如何使用 Pandas 在现有 excel 文件中保存新工作表？

Question

I want to use excel files to store data elaborated with python. My problem is that I can't add sheets to an existing excel file. Here I suggest a sample code to work with in order to reach this issue

import pandas as pd
import numpy as np

path = r"C:\Users\fedel\Desktop\excelData\PhD_data.xlsx"

x1 = np.random.randn(100, 2)
df1 = pd.DataFrame(x1)

x2 = np.random.randn(100, 2)
df2 = pd.DataFrame(x2)

writer = pd.ExcelWriter(path, engine = 'xlsxwriter')
df1.to_excel(writer, sheet_name = 'x1')
df2.to_excel(writer, sheet_name = 'x2')
writer.save()
writer.close()

This code saves two DataFrames to two sheets, named "x1" and "x2" respectively. If I create two new DataFrames and try to use the same code to add two new sheets, 'x3' and 'x4', the original data is lost.

import pandas as pd
import numpy as np

path = r"C:\Users\fedel\Desktop\excelData\PhD_data.xlsx"

x3 = np.random.randn(100, 2)
df3 = pd.DataFrame(x3)

x4 = np.random.randn(100, 2)
df4 = pd.DataFrame(x4)

writer = pd.ExcelWriter(path, engine = 'xlsxwriter')
df3.to_excel(writer, sheet_name = 'x3')
df4.to_excel(writer, sheet_name = 'x4')
writer.save()
writer.close()

I want an excel file with four sheets: 'x1', 'x2', 'x3', 'x4'. I know that 'xlsxwriter' is not the only "engine", there is 'openpyxl'. I also saw there are already other people that have written about this issue, but still I can't understand how to do that.

Here a code taken from this link

import pandas
from openpyxl import load_workbook

book = load_workbook('Masterfile.xlsx')
writer = pandas.ExcelWriter('Masterfile.xlsx', engine='openpyxl') 
writer.book = book
writer.sheets = dict((ws.title, ws) for ws in book.worksheets)

data_filtered.to_excel(writer, "Main", cols=['Diff1', 'Diff2'])

writer.save()

They say that it works, but it is hard to figure out how. I don't understand what "ws.title", "ws", and "dict" are in this context.

Which is the best way to save "x1" and "x2", then close the file, open it again and add "x3" and "x4"?

Answer 1

Thank you. I believe that a complete example could be good for anyone else who have the same issue:

import pandas as pd
import numpy as np

path = r"C:\Users\fedel\Desktop\excelData\PhD_data.xlsx"

x1 = np.random.randn(100, 2)
df1 = pd.DataFrame(x1)

x2 = np.random.randn(100, 2)
df2 = pd.DataFrame(x2)

writer = pd.ExcelWriter(path, engine = 'xlsxwriter')
df1.to_excel(writer, sheet_name = 'x1')
df2.to_excel(writer, sheet_name = 'x2')
writer.save()
writer.close()

Here I generate an excel file, from my understanding it does not really matter whether it is generated via the "xslxwriter" or the "openpyxl" engine.

When I want to write without loosing the original data then

import pandas as pd
import numpy as np
from openpyxl import load_workbook

path = r"C:\Users\fedel\Desktop\excelData\PhD_data.xlsx"

book = load_workbook(path)
writer = pd.ExcelWriter(path, engine = 'openpyxl')
writer.book = book

x3 = np.random.randn(100, 2)
df3 = pd.DataFrame(x3)

x4 = np.random.randn(100, 2)
df4 = pd.DataFrame(x4)

df3.to_excel(writer, sheet_name = 'x3')
df4.to_excel(writer, sheet_name = 'x4')
writer.save()
writer.close()

this code do the job!

Answer 2

In the example you shared you are loading the existing file into book and setting the writer.book value to be book . In the line writer.sheets = dict((ws.title, ws) for ws in book.worksheets) you are accessing each sheet in the workbook as ws . The sheet title is then ws so you are creating a dictionary of {sheet_titles: sheet} key, value pairs. This dictionary is then set to writer.sheets. Essentially these steps are just loading the existing data from 'Masterfile.xlsx' and populating your writer with them.

Now let's say you already have a file with x1 and x2 as sheets. You can use the example code to load the file and then could do something like this to add x3 and x4 .

path = r"C:\Users\fedel\Desktop\excelData\PhD_data.xlsx"
writer = pd.ExcelWriter(path, engine='openpyxl')
df3.to_excel(writer, 'x3', index=False)
df4.to_excel(writer, 'x4', index=False)
writer.save()

That should do what you are looking for.

