Lab 4

Due: 03/18

Exception handling and I/O

Objectives

• 4 Points - Extend some of the classes developed in previous assignments to handle exceptional conditions.
• 16 Points - Implement a class using file I/O.

Description

Extend some of the classes developed in previous assignments to handle exceptional conditions.

In this task we will revisit some of the classes developed in previous assignments. Your task is to identify exceptional conditions and implement code that deals with these conditions or throws exceptions for conditions that cannot be dealt with in the class itself. Copy the following classes to the package lecture.lab4.

• lecture.lab1.BankAccount
• lecture.lab3.BlockWorld

Add a text file explanations.txt into folder lecture/lab4 in which you explain the exceptional conditions you have identified and how you address them

Implement a class using file I/O.

In this class you will implement classes DataFrame and CSVFileAccess in package lecture.lab4. Class DataFrame stores tabular data. CSVFileAccess reads such tabular data from a file and can write the data of a table stored in a DataFrame to disk using the CSV file format (explained below).

DataFrame (6 Points)

A table has a list of columns (its schema) and a a sequence of rows (the data). A row has a value of type String for each column of the table. The schema of a DataFrame should be determined when creating the data frame. That is, you should implement a constructor which takes as input an array of column names:

public DataFrame(String[] columnNames);

Initially, the table stored in a dataframe should be empty. The following methods can be used to access and add data.

• getNumRows returns the number of rows in the table

public int getNumRows(); // returns the number of rows in the table

• getRow - returns the row at position “pos” or throws an exception if no such row exists.

public String[] getRow(int pos) throws Exception;

• getCell - returns the value at cell (a column of a particular row) or throws an exception if no such cell exists.

public String getCell(int row, int column) throws Exception;

• setCell - replaces the value at cell (a column of a particular row) with value or throws an exception if no such cell exists.

public void setCell(int row, int column, String value) throws Exception;

• appendRow add the provided row to the end of the table increasing the number of rows by 1. Throws an exception if row does not have the right number of values.

public void appendRow(String[] row) throws Exception;

• getColumnNames - returns all column names as a String[].
• getColumnName - returns the name of column at position pos or throws and exception if no such column exists.
• getColumnPos - returns the position of the column name named or throws and exception if no such column exists.

public String[] getColumnNames();
public String getColumnName(int pos) throws Exception;
public int getColumnPos(String name) throws Exception;

CSVFileAccess (10 Points)

The CSVFileAccess class should implement two methods:

public DataFrame readCSV(File csvFile) throws Exception;
public void writeToCSV(DataFrame f, File csvFile) throws Exception;

Method readCSV reads a table from a CSV file and returns a DataFrame storing this table.

This method needs to deal with both I/O errors (e.g., the file does not exist) and data format errors, e.g., a row does not have the correct number of values.

Method writeToCSV takes a data frame and writes the table is represents to a CSV file.

This method needs to handle I/O errors. Also if the file already exists an exception should be thrown instead of modifying the file.

Test Case

Write a class TestDataFrameCSV whose main method creates a data frame from the file animals.csv that we have uploaded to your git repos. Add new animals lion and tiger as new rows to the dataframe and then write this updated data frame to a file called new_animals.csv. You should determine reasonable values for the other columns (length, weight, …).

The CSV File Format

The CSV (Comma Separated Values) file format is a textual format for storing tabular data. The first line of a CSV file may be a header which lists the names of columns of the table. All the following lines are the data where each line corresponds to a row in the table. Both the column names and column values of a row are separated using comma (","), hence the name. For instance, shown below is the content of a CSV file which stores a table with 3 columns (Name, Age, Salary) and three rows. For example, the first row is denotes a person with name Peter who is 45 years old and earns 30000.

Name,Age,Salary
Peter,45,30000
Alice,35,100000
Bob,23,20000