====== BASH - Cheat Sheet ======

===== String Manipulation =====

<code bash>
${str:position}         # substring starting at position
${str:position:len}     # substring starting at position with length len
${str#ubstring}         # delete shortest match from front
${str##substring}       # delete longest match from front
${str%substring}        # delete shortest match from back
${str%%substring}       # delete longest match from back
${str/pattern/replacement}  # pattern replace
${str/#pattern/replacement} # pattern replace at front
${str/%pattern/replacement} # pattern replace at end
${str//pattern/replacement} # global pattern replace
</code>

----

===== Parameter Substituion =====

<code bash>
${str-default}             # set 'default' if not set
${str:-default}            # set 'default' if not set (even if declared null)

${str=default}             # set 'default' if not set
${str:=default}            # set 'default' if not set (even if declared null)

${str+value}               # set 'value' if $str is set, otherwise set null
${str:+value}              # set 'value' if $str is set (even if declared null), otherwise set null

${str?error}               # abort with 'error' if not set
${str:?error}              # abort with 'error' if not set (and not null)
</code>

----

===== Arrays =====

Indexed arrays require no declaration

<code bash>
arr=("string 1", "string 2", "string 3")
arr=([1]="string 1", [2]="string 2", [3]="string 3")

arr[4]="string 4"
</code>

Check below under "Hashes" for accessing the different properties of an array.

----

===== Hashes =====

Since Bash v4

<code bash>
# Hashes need declaration!
declare -A arr

# Assigning values to associative arrays
arr[my key]="my value"
arr["my key"]="my value"
arr[$my_key]="my value"

# Fetching values
echo ${arr[my key]}
echo ${arr["my key"]}
echo ${arr[$my_key]}

# Accessing the array
${arr[@]}         # Returns all indizes and their items (doesn't work with associative arrays)
${arr[*]}         # Returns all items
${!arr[*]}        # Returns all indizes
${#arr[*]}        # Number elements
${#arr[$n]}       # Length of $nth item

# Pushing to array
arr+=("new string value", "another new value")
</code>

----

===== Here Documents =====

Bash allows here documents like this

<code bash>
cat <<EOT
[...]
EOT
</code>

To disable substitution in the here doc text quote the marker with single or double quotes.

<code bash>
cat <<'EOT'
</code>

To strip leading tabs use

<code bash>
cat <<-EOT
</code>

----

===== Debugging Scripts =====

For simple tracing add a

<code bash>
set -x
</code>

in the script or append the "-x" to the shebang or run the script like this:

<code bash>
bash -x <script name>
</code>

As "set -x" enables tracing you can disable it with "set +x" again. This allows tracing only a part of the code (e.g. a condition in an inner loop). Additionally to "-x" you may want to set "-v" to see the shell commands that are executed. Combine both to

<code bash>
set -xv
</code>

----

===== Writing Safer Scripts =====

It is a good practice to use

<code bash>
set -euo pipefail
</code>

which ensures

  * that you check all commands that can fail
  * that you check for failing commands within pipes
  * that you do not use undefined variables

----

===== Network Connections =====

<code bash>
# Establish a connection to 91.92.93.94:80 on file handle 4 with
if ! exec 4<> /dev/tcp/91.92.93.94/80; then
    echo "ERROR: Connection failed!"
fi

# Write something
echo -e "GET / HTTP/1.0\n" >&4

# Read something
cat <&4

# Close the socket
exec <&4-
exec >&4-
</code>

----

===== File Operators =====

The complete list of bash 4.2 test operators:

<code bash>
      -a FILE        True if file exists.
      -b FILE        True if file is block special.
      -c FILE        True if file is character special.
      -d FILE        True if file is a directory.
      -e FILE        True if file exists.
      -f FILE        True if file exists and is a regular file.
      -g FILE        True if file is set-group-id.
      -h FILE        True if file is a symbolic link.
      -L FILE        True if file is a symbolic link.
      -k FILE        True if file has its sticky bit set.
      -p FILE        True if file is a named pipe.
      -r FILE        True if file is readable by you.
      -s FILE        True if file exists and is not empty.
      -S FILE        True if file is a socket.
      -t FD          True if FD is opened on a terminal.
      -u FILE        True if the file is set-user-id.
      -w FILE        True if the file is writable by you.
      -x FILE        True if the file is executable by you.
      -O FILE        True if the file is effectively owned by you.
      -G FILE        True if the file is effectively owned by your group.
      -N FILE        True if the file has been modified since it was last read.

