CSE 374 Midterm Exam Sample Solution 2/6/12

Question 1. (12 points) Suppose we have the following subdirectory structure inside the
current directory:

docs
docs/friends
docs/friends/birthdays.txt
docs/friends/messages.txt
docs/cse374
docs/cse374/notes.txt
docs/cse374/hw9
docs/cse374/hw9/package.h
docs/cse374/hw9/package.c
docs/cse374/hw9/main.c
pictures
pictures/big.jpg
pictures/thumb.jpg

(a) What are the final contents of the file answer after the user executes the following
commands in the current directory:

cd docs
ls friends > temp
ls cse374 >> temp
sort temp >> answer

birthdays.txt
hw9
messages.txt
notes.txt

(b) Assuming we are back in the original directory, and without changing the current
directory, give a single command line that will list the number of lines and characters in
each of the C source files (files with names ending in ‘.c’) in the hw9 directory. The
output can contain additional detail or summary information if it is inconvenient to get rid
of it.

wc docs/cse374/hw9/*.c

(Strictly speaking, we should use the -l and –c options to count just lines and
characters, but that is not something you would be expected to memorize.)

(c) Assuming we are back in the original directory, and without changing the current
directory, give a single command line that will make a complete copy of the
docs/cse374 subtree (including all directories and files), and store the copy in a new
directory docs/cse374-bak.

cp -r docs/cse374 docs/cse374-bak

Page 1 of 8
 CSE 374 Midterm Exam Sample Solution 2/6/12

Question 2. (18 points) Write a shell script rmdups that deletes all files that are an
exact byte-for-byte copy of a given file. The command

rmdups f1 f2 ... fn

should remove all of the files f2 through fn that are exact copies of f1. Use cmp to
detect whether two files are identical. The command

cmp -s x y

will exit with a return code of 0 if the two files x and y are identical. The return code
will be 1 if they are different. The -s option instructs cmp to perform the comparison
silently and not produce any output on stdout or stderr.

Your script should verify that there is at least one argument (f1) and that each of the files
in the argument list exist and are regular files (if [[ -f file ]] ). If there are no
arguments or if f1 is not an existing regular file, the script should print an appropriate
message to stderr and exit. If any of the other arguments are not existing regular files,
the script should print an appropriate error message to stderr and ignore that argument,
continuing on to process the remaining files.

Assume that whoever is running the script has sufficient permissions to remove any
necessary files. Also assume that all files are readable if they exist. You do not need to
worry about permissions issues.

Additional space is provided on the next page if you need more room for your answer.

Page 2 of 8
 CSE 374 Midterm Exam Sample Solution 2/6/12

Question 2. (cont.) Additional space for your answer if needed.

#!/bin/bash

if [[ $# < 1 ]]
then
echo $0: at least one argument required >&2
exit 1
fi

if [[ ! -f $1 ]]
then
echo $0: $1 not ordinary file >&2
exit 1
fi

fn=$1
shift
while [[ $# > 0 ]]
do
if [[ -f $1 ]]
then
cmp -s $fn $1
if [[ $? = 0 ]]
then
rm $1
fi
else
echo $0: $1 not ordinary file >&2
fi
shift
done

The exact wording of error messages can be anything appropriate, but they did need
to be written to stderr (stream 2).

Page 3 of 8
 CSE 374 Midterm Exam Sample Solution 2/6/12

Question 3. (14 points) Give basic regular expressions that could be used with grep
(not egrep) to locate words with the following properties in a file containing a list of
words, each line in the file containing a single word (like the words file demonstrated in
class).

(a) All words that would be legal identifiers in Java or C programs: that is, they start
with an upper-case or lower-case letter and contain zero or more additional letters, digits,
and underscore (‘_’) characters. Examples: ‘x’, ‘xyzzy’, ‘a_long_name’, ‘Another_1’,
‘sillyQuestion’, but not ‘1more_thing’ (starts with a digit), etc.

^[a-zA-Z][a-zA-Z0-9_]*$

(b) All words that are 8 characters long and consist of the same 4 characters repeated
twice. Examples: ‘couscous’, ‘beriberi’.

^\(....\)\1$

(c) All words that do not contain any lower-case English vowels. For this question, the
vowels are the characters ‘a’, ‘e’, ‘i’, ‘o’, and ‘u’.

^[^aeiou]*$

An answer that used + instead of * would also be fine.

On all of these problems, characters ^ and $ were needed surrounding each basic
pattern to match only complete words. If those were missed on multiple parts of the
question, points were only deducted once.

