1
1
mirror of https://github.com/adambard/learnxinyminutes-docs.git synced 2024-12-21 14:21:47 +03:00
learnxinyminutes-docs/bash.html.markdown

502 lines
16 KiB
Bash
Raw Normal View History

2013-08-18 02:59:35 +04:00
---
2013-08-19 20:14:02 +04:00
category: tool
tool: bash
2013-08-18 02:59:35 +04:00
contributors:
2013-08-19 20:14:02 +04:00
- ["Max Yankov", "https://github.com/golergka"]
- ["Darren Lin", "https://github.com/CogBear"]
2013-09-22 08:06:35 +04:00
- ["Alexandre Medeiros", "http://alemedeiros.sdf.org"]
2013-12-18 12:37:27 +04:00
- ["Denis Arh", "https://github.com/darh"]
2014-07-23 10:43:18 +04:00
- ["akirahirose", "https://twitter.com/akirahirose"]
2014-08-06 22:30:15 +04:00
- ["Anton Strömkvist", "http://lutic.org/"]
- ["Rahil Momin", "https://github.com/iamrahil"]
- ["Gregrory Kielian", "https://github.com/gskielian"]
2015-04-26 20:22:48 +03:00
- ["Etan Reisner", "https://github.com/deryni"]
- ["Jonathan Wang", "https://github.com/Jonathansw"]
- ["Leo Rudberg", "https://github.com/LOZORD"]
- ["Betsy Lorton", "https://github.com/schbetsy"]
- ["John Detter", "https://github.com/jdetter"]
- ["Harry Mumford-Turner", "https://github.com/harrymt"]
2018-01-16 18:27:30 +03:00
- ["Martin Nicholson", "https://github.com/mn113"]
2013-08-18 02:59:35 +04:00
filename: LearnBash.sh
translators:
- ["Dimitri Kokkonis", "https://github.com/kokkonisd"]
2013-08-18 02:59:35 +04:00
---
2018-12-24 01:59:11 +03:00
Bash is a name of the unix shell, which was also distributed as the shell
for the GNU operating system and as the default shell on most Linux distros.
2018-12-24 01:59:11 +03:00
Nearly all examples below can be a part of a shell script
or executed directly in the shell.
2013-08-18 02:59:35 +04:00
[Read more here.](https://www.gnu.org/software/bash/manual/bashref.html)
2013-08-18 02:59:35 +04:00
```bash
2018-07-26 00:58:31 +03:00
#!/usr/bin/env bash
2018-12-24 01:59:11 +03:00
# First line of the script is the shebang which tells the system how to execute
# the script: https://en.wikipedia.org/wiki/Shebang_(Unix)
2013-08-18 02:59:35 +04:00
# As you already figured, comments start with #. Shebang is also a comment.
# Simple hello world example:
echo Hello world! # => Hello world!
2013-08-18 02:59:35 +04:00
2018-12-24 01:59:11 +03:00
# Each command starts on a new line, or after a semicolon:
2013-08-18 02:59:35 +04:00
echo 'This is the first line'; echo 'This is the second line'
# => This is the first line
# => This is the second line
2013-08-18 02:59:35 +04:00
# Declaring a variable looks like this:
Variable="Some string"
2013-08-18 02:59:35 +04:00
# But not like this:
Variable = "Some string" # => returns error "Variable: command not found"
# Bash will decide that Variable is a command it must execute and give an error
# because it can't be found.
2018-12-24 01:59:11 +03:00
# Nor like this:
Variable= 'Some string' # => returns error: "Some string: command not found"
