Ref
https://guide.bash.academy/expansions/
Definition
bash wraps the kernel so you can launch processes
and
if [ -r 1 ] && [ -s 2 ]
or
if [ $USER == 'bob' ] || [ $USER == 'andy' ]
for loop
for i in $(cat countries.txt); do
echo -n ${i:0:1} | tr '[:lower:]' '[:upper:]';
echo ${i:1};
done >> countries.txt
for loop
for i in $(seq 1 10); do echo $i; done
for loop
for ((i=1;i<10;i++)); do echo $i; done
if statement
if [ <test> ]; then echo foo; fi
if-elif statement
if [ <test> ]; then echo 1; elif [ <test2> ]; then echo 2; else echo 3; fi
if directory exists
if [ -d <directory> ];
if file exists
if [ -f <directory> ];
if symbolic link exists
if [ -L <directory ];
case statements
case $1 in start) echo start;; stop) echo stop;; restart) echo restart;; *) echo no clue;; esac
while statements
while [ <test> ]; do echo 1; done
Read one line at a time
<file while read; do echo "$REPLY"; done
Read one line at a time
<file while read line; do echo "$line"; done
simple test
if [[ $REPLY = y ]]; then echo 1; fi
Asterix (*)
matches any text or no text
Question Mark (?)
matches a single character
Char Set ([chars])
matches single character as long as its within the set
Character Classes [[:classname:]]
alnum, alpha, ascii, blank, cntrl, digit, graph, lower, print, punct, space, upper, word, xdigit
Tilde Home Path (~)
Expands to current user's home directory (i.e. /home/user)
Arbitrary fd Redirection
Use
When multiple patterns are needed to be used.
+(pattern | pattern)
Matches when any of the patterns appears at least once
*(pattern | pattern)
Matches when patterns appear at least once or not at all (what the heck?)
?(pattern | pattern)
Matches once or not at all
@(pattern | pattern)
Matches when any of the patterns appear exactly once
!(pattern | pattern)
Matches only when none of the patterns appear
Example 1
ls *.jp?(e)g
Example 2
ls *.@(jpg|jpeg)
$(cat /etc/passwd)
Launches command in subshell (new bash process)
`cat /etc/passwd`
Deprecated syntax for same thing
var1=myname
echo "My name is $name"
contents="$(cat hello.txt)"
echo "Contents here: ${contents}<--end of file"
name=Britta time=23.73
echo "$name's current record is ${time%.*} seconds and ${time#*.} hundredths."
more expansion
echo "PATH currently contains: ${PATH//:/, }"
${parameter#pattern}
Remove the shortest string that matches pattern (start) "${url#*/}" /www.hooperlabs.xyz/h00p.html
${parameter##pattern}
Remove the longest string that matches pattern (start) "${url##*/}" h00p.html
${parameter%pattern}
Remove the shortest string that matches the pattern (end) "${url%/*}" https://www.hooperlabs.xyz
${parameter%%pattern}
Remove longest string matching pattern (end) "${url%%/*}" https:
${parameter/pattern/replacement}
Replace first string matching pattern "${url/./-}" https://www-hooperlabs.xyz/h00p.html
${parameter//pattern/replacement}
Replace each string matching pattern "${url//./-}" https://www-hooperlabs-xyz/h00p.html
${parameter/#pattern/replacement}
Replace string matching pattern (start) "${url/#*:/http:}" http://www.hooperlabs.xyz/h00p.html
${parameter/%pattern/replacement}
Replace string matching pattern at end "${url/%.html/.jpg}" https://www.hooperlabs.xyz/h00p.jpg
${#parameter}
Expand length of the value (in bytes) "${#url}" 36
${parameter:start:length}
Expand part of value starting at start, length bytes long "${url:7}" www.hooperlabs.xyz/h00p.html
${parameter[^|^^|,|,,]pattern}
Expand the transformed value, either upper or lower (first char or all chars) of string that matches the pattern. "${url^^[ht]}" HTTps://www.Hooperlabs.xyz/h00p.HTml
echo ${PATH//:/ }
Substitute : for spaces in $PATH.
{1...10}
numbers in sequence (separated by spaces)
{01..10..2}
count by twos
"$1"
first positional parameter
"$*"
expands single string, joining all parameters into one, separated by $IFS
"$@"
expands the positional parameters as a list of separate arguments
"$#"
expands into a number indicating the number of positional parameters
"$?"
expands the exit code of last command (0 means success)
"$-"
expands the set of option flags currently active in shell
"$$"
expands PID for current shell process
"$!"
expands unique PID foir last process started in background
"$_"
expands to the last argument of previous command
"!!"
expands to full previous line
"echo !:2"
expands to second argument of last command
"!-2"
second command in history
"!:2-5"
all but first argument of last command
"~+"
current working directory
"~-"
old working directory
BASH
full pathname of command that started the current bash shell
BASH_VERSION
active version of bash
BASH_VERSIONINFO
array of detailed info about current version of bash
BASH_SOURCE
filenames of the scripts currently running (usually empty)
BASHPID
PID of the bash that is parsing script code
UID
uid of user within bash shell
HOME
pathname of the home directory
HOSTNAME
name of computer
LANG
preferred language category
MACHTYPE
full description of type of system
PWD
pathname to current directory
OLDPWD
pathname of directory you were in before
RANDOM
expands random number between 0 and 32767
SECONDS
number of seconds bash shell has been active
LINES
contains height (rows) of terminal display
COLUMNS
contains width (single char spaces) of single row in display
IFS
"Internal Field Separator" - bash uses this for word-splitting of data. By default, spaces, tabs, and newlines
PATH
List of paths bash will search for executable programs when a command is run.
PS1
Contains string that describes what the prompt should look like
PS2
describes what the secondary prompt will look like
array
files=( myscript hello.txt "foo bar.txt" ); rm -rv "${files[@]}"
array (caution)
Use quotes around array!!! don't use ${files[@]}
array append
files+=( selfie.png )
array patterns
files=( *.txt )
array first element
echo "${files[0]}"; echo "$files"
array remove element
unset "files[3]"
array number of elements
echo "${#files[@]}"
array retrieve some elements
echo "${names[@]:1:2}"; echo "${names[@]: -2}"
Return all text between two words
sed -n -e '/BEGIN/./END/ p'
Substitute string for another (and delete first line)
sed -e 1d -e s/foo/bar/g /tmp/foo
Substitute string for another (and delete first line)
sed '1d; s/foo/bar/g' /tmp/foo
Delete lines from file containing string
sed -i '/192\.168\.0\.1/d' /var/log/apache2/access.log
Use awk as cut
echo a,b,c,d,e | awk -F, '{print $1 $2 $3 $5}'