mikkey at dynamo.com.ar
http://www.linuxdoc.org/HOWTO/Bash-Prog-Intro-HOWTO.html
Familiarity with GNU/Linux command lines, and familiarity with basic programming concepts is helpful. While this is not a programming introduction, it explains (or at least tries) many basic concepts.
This document tries to be useful in the following situations
This HOW-TO will try to give you some hints about shell script programming strongly based on examples.
In this section you'll find some little scripts which will hopefully help you to understand some techniques.
#!/bin/bash
echo Hello World
This script has only two lines. The first indicates the system which program to use to run the file.
The second line is the only action performed by this script, which prints 'Hello World' on the terminal.
If you get something like ./hello.sh: Command not found. Probably the first line '#!/bin/bash' is wrong, issue whereis bash or see 'finding bash' to see how sould you write this line.
#!/bin/bash
tar -cZf /var/my-backup.tgz /home/me/
In this script, instead of printing a message on the terminal, we create a tar-ball of a user's home directory. This is NOT intended to be used, a more useful backup script is presented later in this document.
There are 3 file descriptors, stdin, stdout and stderr (std=standard).
Basically you can:
A little note for seeing this things: with the less command you can view both stdout (which will remain on the buffer) and the stderr that will be printed on the screen, but erased as you try to 'browse' the buffer.
This will cause the ouput of a program to be written to a file.
ls -l > ls-l.txt
Here, a file called 'ls-l.txt' will be created and it will contain what you would see on the
screen if you type the command 'ls -l' and execute it.
This will cause the stderr ouput of a program to be written to a file.
grep da * 2> grep-errors.txt
Here, a file called 'grep-errors.txt' will be created and it will contain what you would see
the stderr portion of the output of the 'grep da *' command.
This will cause the stderr ouput of a program to be written to the same filedescriptor than stdout.
grep da * 1>&2
Here, the stdout portion of the command is sent to stderr, you may notice that in differen ways.
This will cause the stderr ouput of a program to be written to the same filedescriptor than stdout.
grep * 2>&1
Here, the stderr portion of the command is sent to stdout, if you pipe to less, you'll see that
lines that normally 'dissapear' (as they are written to stderr) are being kept now (because
they're on stdout).
This will place every output of a program to a file. This is suitable sometimes for cron entries, if you want a command to pass in absolute silence.
rm -f $(find / -name core) &> /dev/null
This (thinking on the cron entry) will delete every file called 'core' in any directory. Notice
that you should be pretty sure of what a command is doing if you are going to wipe it's output.
This section explains in a very simple and practical way how to use pipes, nd why you may want it.
Pipes let you use (very simple, I insist) the output of a program as the input of another one
This is very simple way to use pipes.
ls -l | sed -e "s/[aeio]/u/g"
Here, the following happens: first the command ls -l is executed, and it's output,
instead of being printed, is sent (piped) to the sed program, which in turn, prints
what it has to.
Probably, this is a more difficult way to do ls -l *.txt, but it is here for illustrating pipes, not for solving such listing dilema.
ls -l | grep "\.txt$"
Here, the output of the program ls -l is sent to the grep program, which, in turn, will print
lines which match the regex "\.txt$".
You can use variables as in any programming languages. There are no data types. A variable in bash can contain a number, a character, a string of characters.
You have no need to declare a variable, just assigning a value to its reference will create it.
#!/bin/bash
STR="Hello World!"
echo $STR
Line 2 creates a variable called STR and assigns the string "Hello World!" to it. Then the VALUE of this variable is retrieved by putting the '$' in at the beginning. Please notice (try it!) that if you don't use the '$' sign, the output of the program will be different, and probably not what you want it to be.
#!/bin/bash
OF=/var/my-backup-$(date +%Y%m%d).tgz
tar -cZf $OF /home/me/
This script introduces another thing. First of all, you should be familiarized with the variable creation and assignation on line 2. Notice the expression '$(date +%Y%m%d)'. If you run the script you'll notice that it runs the command inside the parenthesis, capturing its output.
Notice that in this script, the output filename will be different every day, due to the format switch to the date command(+%Y%m%d). You can change this by specifying a different format.
Some more examples:
echo ls
echo $(ls)
Local variables can be created by using the keyword local.
#!/bin/bash
HELLO=Hello
function hello {
local HELLO=World
echo $HELLO
}
echo $HELLO
hello
echo $HELLO
This example should be enought to show how to use a local variable.
Conditionals let you decide whether to perform an action or not, this decision is taken by evaluating an expression.
Conditionals have many forms. The most basic form is: if expression then statement where 'statement' is only executed if 'expression' evaluates to true. '2<1' is an expresion that evaluates to false, while '2>1' evaluates to true.xs
Conditionals have other forms such as: if expression then statement1 else statement2. Here 'statement1' is executed if 'expression' is true,otherwise 'statement2' is executed.