Answer 3

A simple example for writing multiple data to excel at a time. And also when you want to append data to a sheet on a written excel file (closed excel file).

When it is your first time writing to an excel. (Writing "df1" and "df2" to "1st_sheet" and "2nd_sheet")

import pandas as pd 
from openpyxl import load_workbook

df1 = pd.DataFrame([[1],[1]], columns=['a'])
df2 = pd.DataFrame([[2],[2]], columns=['b'])
df3 = pd.DataFrame([[3],[3]], columns=['c'])

excel_dir = "my/excel/dir"

with pd.ExcelWriter(excel_dir, engine='xlsxwriter') as writer:    
    df1.to_excel(writer, '1st_sheet')   
    df2.to_excel(writer, '2nd_sheet')   
    writer.save()    

After you close your excel, but you wish to "append" data on the same excel file but another sheet, let's say "df3" to sheet name "3rd_sheet".

book = load_workbook(excel_dir)
with pd.ExcelWriter(excel_dir, engine='openpyxl') as writer:
    writer.book = book
    writer.sheets = dict((ws.title, ws) for ws in book.worksheets)    

    ## Your dataframe to append. 
    df3.to_excel(writer, '3rd_sheet')  

    writer.save()     

Be noted that excel format must not be xls, you may use xlsx one.

Answer 4

For creating a new file

x1 = np.random.randn(100, 2)
df1 = pd.DataFrame(x1)
with pd.ExcelWriter('sample.xlsx') as writer:  
    df1.to_excel(writer, sheet_name='x1')

For appending to the file, use the argument mode='a' in pd.ExcelWriter .

x2 = np.random.randn(100, 2)
df2 = pd.DataFrame(x2)
with pd.ExcelWriter('sample.xlsx', engine='openpyxl', mode='a') as writer:  
    df2.to_excel(writer, sheet_name='x2')

Default is mode ='w' . See documentation .

Answer 5

I would strongly recommend you work directly with openpyxl since it now supports Pandas DataFrames .

This allows you to concentrate on the relevant Excel and Pandas code.

Answer 6

Can do it without using ExcelWriter, using tools in openpyxl This can make adding fonts to the new sheet much easier using openpyxl.styles

import pandas as pd
from openpyxl import load_workbook
from openpyxl.utils.dataframe import dataframe_to_rows

#Location of original excel sheet
fileLocation =r'C:\workspace\data.xlsx'

#Location of new file which can be the same as original file
writeLocation=r'C:\workspace\dataNew.xlsx'

data = {'Name':['Tom','Paul','Jeremy'],'Age':[32,43,34],'Salary':[20000,34000,32000]}

#The dataframe you want to add
df = pd.DataFrame(data)

#Load existing sheet as it is
book = load_workbook(fileLocation)
#create a new sheet
sheet = book.create_sheet("Sheet Name")

#Load dataframe into new sheet
for row in dataframe_to_rows(df, index=False, header=True):
    sheet.append(row)

#Save the modified excel at desired location    
book.save(writeLocation)

Answer 7

You can read existing sheets of your interests, for example, 'x1', 'x2', into memory and 'write' them back prior to adding more new sheets (keep in mind that sheets in a file and sheets in memory are two different things, if you don't read them, they will be lost). This approach uses 'xlsxwriter' only, no openpyxl involved.

import pandas as pd
import numpy as np

path = r"C:\Users\fedel\Desktop\excelData\PhD_data.xlsx"

# begin <== read selected sheets and write them back
df1 = pd.read_excel(path, sheet_name='x1', index_col=0) # or sheet_name=0
df2 = pd.read_excel(path, sheet_name='x2', index_col=0) # or sheet_name=1
writer = pd.ExcelWriter(path, engine='xlsxwriter')
df1.to_excel(writer, sheet_name='x1')
df2.to_excel(writer, sheet_name='x2')
# end ==>

# now create more new sheets
x3 = np.random.randn(100, 2)
df3 = pd.DataFrame(x3)

x4 = np.random.randn(100, 2)
df4 = pd.DataFrame(x4)

df3.to_excel(writer, sheet_name='x3')
df4.to_excel(writer, sheet_name='x4')
writer.save()
writer.close()

