Search icon CANCEL
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Conferences
Free Learning
Arrow right icon
Arrow up icon
GO TO TOP
Extending Excel with Python and R

You're reading from   Extending Excel with Python and R Unlock the potential of analytics languages for advanced data manipulation and visualization

Arrow left icon
Product type Paperback
Published in Apr 2024
Publisher Packt
ISBN-13 9781804610695
Length 344 pages
Edition 1st Edition
Languages
Tools
Arrow right icon
Authors (2):
Arrow left icon
Steven Sanderson Steven Sanderson
Author Profile Icon Steven Sanderson
Steven Sanderson
David Kun David Kun
Author Profile Icon David Kun
David Kun
Arrow right icon
View More author details
Toc

Table of Contents (20) Chapters Close

Preface 1. Part 1:The Basics – Reading and Writing Excel Files from R and Python FREE CHAPTER
2. Chapter 1: Reading Excel Spreadsheets 3. Chapter 2: Writing Excel Spreadsheets 4. Chapter 3: Executing VBA Code from R and Python 5. Chapter 4: Automating Further – Task Scheduling and Email 6. Part 2: Making It Pretty – Formatting, Graphs, and More
7. Chapter 5: Formatting Your Excel Sheet 8. Chapter 6: Inserting ggplot2/matplotlib Graphs 9. Chapter 7: Pivot Tables and Summary Tables 10. Part 3: EDA, Statistical Analysis, and Time Series Analysis
11. Chapter 8: Exploratory Data Analysis with R and Python 12. Chapter 9: Statistical Analysis: Linear and Logistic Regression 13. Chapter 10: Time Series Analysis: Statistics, Plots, and Forecasting 14. Part 4: The Other Way Around – Calling R and Python from Excel
15. Chapter 11: Calling R/Python Locally from Excel Directly or via an API 16. Part 5: Data Analysis and Visualization with R and Python for Excel Data – A Case Study
17. Chapter 12: Data Analysis and Visualization with R and Python in Excel – A Case Study 18. Index 19. Other Books You May Enjoy

Reading Excel files to R

In this section, we are going to read data from Excel with a few different R libraries. We need to do this before we can even consider performing any type of manipulation or analysis on the data contained in the sheets of the Excel files.

As mentioned in the Technical requirements section, we are going to be using the readxl, openxlsx, and xlsx packages to read data into R.

Installing and loading libraries

In this section, we are going to install and load the necessary libraries if you do not yet have them. We are going to use the openxlsx, xlsx, readxl, and readxlsb libraries. To install and load them, run the following code block:

pkgs <- c("openxlsx", "xlsx", "readxl")
install.packages(pkgs, dependencies = TRUE)
lapply(pkgs, library, character.only = TRUE)

The lapply() function in R is a versatile tool for applying a function to each element of a list, vector, or DataFrame. It takes two arguments, x and FUN, where x is the list and FUN is the function that is applied to the list object, x.

Now that the libraries have been installed, we can get to work. To do this, we are going to read a spreadsheet built from the Iris dataset that is built into base R. We are going to read the file with three different libraries, and then we are going to create a custom function to work with the readxl library that will read all the sheets of an Excel file. We will call this the read_excel_sheets() function.

Let’s start reading the files. The first library we will use to open an Excel file is openxlsx. To read the Excel file we are working with, you can run the code in the chapter1 folder of this book’s GitHub repository called ch1_create_iris_dataset.R Refer to the following screenshot to see how to read the file into R.

You will notice a variable called f_pat. This is the path to where the Iris dataset was saved as an Excel file – for example, C:/User/UserName/Documents/iris_data.xlsx:

Figure 1.1 – Using the openxlsx package to read the Excel file

Figure 1.1 – Using the openxlsx package to read the Excel file

The preceding screenshot shows how to read an Excel file. This example assumes that you have used the ch1_create_iris_datase.R file to create the example Excel file. In reality, you can read in any Excel file that you would like or need.

Now, we will perform the same type of operation, but this time with the xlsx library. Refer to the following screenshot, which uses the same methodology as with the openxlsx package:

Figure 1.2 – Using the xlsx library and the read.xlsx() function to open the Excel file we’ve created

Figure 1.2 – Using the xlsx library and the read.xlsx() function to open the Excel file we’ve created

Finally, we will use the readxl library, which is part of the tidyverse:

Figure 1.3 – Using the readxl library and the read_excel() function to read the Excel file into memory

Figure 1.3 – Using the readxl library and the read_excel() function to read the Excel file into memory

In this section, we learned how to read in an Excel file with a few different packages. While these packages can do more than simply read in an Excel file, that is what we needed to focus on in this section. You should now be familiar with how to use the readxl::read_excel(), xlsx::read.xlsx(), and openxlsx::read.xlsx() functions.

Building upon our expertise in reading Excel files into R, we’ll now embark on the next phase of our journey: unraveling the secrets of efficiently extracting data from multiple sheets within an Excel file.

lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime