Mini CSV (comma separated values) Spreadsheet interpreter

Reads in a csv spreadsheet (format described below), evaluates the value of each cell, and outputs values to STDOUT.

A spreadsheet is defined as a two-dimensional array of cells. Columns are identified using letters and rows by numbers (C2 references a cell in column 3, row 2). Each cell contains either an integer or an expression. Expressions contain integers, cell references, and operators ('+', '-', '*', '/') and are evaluated with the usual rules of evaluation.

Input Format:

• A csv file with m rows and n columns
• The input file will have no headers
• Cells will not be surrounded in double quotes

Output Format:

• A csv file (to stdout) with the same dimensions as the input file
• Each cell is output as a floating point value. Round output values to two decimal places.

Example:

input.csv

B2+2,A1+A2
B2-3,7+5

output.csv

14.00,23.00
9.00,12.00

More specifications

• The spreadsheet should be able to evaluate expressions containing cell references, integers, floating point numbers, and the addition and subtraction operators.
• Support for up to 26 columns (A-Z)
• The spreadsheet should detect circular references and exit appropriately
• Main method in a class named Spreadsheet.java
• Empty cell contents will be treated as null values and be output as empty strings.
• Negative numbers supported.

Design

• The overall design is based on the Gang-Of-Four Interpreter design pattern.
• Factory methods are used to create and manage Expressions and are in the context class.
• The implementation emphasizes simplicity and hence does not implement a full lexer and parser.
• They are jointly implemented via regexes which are sufficient for the 4 binary operators desired.
• The primary SpreadSheetContext class is the "Context" class in the Interpreter pattern.
• The Expression class hierarchy is designed per the Composite design pattern to support easy extensibility

Solution

• Expressions are the fundamental units stored in cells in the spreadsheet.

• Some Expressions are simple terminal ones that contain ints/floats/references to other cells/Expressions.

• Other Expressions are complex and are currently limited to Binary operations.

• SpreadsheetContext objects will hold all the actual data slurped from the csv.

• Spreadsheet cells are stored in a HashMap. The unevaluated text representation is stored as is.

• The cells (Expressions) are evaluated on demand when required for activities such as printing.

• This postponed/lazy evaluation will allow for the construction a more general spreadsheet editor over time.