# Bash will decide that 'Some string' is a command it must execute and give an
# error because it can't be found. (In this case the 'Variable=' part is seen
# as a variable assignment valid only for the scope of the 'Some string'
# command.)
2013-08-18 02:59:35 +04:00
# Using the variable:
echo $Variable # => Some string
echo "$Variable" # => Some string
echo '$Variable' # => $Variable
2013-08-19 20:14:02 +04:00
# When you use the variable itself — assign it, export it, or else — you write
2015-10-02 17:08:27 +03:00
# its name without $. If you want to use the variable's value, you should use $.
2013-09-22 08:29:05 +04:00
# Note that ' (single quote) won't expand the variables!
# Parameter expansion ${ }:
echo ${Variable} # => Some string
# This is a simple usage of parameter expansion
# Parameter Expansion gets a value from a variable.
2018-12-24 01:59:11 +03:00
# It "expands" or prints the value
# During the expansion time the value or parameter can be modified
# Below are other modifications that add onto this expansion
2013-09-22 08:29:05 +04:00
# String substitution in variables
echo ${Variable/Some/A} # => A string
2015-05-26 04:25:07 +03:00
# This will substitute the first occurrence of "Some" with "A"
2013-09-22 08:29:05 +04:00
2014-02-28 20:24:45 +04:00
# Substring from a variable
Length=7
echo ${Variable:0:Length} # => Some st
2014-02-28 20:24:45 +04:00
# This will return only the first 7 characters of the value
2018-01-16 18:31:32 +03:00
echo ${Variable: -5} # => tring
# This will return the last 5 characters (note the space before -5)
# String length
echo ${#Variable} # => 11
2014-02-28 20:24:45 +04:00
# Indirect expansion
OtherVariable="Variable"
echo ${!OtherVariable} # => Some String
2020-04-20 21:42:18 +03:00
# This will expand the value of OtherVariable
2013-12-18 12:37:27 +04:00
# Default value for variable
echo ${Foo:-"DefaultValueIfFooIsMissingOrEmpty"}
# => DefaultValueIfFooIsMissingOrEmpty
# This works for null (Foo=) and empty string (Foo=""); zero (Foo=0) returns 0.
# Note that it only returns default value and doesn't change variable value.
2013-12-18 12:37:27 +04:00
2017-11-14 22:30:57 +03:00
# Declare an array with 6 elements
array0=(one two three four five six)
# Print first element
echo $array0 # => "one"
# Print first element
echo ${array0[0]} # => "one"
# Print all elements
echo ${array0[@]} # => "one two three four five six"
# Print number of elements
echo ${#array0[@]} # => "6"
# Print number of characters in third element
echo ${#array0[2]} # => "5"
# Print 2 elements starting from forth
echo ${array0[@]:3:2} # => "four five"
# Print all elements. Each of them on new line.
for i in "${array0[@]}"; do
echo "$i"
done
# Brace Expansion { }
# Used to generate arbitrary strings
echo {1..10} # => 1 2 3 4 5 6 7 8 9 10
echo {a..z} # => a b c d e f g h i j k l m n o p q r s t u v w x y z
# This will output the range from the start value to the end value
2018-12-24 01:59:11 +03:00
# Built-in variables:
# There are some useful built-in variables, like
2015-10-02 17:08:27 +03:00
echo "Last program's return value: $?"
2013-09-22 08:29:05 +04:00
echo "Script's PID: $$"
2015-10-02 17:08:27 +03:00
echo "Number of arguments passed to script: $#"
echo "All arguments passed to script: $@"
echo "Script's arguments separated into different variables: $1 $2..."
2013-08-18 03:14:00 +04:00
# Now that we know how to echo and use variables,
# let's learn some of the other basics of bash!
# Our current directory is available through the command `pwd`.
# `pwd` stands for "print working directory".
2018-12-24 01:59:11 +03:00
# We can also use the built-in variable `$PWD`.
# Observe that the following are equivalent:
echo "I'm in $(pwd)" # execs `pwd` and interpolates output
echo "I'm in $PWD" # interpolates the variable
# If you get too much output in your terminal, or from a script, the command
# `clear` clears your screen
clear
# Ctrl-L also works for clearing output
2013-08-18 03:14:00 +04:00
# Reading a value from input:
echo "What's your name?"
read Name # Note that we didn't need to declare a new variable
echo Hello, $Name!
2013-08-18 02:59:35 +04:00
# We have the usual if structure:
2018-12-24 01:59:11 +03:00
# use `man test` for more info about conditionals
2016-01-12 02:32:03 +03:00
if [ $Name != $USER ]
2013-08-18 02:59:35 +04:00
then
2013-12-27 19:21:24 +04:00
echo "Your name isn't your username"
2014-08-23 10:54:13 +04:00
else
echo "Your name is your username"
2013-08-18 02:59:35 +04:00
fi
# True if the value of $Name is not equal to the current user's login username
2013-08-18 02:59:35 +04:00
# NOTE: if $Name is empty, bash sees the above condition as:
2016-01-12 02:32:03 +03:00
if [ != $USER ]
# which is invalid syntax
# so the "safe" way to use potentially empty variables in bash is:
2016-01-12 02:32:03 +03:00
if [ "$Name" != $USER ] ...