Yet another form of conditionals is: if expression1 then statement1 else if expression2 then statement2 else statement3. In this form there's added only the "ELSE IF 'expression2' THEN 'statement2'" which makes statement2 being executed if expression2 evaluates to true. The rest is as you may imagine (see previous forms).
A word about syntax:
The base for the 'if' constructions in bash is this:
if [expression];
then
code if 'expression' is true.
fi
#!/bin/bash
if [ "foo" = "foo" ]; then
echo expression evaluated as true
fi
The code to be executed if the expression within braces is true can be found after the 'then' word and before 'fi' which indicates the end of the conditionally executed code.
#!/bin/bash
if [ "foo" = "foo" ]; then
echo expression evaluated as true
else
echo expression evaluated as false
fi
#!/bin/bash
T1="foo"
T2="bar"
if [ "$T1" = "$T2" ]; then
echo expression evaluated as true
else
echo expression evaluated as false
fi
In this section you'll find for, while and until loops.
The for loop is a little bit different from other programming languages. Basically, it let's you iterate over a series of 'words' within a string.
The while executes a piece of code if the control expression is true, and only stops when it is false (or a explicit break is found within the executed code.
The until loop is almost equal to the while loop, except that the code is executed while the control expression evaluates to false.
If you suspect that while and until are very similar you are right.
#!/bin/bash
for i in $( ls ); do
echo item: $i
done
On the second line, we declare i to be the variable that will take the different values contained in $( ls ).
The third line could be longer if needed, or there could be more lines before the done (4).
'done' (4) indicates that the code that used the value of $i has finished and $i can take a new value.
This script has very little sense, but a more useful way to use the for loop would be to use it to match only certain files on the previous example
fiesh suggested adding this form of looping. It's a for loop more similar to C/perl... for.
#!/bin/bash
for i in `seq 1 10`;
do
echo $i
done
#!/bin/bash
COUNTER=0
while [ $COUNTER -lt 10 ]; do
echo The counter is $COUNTER
let COUNTER=COUNTER+1
done
This script 'emulates' the well known (C, Pascal, perl, etc) 'for' structure
#!/bin/bash
COUNTER=20
until [ $COUNTER -lt 10 ]; do
echo COUNTER $COUNTER
let COUNTER-=1
done
As in almost any programming language, you can use functions to group pieces of code in a more logical way or practice the divine art of recursion.
Declaring a function is just a matter of writing function my_func { my_code }.
Calling a function is just like calling another program, you just write its name.
#!/bin/bash
function quit {
exit
}
function hello {
echo Hello!
}
hello
quit
echo foo
Lines 2-4 contain the 'quit' function. Lines 5-7 contain the 'hello' function If you are not absolutely sure about what this script does, please try it!.
Notice that a functions don't need to be declared in any specific order.
When running the script you'll notice that first: the function 'hello' is called, second the 'quit' function, and the program never reaches line 10.
#!/bin/bash
function quit {
exit
}
function e {
echo $1
}
e Hello
e World
quit
echo foo
This script is almost identically to the previous one. The main difference is the funcion 'e'. This function, prints the first argument it receives. Arguments, within funtions, are treated in the same manner as arguments given to the script.
#!/bin/bash
OPTIONS="Hello Quit"
select opt in $OPTIONS; do
if [ "$opt" = "Quit" ]; then
echo done
exit
elif [ "$opt" = "Hello" ]; then
echo Hello World
else
clear
echo bad option
fi
done
If you run this script you'll see that it is a programmer's dream for text based menus. You'll probably notice that it's very similar to the 'for' construction, only rather than looping for each 'word' in $OPTIONS, it prompts the user.
#!/bin/bash
if [ -z "$1" ]; then
echo usage: $0 directory
exit
fi
SRCD=$1
TGTD="/var/backups/"
OF=home-$(date +%Y%m%d).tgz
tar -cZf $TGTD$OF $SRCD
What this script does should be clear to you. The expression in the first conditional tests if the program has received an argument ($1) and quits if it didn't, showing the user a little usage message. The rest of the script should be clear at this point.
In many ocations you may want to prompt the user for some input, and there are several ways to achive this. This is one of those ways:
#!/bin/bash
echo Please, enter your name
read NAME
echo "Hi $NAME!"
As a variant, you can get multiple values with read, this example may clarify this.
#!/bin/bash
echo Please, enter your firstname and lastname
read FN LN
echo "Hi! $LN, $FN !"
On the command line (or a shell) try this:
echo 1 + 1
If you expected to see '2' you'll be disappointed. What if you want BASH to evaluate some numbers you have? The solution is this:
echo $((1+1))
This will produce a more 'logical' output. This is to evaluate an arithmetic expression. You can achieve this also like this:
echo $[1+1]
If you need to use fractions, or more math or you just want it, you can use bc to evaluate arithmetic expressions.
if i ran "echo $[3/4]" at the command prompt, it would return 0 because bash only uses integers when answering. If you ran "echo 3/4|bc -l", it would properly return 0.75.
