Description
1 Summary
Write a program which reformats a text file to fit in a certain number of columns (the page width).
This process is called word wrapping.
Part I (50 points) Write a word-wrapping program ww that reads the contents of a file and prints
it to standard output, wrapped to fit within a specified width.
For example, to reformat “manifesto.txt” to fit in 80 columns:
$ ./ww 80 manifesto.txt
Part II (50 points) Extend ww by allowing the second argument to be a directory. When the
second argument is a directory, ww will reformat each file in the directory and write the output to a
new file with the prefix “wrap.”.
For example, if directory “foo” contains “bar” and “baz”, this command will create files in foo
named “wrap.bar” and “wrap.baz”:
$ ./ww 80 foo
2 Definitions
line A sequence of characters ending with a newline (’\n’). A line may not contain a newline,
except as the last character.
width The number of characters in a line preceding the terminal newline. A file wrapped to width
80 will contain at most 80 characters between one newline and the next. We can consider two
consecutive newlines to be separated by a zero-width line.
1
blank line A line containing only whitespace characters.
word A non-empty sequence of non-whitespace characters. Words are not restricted to alphabetic
characters.
A blank line contains no words. All non-blank lines contain at least one word.
paragraph A sequence of non-blank lines. A paragraph contains one or more words.
Implications Words are separated by one or more whitespace characters. The amount of
whitespace used is not important. Paragraphs are separated by a whitespace sequence containing at
least two newlines. Whitespace in blank lines, and the number of blank lines, is not important.
3 Input
Your program will take one or two arguments: the desired page width and an optional file or
directory name. The page width should be a positive integer.
If the file name is not present, ww will read from standard input and print to standard output.
If the file name is a regular file, ww will read from the file and print to standard output.
If the file name is a directory, ww will open each regular file in the directory and write to a new
file created in the same directory. The name for each output file will be the name of the input file
prefixed with “wrap.”, so that a file “bar” will become “wrap.bar”. If a file by that name already
exists, overwrite it.
Exceptions When reading a directory, ww will ignore any files whose names begin with a period
(.) or the string “wrap.”.
4 Output
While ww should accept any text file, the text it produces should be normalized. In a normalized
text file, lines do not begin or end with whitespace (excluding the terminal newline) and no more
than one space occurs between words. As a consequence, all whitespace sequences in a normalized
file will be either (a) a single space, (b) a single newline, or (c) two consecutive newlines. Since no
line begins with whitespace, the first character in a non-empty file will be non-whitespace. Since
each line is terminated by a newline, the final character in a non-empty file will always be a newline.
(We consider a file with no words to contain no lines. Thus, the smallest non-empty file will contain
at least two characters.)
Implications As a result of normalization, two input files consisting of the same words in the
same paragraphs will produce identical output when wrapped to the same width. In particular, if
we wrap a file to some width, and then re-wrap the wrapped file to the same width, the wrapped
and re-wrapped files should be byte-identical.
$ ./ww 80 input_file > output1
$ ./ww 80 output1 > output2
$ cmp output1 output2
$
2
In this example, cmp prints nothing because the two files have no differences.
5 Method
We will use a greedy word-wrap algorithm: track the number of characters printed on the current
line. Before printing a word, check whether it will fit on the current line. If so, print it on the
current line. If not, start a new line.
Use open, read. and write to access files. Make no assumptions about the maximum lengths for
a file, line, or word. (You may assume that the lengths will not exceed INT_MAX.) Do not attempt to
read the entire file in a single call to read.
Use stat and the macros S_ISDIR and S_ISREG to determine whether the second argument is a
file name or directory. When listing the contents of a directory, you may use stat or the d_type
field to distinguish files and subdirectories.
Free all memory allocated by your program and close all files opened by your program.
6 Error conditions
If the file does not exist or cannot be opened use perror to report a message to the user and
terminate with exit status EXIT_FAILURE.
If the input contains a word that is longer than the page width, print it on a single line by itself
that will exceed the page width. Do not truncate or divide the word. If your program needs to exceed
the page width, continue printing the input file and then finish with exit status EXIT_FAILURE.
7 Notes
If you are using a char array as a buffer, use a macro to define its length, and then test your code
by changing the definition. Your code should still work even if the buffer length is 1.
In general, word boundaries will not coincide with the ends of the buffer. Make sure you can
handle receiving a word split between the data returned by consecutive calls to read(). In particular,
make sure you can handle a word that is too big to fit in your buffer.
You may use any of the C or Posix standard libraries available on the iLab machines. You may
use code originally written by you or your partner for other assignments in this class. You may not
include or copy code from other students or found on-line.
8 Examples
Here is some text that we might have in a file:
This is very good text that has been put into a file for the purposes
of being
an
example.
That is
good.
3
Wrapped to width 20:
This is very good
text that has been
put into a file for
the purposes of
being an example.
That is good.
Wrapped to width 30:
This is very good text that
has been put into a file for
the purposes of being an
example.
That is good.
9 Submission
Submit a Tar archive containing your source code, makefile, and a README. The README must
contain:
• Your name and NetID
• Your partner’s name and NetID, if you worked with a partner.
• A brief description of your testing strategy. How did you determine that your program is
correct? What sorts of files and scenarios did you check?
Do not include executables, object files, or testing data. Before submitting, be sure to confirm
that your archive contains the correct files!
10 Grading
Your program will be scored out of 100 points: 50 points for each part. Your grade will be based on
test cases and on manual examination. Testing will be performed on iLab machines, so be certain
that your code will compile and execute on the iLab!
Your code should be free of memory errors and undefined behavior. We may compile your code
using AddressSanitizer, UBSan, Valgrind, or other analysis tools. You will lose points if these tools
report memory errors, space leaks, or undefined behavior. You are advised to take advantage of
AddressSanitizer and UBSan when compiling your program for testing.
4
