For this assignment you are to write a program that reads a text file, identifies the words in
it and then checks the spelling of each word against a small dictionary of words. Any words
not found the dictionary are printed out. Finally, the total number of words in the file is
printed out and the total number with unknown spelling.
DEFINITIONS
Text Files
For our purposes, a text file consists of lines of text containing words, spaces and
punctuation with a newline character at the end of each line. All characters will be from the
basic ASCII character set.
Words
• Words are sequences of alphabetic and numeric characters, apostrophe, the underscore
and hyphen characters that are separated by sequences of one or more separation
characters. See below for a list of the separator characters.
• The input for this assignment will consist of normal words plus integer and floating point
numbers.
Separator Characters
• The list of word separator characters includes the following :
space, tab (\t), newline (\n), carriage return (\r)
and the common punctuation characters
,;:”~!#%^*()=+[]{}\|<>?/
• If the . character has a space, tab, newline, digit, minus or plus on the left and a digit on
the right then it is treated as a decimal point and thus part of a number. Otherwise it is
treated as a full stop and a word separator.
Assignment 2 – Autumn 2015 Page 2 of 3
IMPLEMENTATION DETAILS
1. Assuming your compiled program is called a.out your program will use command line
argument as follows
./a.out dictionaryfile textfile
Where dictionaryfile is a file containing a list of correctly spelled words and
textfile is the file containing the text you are going to spell check.
2. In the SUBNET assign directory you will find a file called dict.txt. This file contains the
1000 most commonly used words in the English language, all in lower case and with one
word per line. You will use this as your dictionary file. Do NOT change the contents of it.
3. Your program should start by reading the dictionary file, loading each word into a C++
string and then storing that string in a binary tree. A copy of the bintree.h and
binnode.h files can be found in the SUBNET assign directory.
PLEASE NOTE 1. You must use these binary tree classes and you may not alter them in
any way. Do NOT use any other data structures to store the word list.
PLEASE NOTE 2. To use them you will just #include bintree.h in your code. Do NOT
cut and paste the entire bintree code into your code.
4. Your program will then read the text file, line by line, and extract every word from the
line. Each extracted word will be checked to see if it exists in the tree. If not then the
word will be printed. Please note the following regarding extracted words
• Before checking a word against the tree, all letters in it will be changed to lower
case. That is “Hello” will be changed to “hello” before seeing if it exists in the
tree.
• If any extracted word is a number, either an integer or real, it will be ignored. That
is, it won’t be checked for spelling. Look up the strtod library function to work out
how to do this.
5. Finally, your program will print the total number of words in the file and then print the
total number of words with unknown spelling. Numbers are not considered to be words
and thus not included in the count of words.
BENCHMARK PROGRAM
To clarify how the program works, a benchmark executable version has been placed on the
server. You can run it by typing the following command
/home/glingard/spellcheck
You will need to give it a dictionaryfile and textfile to check for spelling. To give you an idea
of the size of this assignment, the source code for the benchmark program, including
whitespace and comments, but not including the bintree code, takes approximately 280
lines of code.
ASSIGNMENT OBJECTIVES
The purpose of this assignment is to demonstrate competence in the following skills.
• Program design – including class design
• Using command line parameters
• String manipulation
• File handling
