BASH

1: Write a script called simple-useradd.sh that adds a local user to the system. This
script should:
i) Take only one argument, or else exit after printing a usage message.

ii) Check /etc/passwd and decide on the first free user ID. Print a message
containing this ID.

iii) Create a private group for this user, checking the /etc/group file. Print a
message containing the group ID.

iv) Gather information from the operator user: a comment describing this user,
choice from a list of shells (CSH and BASH will do, test for acceptability, else
exit printing a message), expiration date for this account, extra groups of
which the new user should be a member.

v) With the obtained information, add a line to /etc/passwd, /etc/group and
/etc/shadow; create the user's home directory (with correct permissions!); add
the user to the desired secondary groups.

vi) Set the password for this user to a default known string.
Answer :
#!/bin/bash
# Script to add a user to Linux system
HOME_BASE=/home/
SHELL=/bin/bash

read -p Username: USER
read -s -p Password: PASSWORD
egrep -w ^$USER /etc/passwd >/dev/null
if [ $# -ne 1 ]
echo "Command arguement cannot be more than one or less!"
exit 1
else if [ $? -eq 0 ]; then
echo "$username exists!"
exit 1
else
PASS=$(perl -e print crypt($ARGV[0], password) $PASSWORD)
useradd -p ${PASS} -s $SHELL -m -d ${HOME_BASE}${USER} ${USER}
[ $? -eq 0 ] && echo -e \nUser has been added to system!\n || echo -e \nFailed to
add a user!\n
fi


2: Write a script called homebackup that automates tar so the person
executing the script always uses the desired options (cvp) and backup
destination directory (/var/backups) to make a backup of his or her home directory.
Implement the following features:

i) Test for the number of arguments. The script should run without arguments.
If any arguments are present, exit after printing a usage message.

ii) Determine whether the backups directory has enough free space to hold
the backup.

iii) Ask the user whether a full or an incremental backup is wanted. If the user
does not have a full backup file yet, print a message that a full backup will be
taken. In case of an incremental backup, only do this if the full backup is not
older than a week.

iv) Compress the backup using any compression tool. Inform the user that the
script is doing this, because it might take some time, during which the user
might start worrying if no output appears on the screen.

v) Print a message informing the user about the size of the compressed
backup.

#!/bin/sh
if [ $# -ne 0 ]
echo "No Command arguement can be present at this moment!"
exit 1
fi

# What to backup.
datapath='/home'

# Where to backup to.
backuppath="/var/backup"

# checking the size and space.
backupsize=$df -h "/home/"
availspace=$df -h "/dev/hda1"

if [ $availspace < $backupsize ]
echo "Not enough free space to hold the backup"
exit 1
fi

# checking the file exsit.
ckfile=$ls -l "$dest"

# Backup type

if [ $ckfile < 1 ]; then
echo "No previoud backup found, System is doing Full Backup from $backup_files to
$dest/$archive_file"
echo "Tar compression tool will be used, it might take some times."
else

echo "Select a backup type"
echo "Key in 1 for Full backup or 2 for Incremental backup"
read bktype
if [ $bktype -eq 1 ]; then
echo "System is now doing a Full Backup from $backup_files to $dest/$archive_file"
echo "Tar compression tool will be used, it might take some times."

# check last modified date
if [ "$(date +%a)" = "Mon" && $bktype -eq 2 ]; then
ckdate=$(stat -c %y $hostname-$day.tgz)"
ckdate=($ckdate - $day)
# Print start status message.
echo "System is now doing a Incremental Backup from $backup_files to
$dest/$archive_file"
echo "Tar compression tool will be used, it might take some times."
# move the old backup file to last week
fullbackup='-full' #extend file name
rm $backuppath/lastweek/* -R
mkdir lastweek
mv $backuppath/data-* $backuppath/lastweek/
fi
# Backup the files using tar.
filename=$(date +%u-%a-%Y%m%d)
tar -zcf $backuppath/$filename $fullbackup.tar.gz -g

# Print end status message.
echo
echo "Backup finished"
date

else if [ $bktype -gt 2 ]; then
echo "Invalid input"
exit 1

fi


PERL scripts should be written to run under Linux/OSX

1: Write a Perl script that will notify a user if they have received email. The
script should be run under cron at specific intervals and return the from
address, subject and date/time email was received in a readable format.