Page 4 of 8
 CSE 374 Midterm Exam Sample Solution 2/6/12

Question 4. (18 points) Several months before each quarter starts, faculty members are
asked to send in a list of books for their upcoming courses. The bookstore only needs the
10-character ISBN codes (the codes in the barcode label on the back of each book), but
usually faculty members send messages with extra information. For example:

This year please get the Kernighan & Ritchie C book, isbn is 0-13-110362-8, 2nd ed
Also include the Linux Pocket Guide 0-596-00628-0 by Barrett.
For my other course please order Pragmatic Programmer 0-201-61622-X

For this problem, write a shell script whose argument is the name of a file containing
lines like the ones above. The output should be a list of ISBN codes extracted from the
text. For example, given the above input, the output should be:

0131103628
0596006280
020161622X

ISBN format: All ISBN codes in the input file contain 10 characters (mostly digits) plus 3
hyphens. There are 4 parts to the code separated by the 3 hyphens:

region-publisher-title-check

The region is always 0, and the check character is a single digit 0-9 or the upper-case
letter X. The publisher and title codes are all digits 0-9 and there are 8 digits total (you
do not need to check for the exact number of digits in these parts). The number of digits
in these two parts are different for different publishers. Large publishers have short
publisher codes with fewer digits and more digits in the title code; small publishers have
more digits in the publisher code and fewer digits for the title code. Both the publisher
and title codes have at least one digit each.

You may assume that each line in the input file contains exactly one ISBN code some-
where in the line, and that nothing else in the line is formatted exactly like an ISBN code.

You should assume that the script has one argument and it is the name of a file that exists
and can be read (i.e., you don’t need to check for a valid argument list in this problem).

Hint: sed

#!/bin/bash
sed 's/.*0-\([0-9][0-9]*\)-\([0-9][0-9]*\)-\([0-9X]\).*/0\1\2\3/' $1

Note: Basic sed regular expressions do not include the + operator, but we allowed
[0-9]+ to match the book and publisher fields. Just [0-9]* would not be
sufficient because it would match strings without digits in those positions.

Page 5 of 8
 CSE 374 Midterm Exam Sample Solution 2/6/12

Question 5. (18 points) Consider the following C program.

#include <stdio.h>

void mystery(int *x, int *y, int *z) {

*x = *y + *z;
}

int main(int argc, char** argv) {

int x, y;
int *p1;
int *p2;
int *p3;

p1 = &x;
*p1 = 1;
p3 = p1;

p2 = &y;
*p2 = 3;

printf("%d %d %d \n", *p1, *p2, *p3);

*p3 = (*p2) + 1;

printf("%d %d %d \n", *p1, *p2, *p3);

mystery(p1,p2,p3);

printf("%d %d %d \n", *p1, *p2, *p3);

return 0;
}

What output does this program produce when it is executed? (It does execute
successfully.) It would be useful to draw diagrams showing variables and pointers to
help answer the question and to help us award partial credit if needed.

1 3 1
4 3 4
7 3 7

Page 6 of 8
 CSE 374 Midterm Exam Sample Solution 2/6/12

Question 6. (20 points) (The small C programming exercise.) Write a on the next page
a program that will read a file whose name is given as an argument on the command line
and will print the following on stdout:

 The name of the file,

 The total number of lines in the file, and
 The contents of the last line in the file.

The output should be formatted as shown in the following examples.

Examples: If we run the program with the file hamlet.txt as input (which contains
4921 lines of text), the output should be:

hamlet.txt line 4921

FINIS. The tragedie of HAMLET, Prince of Denmarke.

If we run the program with a different file fairy-tale.txt, the output might be:

fairy-tale.txt line 263

And they all lived happily ever after.

Details:

 If the program is given no arguments or more than one argument, print a suitable
message to stderr and exit.
 If there is one argument, you can assume it names an existing file that can be
successfully opened for reading. You don’t need to check for this.
 You can assume that the input file contains lines of text and no input line contains
more than 200 characters, including any trailing newline (\n) and terminator (\0)
characters.
 Your answer may be written as a single main function, but you can define
additional functions if you wish.
 Assume that any #include directives for any necessary standard C library files
are already provided – you do not need to write these.

Some useful functions: fopen(filename, “r”), fgets(line, max_length, file)

You may not need nearly as much space as is available for your answer.

Write your answer on the next page

(You may detach this page from the exam if that is convenient.)

Page 7 of 8
 CSE 374 Midterm Exam Sample Solution 2/6/12

Question 6. (cont). Write your answer here.

#define MAX_LINE_SIZE 200

int main(int argc, char ** argv) {

int nlines; /* number of lines read */

char line[MAX_LINE_SIZE]; /* contents of line nlines */
FILE *in; /* input file */

/* exit if not exactly 1 filename argument */

if (argc != 2) {
fprintf(stderr, "exactly one argument required\n");
return 1;
}

/* open file, read and count lines */

in = fopen(argv[1], "r");
nlines = 0;
while (fgets(line, MAX_LINE_SIZE, in)) {
nlines++;
}

/* print number of lines and last line in file */

printf("%s line %d\n%s", argv[1], nlines, line);

return 0;
}

Note: Solutions on an exam would not be expected to contain extensive comments or

#define constants as shown above.

Page 8 of 8