      FILE1 -nt FILE2  True if file1 is newer than file2 (according to
                       modification date).

      FILE1 -ot FILE2  True if file1 is older than file2.

      FILE1 -ef FILE2  True if file1 is a hard link to file2.
</code>

----

===== String Operators =====

The complete list of bash 4.2 string operators:

<code bash>
      -z STRING      True if string is empty.

      -n STRING
         STRING      True if string is not empty.

      STRING1 = STRING2
                     True if the strings are equal.
      STRING1 != STRING2
                     True if the strings are not equal.
      STRING1 < STRING2
                     True if STRING1 sorts before STRING2 lexicographically.
      STRING1 > STRING2
                     True if STRING1 sorts after STRING2 lexicographically.
</code>

----

===== Simulate Reading From a File =====

Sometimes you might need to pass a file name when you want to pipe output from a commands.

Then you could write to a file first and then used it, but you can also use the ">()" or "\<()" operator.

This can be used with all tools that demand a file name parameter:

<code bash>
diff <(echo abc;echo def) <(echo abc;echo abc)
</code>

----

===== History =====

==== History Handling ====

Here are some improvements for the bash history handling:

<code bash>
unset HISTFILE      # Stop logging history in this bash instance
HISTIGNORE="[ ]*"   # Do not log commands with leading spaces
HISTIGNORE="&"      # Do not log a command multiple times

# Change up/down arrow key behavior to navigate only similar commands
bind '"\e[A":history-search-backward'
bind '"\e[B":history-search-forward'
bind '"\e[C":forward-char'
bind '"\e[D":backward-char'
</code>


----

==== Adding Timestamps ====

To add timestamps to your history set the following environment variable:

<code bash>
HISTTIMEFORMAT="%Y-%m-%d %H:%M:%S " # Log with timestamps
</code>

----

==== Easier History Navigation ====

If you do not like Ctrl-R to navigate the history you can define other keys as PgUp and PgDown in /etc/inputrc:

<code bash>
   "\e[5~": history-search-backward
   "\e[6~": history-search-forward
</code>

----

==== History Hardening ====

For a secure bash configuration add the following settings to your global/users bashrc

<code bash>
HISTIGNORE=""
HISTCONTROL=""
HISTTIMEFORMAT='%Y-%m-%d %H:%M:%S '
HISTFILE=~/.bash_history
HISTFILESIZE=2000
readonly HISTFILE
readonly HISTSIZE
readonly HISTFILESIZE
readonly HISTIGNORE
readonly HISTCONTROL
readonly HISTTIMEFORMAT
shopt -s histappend
</code>

and finally mark the history file append only:

<code bash>
chattr +a $HISTFILE
</code>

----

===== Misc =====

==== Waiting for child processes ====

See: https://spin.atomicobject.com/2017/08/24/start-stop-bash-background-process/

<code bash>
trap "exit" INT TERM ERR
trap "kill 0" EXIT

background_something &

wait
</code>

----

==== Kill childs on exit ====

<code bash>
trap 'kill $(jobs -p)' EXIT
</code>

----

==== Command Completion ====

Setup your own bash completion schemas.

Here is a git example:

<code bash>
complete -W 'add branch checkout clone commit diff grep init log merge mv pull push rebase rm show status tag' git

complete -p    # To list defined completion schemes
</code>

<WRAP info>
**NOTE:**  This example probably already comes prepared with your Linux distribution.

You might want to check default definitions installed in /etc/bash_completion.d for a good starting point.

</WRAP>

----

==== Kill all childs on exit ====

<code bash>
trap true TERM
kill -- -$$
</code>

----

==== Apply ulimit Changes Instantly ====

Try to use the "-i" switch:

<code bash>
sudo -i -u <user>
</code>

<WRAP info>
**NOTE:**  If it does not work,  you might need to investigate and change the PAM configuration.
</WRAP>


----

==== PS1: Escape Non-Print Chars ====

To avoid incorrect line break behaviour when editing the command line you need to escape control characters in PS1 like this:

<code bash>
\[color definition\]
</code>

For example:

<code bash>
\[\033[31m\] some text \[\033[0m\]
</code>