# which, when $Name is empty, is seen by bash as:
2016-01-12 02:32:03 +03:00
if [ "" != $USER ] ...
# which works as expected
2013-09-22 08:06:35 +04:00
# There is also conditional execution
2013-12-27 19:21:24 +04:00
echo "Always executed" || echo "Only executed if first command fails"
# => Always executed
2013-09-22 08:06:35 +04:00
echo "Always executed" && echo "Only executed if first command does NOT fail"
# => Always executed
# => Only executed if first command does NOT fail
2013-08-23 00:26:26 +04:00
2014-08-06 22:30:15 +04:00
# To use && and || with if statements, you need multiple pairs of square brackets:
if [ "$Name" == "Steve" ] && [ "$Age" -eq 15 ]
2014-08-06 22:30:15 +04:00
then
echo "This will run if $Name is Steve AND $Age is 15."
2014-08-06 22:30:15 +04:00
fi
if [ "$Name" == "Daniya" ] || [ "$Name" == "Zach" ]
2014-08-06 22:30:15 +04:00
then
echo "This will run if $Name is Daniya OR Zach."
2014-08-06 22:30:15 +04:00
fi
2018-01-16 19:03:13 +03:00
2018-12-24 01:59:11 +03:00
# There is also the `=~` operator, which tests a string against a Regex pattern:
2018-01-16 19:03:13 +03:00
Email=me@example.com
if [[ "$Email" =~ [a-z]+@[a-z]{2,}\.(com|net|org) ]]
then
echo "Valid email!"
fi
# Note that =~ only works within double [[ ]] square brackets,
# which are subtly different from single [ ].
# See https://www.gnu.org/software/bash/manual/bashref.html#Conditional-Constructs for more on this.
2014-08-06 22:30:15 +04:00
2018-12-24 01:59:11 +03:00
# Redefine command `ping` as alias to send only 5 packets
2017-11-14 22:30:57 +03:00
alias ping='ping -c 5'
2018-12-24 01:59:11 +03:00
# Escape the alias and use command with this name instead
2017-11-14 22:30:57 +03:00
\ping 192.168.1.1
# Print all aliases
alias -p
2014-08-06 22:30:15 +04:00
2013-08-18 03:14:00 +04:00
# Expressions are denoted with the following format:
echo $(( 10 + 5 )) # => 15
2013-08-18 03:14:00 +04:00
2015-10-02 17:08:27 +03:00
# Unlike other programming languages, bash is a shell so it works in the context
# of a current directory. You can list files and directories in the current
2013-12-27 19:21:24 +04:00
# directory with the ls command:
ls # Lists the files and subdirectories contained in the current directory
2018-12-24 01:59:11 +03:00
# This command has options that control its execution:
ls -l # Lists every file and directory on a separate line
ls -t # Sorts the directory contents by last-modified date (descending)
ls -R # Recursively `ls` this directory and all of its subdirectories
# Results of the previous command can be passed to the next command as input.
2018-12-24 01:59:11 +03:00
# The `grep` command filters the input with provided patterns.
# That's how we can list .txt files in the current directory:
ls -l | grep "\.txt"
# Use `cat` to print files to stdout:
cat file.txt
# We can also read the file using `cat`:
Contents=$(cat file.txt)
# "\n" prints a new line character
# "-e" to interpret the newline escape characters as escape characters
echo -e "START OF FILE\n$Contents\nEND OF FILE"
# => START OF FILE
# => [contents of file.txt]
# => END OF FILE
# Use `cp` to copy files or directories from one place to another.
# `cp` creates NEW versions of the sources,
# so editing the copy won't affect the original (and vice versa).
# Note that it will overwrite the destination if it already exists.
cp srcFile.txt clone.txt
cp -r srcDirectory/ dst/ # recursively copy
# Look into `scp` or `sftp` if you plan on exchanging files between computers.
# `scp` behaves very similarly to `cp`.
# `sftp` is more interactive.
# Use `mv` to move files or directories from one place to another.
# `mv` is similar to `cp`, but it deletes the source.
# `mv` is also useful for renaming files!
mv s0urc3.txt dst.txt # sorry, l33t hackers...
