How do I write a 'for' loop in Bash?

I'm looking for the basic loop like:

for(int i = 0; i < MAX; i++) {
  doSomething(i);
}

but for Bash.

From this site:

for i in $(seq 1 10);
do
    echo $i
done

for ((i = 0 ; i < max ; i++ )); do echo "$i"; done

The Bash for consists on a variable (the iterator) and a list of words where the iterator will, well, iterate.

So, if you have a limited list of words, just put them in the following syntax:

for w in word1 word2 word3
do
  doSomething($w)
done

Probably you want to iterate along some numbers, so you can use the seq command to generate a list of numbers for you: (from 1 to 100 for example)

seq 1 100

and use it in the for loop:

for n in $(seq 1 100)
do
  doSomething($n)
done

Note the $(...) syntax. It's a Bash behaviour, and it allows you to pass the output from one command (in our case from seq) to another (the for).

This is really useful when you have to iterate over all directories in some path, for example:

for d in $(find $somepath -type d)
do
  doSomething($d)
done

The possibilities are infinite to generate the lists.

Bash 3.0+ can use this syntax:

for i in {1..10} ; do ... ; done

...which avoids spawning an external program to expand the sequence (such as seq 1 10).

Of course, this has the same problem as the for(()) solution, being tied to Bash and even a particular version (if this matters to you).

Try the Bash built-in help:

help for

Output

for: for NAME [in WORDS ... ;] do COMMANDS; done
    The `for' loop executes a sequence of commands for each member in a
    list of items.  If `in WORDS ...;' is not present, then `in "$@"' is
    assumed.  For each element in WORDS, NAME is set to that element, and
    the COMMANDS are executed.

for ((: for (( exp1; exp2; exp3 )); do COMMANDS; done
    Equivalent to
        (( EXP1 ))
        while (( EXP2 )); do
            COMMANDS
            (( EXP3 ))
        done
    EXP1, EXP2, and EXP3 are arithmetic expressions.  If any expression is
    omitted, it behaves as if it evaluates to 1.

#! /bin/bash

function do_something {
   echo value=${1}
}

MAX=4
for (( i=0; i<MAX; i++ )) ; {
   do_something ${i}
}

---

Here's an example that can also work in older shells, while still being efficient for large counts:

Z=$(date) awk 'BEGIN { for ( i=0; i<4; i++ ) { print i,"hello",ENVIRON["Z"]; } }'

But good luck doing useful things inside of awk: https://stackoverflow.com/q/19075671/86967

I commonly like to use a slight variant on the standard for loop. I often use this to run a command on a series of remote hosts. I take advantage of Bash's brace expansion to create for loops that allow me to create non-numerical for loops.

Example:

I want to run the uptime command on frontend hosts 1-5 and backend hosts 1-3:

% for host in {frontend{1..5},backend{1..3}}.mycompany.com
    do ssh $host "echo -n $host; uptime"
  done

I typically run this as a single-line command with semicolons on the ends of the lines instead of the more readable version above. The key usage consideration are that braces allow you to specify multiple values to be inserted into a string (e.g. pre{foo,bar}post results in prefoopost, prebarpost) and allow counting/sequences by using the double periods (you can use a..z, etc.). However, the double period syntax is a new feature of Bash 3.0; earlier versions will not support this.

If you're interested only in Bash, the "for(( ... ))" solution presented above is the best, but if you want something POSIX SH compliant that will work on all Unices, you'll have to use "expr" and "while", and that's because "(())" or "seq" or "i=i+1" are not that portable among various shells.

I use variations of this all the time to process files... > for files in *.log; do echo "Do stuff with: $files"; echo "Do more stuff with: $files"; done;

If processing lists of files is what you're interested in, look into the -execdir option for files.