LDATA607LAB6

Introduction

In this project, I will explore two common formats for storing and exchanging data: HTML and JSON. Three programming-related books were selected and manually recorded in both formats. This assignment required knowledge of and familiarity with the structure of HTML tables and JSON objects, as well as the ability to import data from these formats into R. After loading both files into separate data frames, the results are compared to determine whether they contain the same information.

Approach To complete this project, I selected three books related to computer programming: Head First Java, HTML and CSS: Design and Build Websites, and Python Crash Course. For each book, I recorded the title, author(s), publication year, publisher, and programming language.

The book information was manually entered into two separate files:

• An HTML file containing a table of book data.

• A JSON file containing the same information in JSON format.

Using R, I  will load the HTML file into a data frame with the rvest package and load the JSON file into a separate data frame with the jsonlite package. Finally, I will compare the two data frames to verify that they contain identical data.

Challenges

One challenge in this project was ensuring that the HTML and JSON files contained exactly the same information and structure. Although the data may appear identical, differences in column names or data types can cause the resulting data frames to differ when imported into R. Another challenge was understanding the different ways that data is represented in HTML and JSON. HTML stores the information in a table format, while JSON stores the same information using key-value pairs. Creating both files manually helped reinforce the structure and purpose of each format.

Import Library

library(tidyverse)
library(tidyr)
library(dplyr)
library(ggplot2)
library(tibble)
library(jsonlite)
library(xml2)
library(rvest)

Load HTML Data

In this section, The HTML file containing the programming book information is loaded into R. The table is extracted from the HTML file from my GitHub repository and stored as a data frame for further analysis.

html_df <- read_html("https://raw.githubusercontent.com/lioneljr17/LDATA607/refs/heads/main/607LAB6/Books.html") |>
  html_table((fill = TRUE))

html_df  <- html_df[[1]]

html_df <-as.data.frame(html_df)

Load JSON Data

In this section. The JSON file containing the same book information is loaded into a separate data frame using the jsonlite package.

json_df <- fromJSON("https://raw.githubusercontent.com/lioneljr17/LDATA607/refs/heads/main/607LAB6/Books.json")

json_df

                                    Title                  Authors
1                         Head First Java Kathy Sierra, Bert Bates
2 HTML and CSS: Design and Build Websites              Jon Duckett
3                     Python Crash Course             Eric Matthes
  Publication_Year       Publisher Language
1             2022  O'Reilly Media     Java
2             2011           Wiley HTML/CSS
3             2023 No Starch Press   Python

str(html_df)

'data.frame':   3 obs. of  5 variables:
 $ Title           : chr  "Head First Java" "HTML and CSS: Design and Build Websites" "Python Crash Course"
 $ Authors         : chr  "Kathy Sierra, Bert Bates" "Jon Duckett" "Eric Matthes"
 $ Publication Year: int  2022 2011 2023
 $ Publisher       : chr  "O'Reilly Media" "Wiley" "No Starch Press"
 $ Language        : chr  "Java" "HTML/CSS" "Python"

str(json_df)

'data.frame':   3 obs. of  5 variables:
 $ Title           : chr  "Head First Java" "HTML and CSS: Design and Build Websites" "Python Crash Course"
 $ Authors         : chr  "Kathy Sierra, Bert Bates" "Jon Duckett" "Eric Matthes"
 $ Publication_Year: int  2022 2011 2023
 $ Publisher       : chr  "O'Reilly Media" "Wiley" "No Starch Press"
 $ Language        : chr  "Java" "HTML/CSS" "Python"

Standardize Column Names

In this section, The HTML and JSON files use slightly different column names. To ensure a valid comparison, the column names in the JSON data frame are updated to match those in the HTML data frame

names(json_df)

[1] "Title"            "Authors"          "Publication_Year" "Publisher"       
[5] "Language"        

names(html_df)

[1] "Title"            "Authors"          "Publication Year" "Publisher"       
[5] "Language"        

names(json_df) <- names(html_df)

#html_df$Publication.Year <- as.numeric(html_df$Publication.Year)

Compare the Data Frames

Once both data frames have the same structure, they are compared to determine whether they contain identical information. The comparison returned True, indicating that both the HTML and JSON files contained the same book information

identical(html_df, json_df)

[1] TRUE

all.equal(html_df, json_df)

[1] TRUE

comparison_result <- all.equal(html_df, json_df)
comparison_result

[1] TRUE

Conclusion

Both the HTML and JSON files were successfully created, imported into R, and converted into data frames. After standardizing the column names and object types, the two data frames were compared and found to contain identical information. This project provided valuable experience working with HTML and JSON data formats and demonstrated how R can be used to validate data consistency across multiple sources.