# Since bash works in the context of a current directory, you might want to
# run your command in some other directory. We have cd for changing location:
cd ~ # change to home directory
2018-10-03 05:23:45 +03:00
cd # also goes to home directory
cd .. # go up one directory
# (^^say, from /home/username/Downloads to /home/username)
cd /home/username/Documents # change to specified directory
cd ~/Documents/.. # still in home directory..isn't it??
2018-10-03 05:23:45 +03:00
cd - # change to last directory
# => /home/username/Documents
# Use subshells to work across directories
(echo "First, I'm here: $PWD") && (cd someDir; echo "Then, I'm here: $PWD")
pwd # still in first directory
# Use `mkdir` to create new directories.
mkdir myNewDir
# The `-p` flag causes new intermediate directories to be created as necessary.
mkdir -p myNewDir/with/intermediate/directories
# if the intermediate directories didn't already exist, running the above
# command without the `-p` flag would return an error
# You can redirect command input and output (stdin, stdout, and stderr).
# Read from stdin until ^EOF$ and overwrite hello.py with the lines
# between "EOF":
cat > hello.py << EOF
#!/usr/bin/env python
from __future__ import print_function
import sys
print("#stdout", file=sys.stdout)
print("#stderr", file=sys.stderr)
for line in sys.stdin:
print(line, file=sys.stdout)
EOF
# Variables will be expanded if the first "EOF" is not quoted
# Run the hello.py Python script with various stdin, stdout, and
# stderr redirections:
python hello.py < "input.in" # pass input.in as input to the script
2018-12-24 01:59:11 +03:00
python hello.py > "output.out" # redirect output from the script to output.out
2018-12-24 01:59:11 +03:00
python hello.py 2> "error.err" # redirect error output to error.err
2018-12-24 01:59:11 +03:00
python hello.py > "output-and-error.log" 2>&1
# redirect both output and errors to output-and-error.log
python hello.py > /dev/null 2>&1
# redirect all output and errors to the black hole, /dev/null, i.e., no output
# The output error will overwrite the file if it exists,
# if you want to append instead, use ">>":
python hello.py >> "output.out" 2>> "error.err"
# Overwrite output.out, append to error.err, and count lines:
info bash 'Basic Shell Features' 'Redirections' > output.out 2>> error.err
wc -l output.out error.err
# Run a command and print its file descriptor (e.g. /dev/fd/123)
# see: man fd
echo <(echo "#helloworld")
# Overwrite output.out with "#helloworld":
cat > output.out <(echo "#helloworld")
echo "#helloworld" > output.out
echo "#helloworld" | cat > output.out
echo "#helloworld" | tee output.out >/dev/null
# Cleanup temporary files verbosely (add '-i' for interactive)
# WARNING: `rm` commands cannot be undone
rm -v output.out error.err output-and-error.log
rm -r tempDir/ # recursively delete
# You can install the `trash-cli` Python package to have `trash`
# which puts files in the system trash and doesn't delete them directly
# see https://pypi.org/project/trash-cli/ if you want to be careful
# Commands can be substituted within other commands using $( ):
2013-08-19 20:14:02 +04:00
# The following command displays the number of files and directories in the
# current directory.
2013-08-18 04:33:32 +04:00
echo "There are $(ls | wc -l) items here."
# The same can be done using backticks `` but they can't be nested -
# the preferred way is to use $( ).
echo "There are `ls | wc -l` items here."
2018-12-24 01:59:11 +03:00
# Bash uses a `case` statement that works similarly to switch in Java and C++:
case "$Variable" in
# List patterns for the conditions you want to meet
0) echo "There is a zero.";;
1) echo "There is a one.";;
*) echo "It is not null.";; # match everything
2013-08-18 04:33:32 +04:00
esac
2018-12-24 01:59:11 +03:00
# `for` loops iterate for as many arguments given:
# The contents of $Variable is printed three times.
for Variable in {1..3}
2013-08-18 04:33:32 +04:00
do
echo "$Variable"
2013-08-18 04:33:32 +04:00
done
# => 1
# => 2
# => 3
2013-08-18 04:33:32 +04:00
# Or write it the "traditional for loop" way:
for ((a=1; a <= 3; a++))