From a message from mike (see Thanks to)
you always use #!/bin/bash .. you might was to give an example of
how to find where bash is located.
'locate bash' is preferred, but not all machines have locate.
'find ./ -name bash' from the root dir will work, usually.
Suggested locations to check:
ls -l /bin/bash
ls -l /sbin/bash
ls -l /usr/local/bin/bash
ls -l /usr/bin/bash
ls -l /usr/sbin/bash
ls -l /usr/local/sbin/bash
(can't think of any other dirs offhand... i've found it in
most of these places before on different system).
You may try also 'which bash'.
In bash, the return value of a program is stored in a special variable called $?.
This illustrates how to capture the return value of a program, I assume that the directory dada does not exist. (This was also suggested by mike)
#!/bin/bash
cd /dada &> /dev/null
echo rv: $?
cd $(pwd) &> /dev/null
echo rv: $?
This little scripts show all tables from all databases (assuming you got MySQL installed). Also, consider changing the 'mysql' command to use a valid username and password.
#!/bin/bash
DBS=`mysql -uroot -e"show databases"`
for b in $DBS ;
do
mysql -uroot -e"show tables from $b"
done
You can use multiple files with the command source.
__TO-DO__
(1) s1 = s2
(2) s1 != s2
(3) s1 < s2
(4) s1 > s2
(5) -n s1
(6) -z s1
(1) s1 matches s2
(2) s1 does not match s2
(3) __TO-DO__
(4) __TO-DO__
(5) s1 is not null (contains one or more characters)
(6) s1 is null
Comparing two strings.
#!/bin/bash
S1='string'
S2='String'
if [ $S1=$S2 ];
then
echo "S1('$S1') is not equal to S2('$S2')"
fi
if [ $S1=$S1 ];
then
echo "S1('$S1') is equal to S1('$S1')"
fi
I quote here a note from a mail, sent buy Andreas Beck, refering to use if [ $1 = $2 ].
This is not quite a good idea, as if either $S1 or $S2 is empty, you will get a parse error. x$1=x$2 or "$1"="$2" is better.
+
-
*
/
% (remainder)
-lt (<)
-gt (>)
-le (<=)
-ge (>=)
-eq (==)
-ne (!=)
C programmer's should simple map the operator to its corresponding parenthesis.
This section was re-written by Kees (see thank to...)
Some of these command's almost contain complete programming languages. From those commands only the basics will be explained. For a more detailed description, have a closer look at the man pages of each command.
sed (stream editor)
Sed is a non-interactive editor. Instead of altering a file by moving the cursor on the screen, you use a script of editing instructions to sed, plus the name of the file to edit. You can also describe sed as a filter. Let's have a look at some examples:
$sed 's/to_be_replaced/replaced/g' /tmp/dummy
Sed replaces the string 'to_be_replaced' with the string 'replaced' and reads from the /tmp/dummy file. The result will be sent to stdout (normally the console) but you can also add '> capture' to the end of the line above so that sed sends the output to the file 'capture'.
$sed 12, 18d /tmp/dummy
Sed shows all lines except lines 12 to 18. The original file is not altered by this command.
awk (manipulation of datafiles, text retrieval and processing)
Many implementations of the AWK programming language exist (most known interpreters are GNU's gawk and 'new awk' mawk.) The principle is simple: AWK scans for a pattern, and for every matching pattern a action will be performed.
Again, I've created a dummy file containing the following lines:
"test123
test
tteesstt"
$awk '/test/ {print}' /tmp/dummy
test123
test
The pattern AWK looks for is 'test' and the action it performs when it found a line in the file /tmp/dummy with the string 'test' is 'print'.
$awk '/test/ {i=i+1} END {print i}' /tmp/dummy
3
When you're searching for many patterns, you should replace the text between the quotes with '-f file.awk' so you can put all patterns and actions in 'file.awk'.
grep (print lines matching a search pattern)
We've already seen quite a few grep commands in the previous chapters, that display the lines matching a pattern. But grep can do more.
$grep "look for this" /var/log/messages -c
12
The string "look for this" has been found 12 times in the file /var/log/messages.
[ok, this example was a fake, the /var/log/messages was tweaked :-)]
wc (counts lines, words and bytes)
In the following example, we see that the output is not what we expected. The dummy file, as used in this example, contains the following text: "bash introduction howto test file"
$wc --words --lines --bytes /tmp/dummy
2 5 34 /tmp/dummy
Wc doesn't care about the parameter order. Wc always prints them in a standard order, which is, as you can see: .
sort (sort lines of text files)
This time the dummy file contains the following text:
"b
c
a"
$sort /tmp/dummy
This is what the output looks like:
a
b
c
Commands shouldn't be that easy :-) bc (a calculator programming language)
Bc is accepting calculations from command line (input from file. not from redirector or pipe), but also from a user interface. The following demonstration shows some of the commands. Note that
I start bc using the -q parameter to avoid a welcome message.