If you want to preserve all existing sheets, you can replace above code between begin and end with:

# read all existing sheets and write them back
writer = pd.ExcelWriter(path, engine='xlsxwriter')
xlsx = pd.ExcelFile(path)
for sheet in xlsx.sheet_names:
    df = xlsx.parse(sheet_name=sheet, index_col=0)
    df.to_excel(writer, sheet_name=sheet)

Answer 8

Another fairly simple way to go about this is to make a method like this:

def _write_frame_to_new_sheet(path_to_file=None, sheet_name='sheet', data_frame=None):
    book = None
    try:
        book = load_workbook(path_to_file)
    except Exception:
        logging.debug('Creating new workbook at %s', path_to_file)
    with pd.ExcelWriter(path_to_file, engine='openpyxl') as writer:
        if book is not None:
            writer.book = book
        data_frame.to_excel(writer, sheet_name, index=False)

The idea here is to load the workbook at path_to_file if it exists and then append the data_frame as a new sheet with sheet_name . If the workbook does not exist, it is created. It seems that neither openpyxl or xlsxwriter append, so as in the example by @Stefano above, you really have to load and then rewrite to append.

Answer 9

#This program is to read from excel workbook to fetch only the URL domain names and write to the existing excel workbook in a different sheet..
#Developer - Nilesh K
import pandas as pd
from openpyxl import load_workbook #for writting to the existing workbook

df = pd.read_excel("urlsearch_test.xlsx")

#You can use the below for the relative path.
# r"C:\Users\xyz\Desktop\Python\

l = [] #To make a list in for loop

#begin
#loop starts here for fetching http from a string and iterate thru the entire sheet. You can have your own logic here.
for index, row in df.iterrows():
    try: 
        str = (row['TEXT']) #string to read and iterate
        y = (index)
        str_pos = str.index('http') #fetched the index position for http
        str_pos1 = str.index('/', str.index('/')+2) #fetched the second 3rd position of / starting from http
        str_op = str[str_pos:str_pos1] #Substring the domain name
        l.append(str_op) #append the list with domain names

    #Error handling to skip the error rows and continue.
    except ValueError:
            print('Error!')
print(l)
l = list(dict.fromkeys(l)) #Keep distinct values, you can comment this line to get all the values
df1 = pd.DataFrame(l,columns=['URL']) #Create dataframe using the list
#end

#Write using openpyxl so it can be written to same workbook
book = load_workbook('urlsearch_test.xlsx')
writer = pd.ExcelWriter('urlsearch_test.xlsx',engine = 'openpyxl')
writer.book = book
df1.to_excel(writer,sheet_name = 'Sheet3')
writer.save()
writer.close()

#The below can be used to write to a different workbook without using openpyxl
#df1.to_excel(r"C:\Users\xyz\Desktop\Python\urlsearch1_test.xlsx",index='false',sheet_name='sheet1')

Answer 10

Every time you want to save a Pandas DataFrame to an Excel, you may call this function:

import os

def save_excel_sheet(df, filepath, sheetname, index=False):
    # Create file if it does not exist
    if not os.path.exists(filepath):
        df.to_excel(filepath, sheet_name=sheetname, index=index)

    # Otherwise, add a sheet. Overwrite if there exists one with the same name.
    else:
        with pd.ExcelWriter(filepath, engine='openpyxl', if_sheet_exists='replace', mode='a') as writer:
            df.to_excel(writer, sheet_name=sheetname, index=index)

Answer 11

if you want to add empty sheet

xw = pd.ExcelWriter(file_path, engine='xlsxwriter')    
pd.DataFrame().to_excel(xw, 'sheet11')

if you get empty sheet

sheet = xw.sheets['sheet11']

Answer 12

The following solution worked for me:

df = pd.DataFrame({"A":[1,2], "B":[3,4]})

path = "./..../..../.../test.xlsx"

if os.path.isfile(path):  
  with pd.ExcelWriter(path, mode='a') as writer:
    df.to_excel(writer, sheet_name= "sheet_2")
else:
  with pd.ExcelWriter(path) as writer:
    df.to_excel(writer, sheet_name= "sheet_1")

Answer 13

import pandas as pd
import openpyxl

writer = pd.ExcelWriter('test.xlsx', engine='openpyxl')
data_df.to_excel(writer, 'sheet_name')
writer.save()
writer.close()