Answer:

#!/usr/bin/perl
use strict;
use Mysql;
use Mail::POP3Client;

sub DoAccount($);

my @Accounts;

$Accounts[0] = {'Name' => 'allenwong@curtin.my',
'Server' => 'mail.curtin.my',
'Username' => 'allenwong',
'Password' => '123456'};


MAIN:
{
foreach my $I (@Accounts)
{
print("Account: $I->{'Name'}\n");
DoAccount($I);
}
}

sub DoAccount($)
{
my ($AN) = @_;
my $CreateString;
my $Message;
my $Sender;
my $Recipient;
my $Temp;
my $TMPFile;

$CreateString = "\tUSER => '$AN->{'Username'}',
PASSWORD => '$AN->{'Password'}',
HOST => '$AN->{'Server'}'";

$CreateString .= ",\n\tUSESSL => 'true'" if ($AN->{'SSL'});
$CreateString .= ",\n\tPORT => '$AN->{'PORT'}'" if ($AN->{'PORT'});

my $pop = new Mail::POP3Client(eval($CreateString));

print("Logging in....");
if (!($pop->Login()))
{
print("Could not connect to account: $AN->{'Name'}: $!\n");
return 1;
}
print("Logged in!\t");

print("Total Messages: " . $pop->Count() . "\n");

foreach my $I (1 .. $pop->Count())
{
$Message = $pop->HeadAndBody($I);

foreach my $Line (split(/\n/, $Message))
{
if (($Line =~ m/^From:.*$/ig) && (!($Sender)))
{
$Temp = $&;
print("Temp: $Temp\n");
if ("$Temp" =~ m/[A-Za-z0-9-]+\@[A-Za-z0-9-]+\.[A-Za-z0-9-]+/ig)
{
$Sender = $&;
}
}
last if ($Sender);
}

if (!($Sender))
{
print("Error finding sender!\n");
next;
}

do
{
$TMPFile = "/tmp/MAILSYSTEM-" . rand(1000000000) . ".eml";
} while (-e $TMPFile);

open(TMP, ">$TMPFile");
print TMP $Message;
close(TMP);

if (system("/server/Mailque/bin/newfilter.pl \"$Sender\" \"exodist\@open-
exodus.net\" < $TMPFile"))
{
print("Error sending message!!!\n");
}
else
{
$pop->Delete($I);
}
unlink($TMPFile);
}
}




























2: Write a Perl script that enters a directory and
a. Searches all the files in a directory for the occurrence of a string
passed to the script from the command line.
b. It should return the name of the files that contain the string.

Answer:
#!/usr/bin/perl

#The filenames from a directory are found in the array @dots.
use strict;
use warnings;

my $dir = '/etc';

opendir(DIR, $dir) or die $!;

my @dots
= grep {
/^\./ # Begins with a period
&& -f "$dir/$_" # and is a file
} readdir(DIR);

# Loop through the array printing out the filenames
foreach my $file (@dots) {
print "$file\n";
}
closedir(DIR);

#Print out the name of the file of all the occurrence of a string in the directory

my $line=("omg") # string that we are searching for
foreach my $search (@dots)
{
open(FILE, @dots); # we open each file from command line
while ($line = <FILE>) # we iterate through each line in the file.
{
print "$line found on $search\n"; # we print out the string on the file we are looking at
}
close(FILE); # we close the file.
}

print "Complete.\n";

exit 0;























3: All web pages should meet the W3C standards on accessibility, which includes
providing alt tags for all images (a text description of an image). Write a Perl script
that parses the source of a web page and displays the text of all alt tags. If an image
exists and there is no alt tag text assigned to that image, the user should be notified.
Hint: use the curl utility to garb the html.


#!/usr/bin/perl
#global vars
my $path_sep = "";


#scan_dir
sub scan_dir {
my ($dir) = @_;
my (@files);
chomp $dir;

# list all files inside $dir
opendir( ROOT, $dir ) || die "cant open $dir\n";
@files = readdir( ROOT );
closedir( ROOT );

FILE: foreach (@files) {
# skip the current and the upper directory in the file listing
next FILE if /^\.|\.\.$/;

# build the complete file path
$file_path = $dir.$path_sep.$_;
if (-d $file_path) {
# if the current file is a directory, recursively scan this directory too
&scan_dir( $file_path, $last_scan_time );
next;
}
# update only html files
if ($file_path =~ /\.html$/i) {
add_alt_tags( $file_path );
}
}
}

#add_alt_tags

sub add_alt_tags {
my ($file) = @_;

# print out the current file name
print STDOUT "$file\n";

# read the file
open( INFILE, "<$file" );
@all = <INFILE>;
close( INFILE );

# reopen it for output
open( OUTFILE, ">$file" );

for $line (@all) {
# match all lines with img tags, but without alt attributes
if (($line=~m/(<\s*img\s+src=)(.*?)(>)/) && ($2!~m/alt=/)) {
# split the matched img tag into two parts, and substitute
# the entire tag with an inserted alt attribute...
$line =~ s/(<\s*img\s+src=.*?)(>)/$1 alt=""$2/;
}
print OUTFILE $line;
}
close OUTFILE;
}

#main
sub main {
umask (000);
if( $ENV{"OS"} eq "Windows_NT" ) {
$path_sep = "\\";
}
else {
$path_sep = "/";
}
if ($#ARGV==-1) {
print STDOUT "Usage: $0 <absolute path to scan for .html files>\n"; exit
1;
}
scan_dir( $ARGV[0] );
}
&main;
exit;