$bc -q
1 == 5
0
0.05 == 0.05
1
5 != 5
0
2 ^ 8
256
sqrt(9)
3
while (i != 9) {
i = i + 1;
print i
}
123456789
quit
tput (initialize a terminal or query terminfo database)
A little demonstration of tput's capabilities:
$tput cup 10 4
The prompt appears at (y10,x4).
$tput reset
Clears screen and prompt appears at (y1,x1). Note that (y0,x0) is the upper left corner.
$tput cols
80
Shows the number of characters possible in x direction.
It it higly recommended to be familiarized with these programs (at least). There are tons of little programs that will let you do real magic on the command line.
[some samples are taken from man pages or FAQs]
#!/bin/bash
SRCD="/home/"
TGTD="/var/backups/"
OF=home-$(date +%Y%m%d).tgz
tar -cZf $TGTD$OF $SRCD
#!/bin/sh
# renna: rename multiple files according to several rules
# written by felix hudson Jan - 2000
#first check for the various 'modes' that this program has
#if the first ($1) condition matches then we execute that portion of the
#program and then exit
# check for the prefix condition
if [ $1 = p ]; then
#we now get rid of the mode ($1) variable and prefix ($2)
prefix=$2 ; shift ; shift
# a quick check to see if any files were given
# if none then its better not to do anything than rename some non-existent
# files!!
if [$1 = ]; then
echo "no files given"
exit 0
fi
# this for loop iterates through all of the files that we gave the program
# it does one rename per file given
for file in $*
do
mv ${file} $prefix$file
done
#we now exit the program
exit 0
fi
# check for a suffix rename
# the rest of this part is virtually identical to the previous section
# please see those notes
if [ $1 = s ]; then
suffix=$2 ; shift ; shift
if [$1 = ]; then
echo "no files given"
exit 0
fi
for file in $*
do
mv ${file} $file$suffix
done
exit 0
fi
# check for the replacement rename
if [ $1 = r ]; then
shift
# i included this bit as to not damage any files if the user does not specify
# anything to be done
# just a safety measure
if [ $# -lt 3 ] ; then
echo "usage: renna r [expression] [replacement] files... "
exit 0
fi
# remove other information
OLD=$1 ; NEW=$2 ; shift ; shift
# this for loop iterates through all of the files that we give the program
# it does one rename per file given using the program 'sed'
# this is a sinple command line program that parses standard input and
# replaces a set expression with a give string
# here we pass it the file name ( as standard input) and replace the nessesary
# text
for file in $*
do
new=`echo ${file} | sed s/${OLD}/${NEW}/g`
mv ${file} $new
done
exit 0
fi
# if we have reached here then nothing proper was passed to the program
# so we tell the user how to use it
echo "usage;"
echo " renna p [prefix] files.."
echo " renna s [suffix] files.."
echo " renna r [expression] [replacement] files.."
exit 0
# done!
#!/bin/bash
# renames.sh
# basic file renamer
criteria=$1
re_match=$2
replace=$3
for i in $( ls *$criteria* );
do
src=$i
tgt=$(echo $i | sed -e "s/$re_match/$replace/")
mv $src $tgt
done
A nice thing to do is to add on the first line
#!/bin/bash -x
This will produce some intresting output information
Feel free to make suggestions/corrections, or whatever you think it would be interesting to see in this document. I'll try to update it as soon as I can.
This documents comes with no warranty of any kind. and all that
Italian: by William Ghelfi (wizzy at tiscalinet.it) is here
French: by Laurent Martelli is missed
Korean: Minseok Park http://kldp.org
Korean: Chun Hye Jin unknown
Spanish: unknow http://www.insflug.org
I guess there are more translations, but I don't have any info of them, if you have it, please, mail it to me so I update this section.
New translations included and minor correcitons.
Added the section usefull commands re-writen by Kess.
More corrections and suggestions incorporated.
Samples added on string comparison.
v0.8 droped the versioning, I guess the date is enought.
v0.7 More corrections and some old TO-DO sections written.
v0.6 Minor corrections.
v0.5 Added the redirection section.
v0.4 disapperd from its location due to my ex-boss and thid doc found it's new place at the proper url: www.linuxdoc.org.
prior: I don't rememeber and I didn't use rcs nor cvs :(
Introduction to bash (under BE) http://org.laol.net/lamug/beforever/bashtut.htm
Bourne Shell Programming http://207.213.123.70/book/