do
echo $a
done
# => 1
# => 2
# => 3
2014-08-08 21:24:43 +04:00
# They can also be used to act on files..
2018-12-24 01:59:11 +03:00
# This will run the command `cat` on file1 and file2
for Variable in file1 file2
2014-08-08 21:24:43 +04:00
do
cat "$Variable"
2014-08-08 21:24:43 +04:00
done
# ..or the output from a command
2018-12-24 01:59:11 +03:00
# This will `cat` the output from `ls`.
for Output in $(ls)
2014-08-08 21:24:43 +04:00
do
cat "$Output"
2014-08-08 21:24:43 +04:00
done
# Bash can also accept patterns, like this to `cat`
# all the Markdown files in current directory
for Output in ./*.markdown
do
cat "$Output"
done
2013-11-25 19:42:37 +04:00
# while loop:
2014-08-08 21:11:17 +04:00
while [ true ]
2013-11-25 19:42:37 +04:00
do
echo "loop body here..."
break
done
# => loop body here...
2013-11-25 19:42:37 +04:00
2013-09-22 18:28:59 +04:00
# You can also define functions
# Definition:
function foo ()
2013-09-22 18:28:59 +04:00
{
echo "Arguments work just like script arguments: $@"
echo "And: $1 $2..."
echo "This is a function"
return 0
}
# Call the function `foo` with two arguments, arg1 and arg2:
foo arg1 arg2
# => Arguments work just like script arguments: arg1 arg2
# => And: arg1 arg2...
# => This is a function
2013-09-22 18:28:59 +04:00
# or simply
bar ()
{
echo "Another way to declare functions!"
return 0
}
# Call the function `bar` with no arguments:
bar # => Another way to declare functions!
2013-09-22 18:28:59 +04:00
# Calling your function
foo "My name is" $Name
2013-09-22 18:35:04 +04:00
# There are a lot of useful commands you should learn:
# prints last 10 lines of file.txt
tail -n 10 file.txt
2013-09-22 18:35:04 +04:00
# prints first 10 lines of file.txt
head -n 10 file.txt
2013-09-22 18:35:04 +04:00
# sort file.txt's lines
sort file.txt
2013-09-22 18:35:04 +04:00
# report or omit repeated lines, with -d it reports them
uniq -d file.txt
2013-09-22 18:35:04 +04:00
# prints only the first column before the ',' character
cut -d ',' -f 1 file.txt
# replaces every occurrence of 'okay' with 'great' in file.txt
# (regex compatible)
sed -i 's/okay/great/g' file.txt
# be aware that this -i flag means that file.txt will be changed
# -i or --in-place erase the input file (use --in-place=.backup to keep a back-up)
2015-04-24 00:06:54 +03:00
# print to stdout all lines of file.txt which match some regex
# The example prints lines which begin with "foo" and end in "bar"
grep "^foo.*bar$" file.txt
# pass the option "-c" to instead print the number of lines matching the regex
grep -c "^foo.*bar$" file.txt
# Other useful options are:
grep -r "^foo.*bar$" someDir/ # recursively `grep`
grep -n "^foo.*bar$" file.txt # give line numbers
grep -rI "^foo.*bar$" someDir/ # recursively `grep`, but ignore binary files
# perform the same initial search, but filter out the lines containing "baz"
grep "^foo.*bar$" file.txt | grep -v "baz"
2015-04-24 00:06:54 +03:00
# if you literally want to search for the string,
# and not the regex, use `fgrep` (or `grep -F`)
2016-05-12 14:07:50 +03:00
fgrep "foobar" file.txt
# The `trap` command allows you to execute a command whenever your script
# receives a signal. Here, `trap` will execute `rm` if it receives any of the
# three listed signals.
trap "rm $TEMP_FILE; exit" SIGHUP SIGINT SIGTERM
# `sudo` is used to perform commands as the superuser
# usually it will ask interactively the password of superuser
NAME1=$(whoami)
NAME2=$(sudo whoami)
echo "Was $NAME1, then became more powerful $NAME2"
2018-12-24 01:59:11 +03:00
# Read Bash shell built-ins documentation with the bash `help` built-in:
help
help help
help for
help return
help source
help .
2018-12-24 01:59:11 +03:00
# Read Bash manpage documentation with `man`
apropos bash
man 1 bash
man bash
2018-12-24 01:59:11 +03:00
# Read info documentation with `info` (`?` for help)
apropos info | grep '^info.*('
man info
info info
info 5 info
# Read bash info documentation:
info bash
info bash 'Bash Features'
info bash 6
info --apropos bash
2013-08-18 04:33:32 +04:00
```