中物品名称列，获取，并写入另外一个Excel中
import pandas as pd

df_list = pd.read_excel("A.xlsx",sheet_name=None)

df_all = pd.concat(df_list.values())

df_names = pd.DataFrame(data={
    
    "物品名称：":list(df_all["物品名称"].unique())})

df_names.to_excel("Result.xlsx",index=False)

#  批量分类统计
import pandas as pd
import xlwings as xw
import os
app = xw.App(visible=False, add_book=False)

for file in os.listdir("List"):
    if file.endswith(".xlsx") and not file.startswith("~$"):
        workbook = app.books.open(f"List/{
      
      file}")
        #  第一个sheet表
        worksheet = workbook.sheets[0]
        # 将A1转换DataFrame对象
        df = worksheet.range("A1").options(pd.DataFrame, expand='table').value
        # 把GDP数据类型设置Float
        df["GDP"] = df["GDP"].astype(float)
        df_agg = df.groupby("城市")["GDP"].sum()
        #  默认是按行
        worksheet.range("F1").value = df_agg
        workbook.save()
        workbook.close()

app.quit()

结果

#  实现多个Excel vlookup
import pandas as pd
import xlwings as xw
import os
app = xw.App(visible=False, add_book=False)
workbook = app.books.open("GDP.xlsx")
df_total = workbook.sheets[0].range("A1").options(pd.DataFrame, expand='table', index=False).value
df_city_list = []
for file in os.listdir("List"):
    if file.endswith(".xlsx") and not file.startswith("~$") and "GDP" in file:
        workbook_list = app.books.open(f"List/{
      
      file}")
        #  第一个sheet表
        df_city = workbook_list.sheets[0].range("A1").options(pd.DataFrame, expand='table', index=False).value
        df_city["省份"] = file.replace("GDP.xlsx", "")
        df_city_list.append(df_city)
        workbook_list.close()
        df_city_all = pd.concat(df_city_list)
        # 把GDP数据类型设置Float
        # left、right:需要连接的两个DataFrame或Series，一左一右
        # left_on:左表的连接键字段        # 
        # right_on:右表的连接键字段
        df_merge = pd.merge(
            left=df_total,
            right=df_city_all,
            left_on=["省份", "城市"],
            right_on=["省份", "城市"]
        )
        df_merge["GDP"]=df_merge["GDP1"]
        df_merge.drop(columns="GDP1", inplace=True)
        workbook.sheets[0].range("A1").options(index=False).value = df_merge
    else:
        continue

workbook.save()
workbook.close()
app.quit()

import pandas as pd
import xlwings as xw
import os
app = xw.App(visible=False, add_book=False)
data_list = []
for file in os.listdir("List"):
    if file.endswith(".xlsx"):
        workbook = app.books.open(f"List/{
      
      file}")
        df_list = workbook.sheets[0].range("A1").options(pd.DataFrame, expand='table').value
        df_list["品牌"] = file.replace("手机.xlsx","")
        data_list.append(df_list)
        workbook.close()
        # if 之间的代码可以简写这样
        df = pd.read_excel(f"List/{
      
      file}")
        df["品牌"]=file.replace("手机.xlsx","")
        print(df)
        data_list.append(df)

def compute(df_sub):
    return pd.Series({
    
    
        "总和": round(df_sub["售价"].sum(), 2),
        "最小": round(df_sub["售价"].min(), 2),
        "最大": round(df_sub["售价"].max(), 2),
        "平均": round(df_sub["售价"].mean(), 2)
    })


data_all = pd.concat(data_list)
# print(data_all)
#  apply(compute)  compute 自定义函数，没有（参数）
df_group = data_all.groupby("品牌").apply(compute)


df_group.to_excel("按品牌汇总统计.xlsx")

app.quit()

#  数据透视表：把列式数据转换成二位交叉形式，便于分析
#  姓名	课程	分数	转换成	姓名  语文	数学	英语  


# 数据透视表

import pandas as pd

import os
# pd.read_excel 结果是  DataFrame
data_all = pd.read_excel('Result.xlsx')
# index 是列表 ['姓名','学号']
# PIVOT 在数据库 表示列行转换
data_pivot = pd.pivot_table(data_all,
               index=["姓名"],
               columns="课程",
               values="分数"
               )
data_pivot.to_excel("透视表.xlsx")



#   效果同上
import pandas as pd

import os

data_all = []
for file in os.listdir('.'):
    if file.endswith('.xlsx'):
        data_all.append(pd.read_excel(file))
# index 是列表 ['姓名','学号']
# PIVOT 在数据库 表示列行转换
# print(type(data_all))   list
# print(type(pd.concat(data_all)))  pandas.core.frame.DataFrame
data_pivot = pd.pivot_table(pd.concat(data_all),
               index=["姓名"],
               columns="课程",
               values="分数"
               )
data_pivot.to_excel("透视表.xlsx")

# 一个Excel多个sheet表，合并透视表，追加合计
import pandas as pd
dfs= pd.read_excel("Result.xlsx",sheet_name=None)
df_list= []
for sheet_name, df in dfs.items():
    print(sheet_name)
    print(df)
    df["月份"]=sheet_name
    df_list.append(df)

data_all = pd.concat(df_list)

data_pivot = pd.pivot_table(data_all,
                            index=['产品名称'],
                            columns='月份',
                            values='销售金额',
                            aggfunc="sum",
                            fill_value=0,
                            margins=True,
                            margins_name="合计"
                            )
data_pivot.to_excel("透视表.xlsx")

pandas的nlargest(n,“排序的列”)，只能求最大N个值

import pandas as pd
dfs= pd.read_excel("Result.xlsx",sheet_name=None)
df_list= []
for sheet_name, df in dfs.items():
    print(sheet_name)
    print(df)
    df["班级"]=sheet_name
    df_list.append(df)

data_all = pd.concat(df_list)

data_all.groupby("班级").apply(lambda x: x.nlargest(2, "分数")).to_excel("透视表.xlsx")

import xlwings as xw
import numpy

app = xw.App(visible=False,add_book=False)
workbook =app.books.open("Result.xlsx")
sheet = workbook.sheets[0]
# 统计员工人数
employ_total = sheet.range("A3").expand('table').shape[0]
# permutation(10),随机生成0-9 10位随机数
employ_GH = numpy.random.permutation(employ_total)+1
# options(transpose=True)列模式
sheet.range("B3").options(transpose=True).value = employ_GH
workbook.save()
workbook.close()
app.quit()