Table of Contents
Shell
Block comment
With “END” being just an example of any random string that must not appear in the commented section:
: << 'END'
sleep 1
END
Iterate over a list as a string
Different options:
# option 1, without variable for x in a b c; do echo $x; done # option 2, with string list="a b c" for x in $list; do echo $x; done # only works with sh and bash for x in ${=list}; do echo $x; done # only works with zsh for x in $(echo $list}; do echo $x; done # works with any shell # option 3, with an array list=(a b c) for x in $list; do echo $x; done
Do something recursively on all files
Use find
, with {}
being the file name:
find . -type f -exec <command> {} \; find -iname "<pattern>" -exec <command> {} \;
Manual solution:
#!/bin/sh function search_dir() { DIR=$1; for file in $DIR/*; do if [[ -f $file ]]; then echo "Do something with $file" fi; done for element in $DIR/* ; do if [[ -e $element && -d $element && \ $(basename "$element") != ".." && \ $(basename "$element") != "." ]]; then search_dir "$element"; fi; done; } search_dir "."
Misc
- Default value for variable:
${<variable>:-<default-value>}
- Get directory of current script (for instance in order to run a script that is stored in the same directory, or in a known relative path):
BASEDIR=$(dirname "$0")
- echo to stderr, just redirect the output:
echo "foo" 1>&2
- Convert dash-time to ISO (2024-04-21_15-50-30 to 2024-04-21T15:50:30, for instance to use it as input for
date
):sed -r "s/([0-9]{4})-([0-9]{2})-([0-9]{2})_([0-9]{2})-([0-9]{2})-([0-9]{2})/\1-\2\-3T\4:\5:\6/"
- Get target name of symbolic link:
readlink <symbolic-link>
- Execute command:
var=`cat file` var=$(cat file)
- Evaluate mathematical expression (
zsh
also supports floating-point operations):echo $((2*3))
- time with redirected output (also works for groups of commands):
time (cat file1 > file2) time (cat file1 > /dev/null; sync)
- Check number of arguments and print usage:
if [[ $# -lt 2 ]]; then echo "Usage: foo.sh <arg1> <arg2>" 1>&2 exit 1 fi
- Load command output as a file:
diff <(grep x file1) <(grep y file2)