- 1 BASIC COMMANDS
- 1.1 Where stuff is located
- 2 PERMISSIONS
- 2.1 Commands
- 3 FILES AND FOLDERS MANIPULATION
- 3.1 Wildcards
- 3.2 Commands
- 4 REDIRECTIONS
- 5 THE SHELL
- 5.1 Expressions
- 5.2 Keyboard tricks
- 6 PROCESSES AND MEMORY
- 7 CONFIGURATION AND ENVIRONMENT
- 7.1 Customize the prompt
- 7.1.1 Colors
- 8 PACKAGE MANAGEMENT
- 9 MEDIA STORAGE
- 9.1 Commands
- 10 NETWORKS
- 11 SEARCHING FOR FILES
- 12 ARCHIVING AND BACKUP
- 13 REGULAR EXPRESSIONS
- 13.1 Ranges
- 13.2 Counters
- 14 TEXT PROCESSING
- 15 FORMATTING OUTPUT
- 16 BASH SCRIPTING BASICS
- 16.1 Flow control
- 16.1.1 Strings Expressions evaluation
- 16.1.2 Integers Expressions evaluation
- 16.1.3 Logical operators
- 16.2 Reading inputs
- 16.2.1 Positionals params
- 16.3 Flow control
- 16.4 Variables
- 16.5 Strings
- 16.6 Numbers
- 16.7 Arrays
To chain more than one command use ;
or &&
cal
date
df
show free space of each partition-h
show sizes in human readable format
free
show free space on disk-h
show sizes in human readable format
file
filename shows the MIMEtype of a filels
-a
show hiddens-d
show only directories-F
classify (if it ends with a/
is a directory)l
alone does the same-h
show sizes in human readable format-l
long format-r
reverse order-S
sort by size-t
sort by timetype
display the command typewhich
display executable locationman
command display help for the commandapropos
command list of man pages for given commandwhatis
command display short command descriptioninfo
command the same ofman
alias
list all the aliasesalias alias_name='command chain'
There are not spaces around=
unalias alias_name
destroy aliashistory
shows the list of the most recent commands with an history number,!history_number
executes that commandalt+>
show the first command in historyalt+<
show the last command in history
/
The system root
/bin
System binaries
/boot
Linux kernel
/dev
List of connected devices
/etc
System configuration files and scripts
/home
Each users has his folder here
/lib
System shared libraries
/lost
+found recovery files when crash happens
/media
Mounting point of external devices
/opt
Optional software installed by users
/proc
Kernel stuff
/sbin
System binaries
/tmp
Temporary files
/usr
Users stuff
/usr/bin
, /usr/lib
, /usr/local
, /usr/sbin
Stuff belonging to user's programs
/usr/share
, /usr/share/doc
Shared user datas and user programs libs
/var
Databases and logs are all here
There are 10 symbols. The first one is for the type the number (d/L/c/b). The number 2,3 and 4 are the owner's permissions, the 5, 6 and 7 are for the group, 8, 9 and 10 are the permissions of the rest of the world. r
means read, w
write, x
execute.
Value | Symbols |
---|---|
0 | --- |
1 | --x |
2 | -w- |
3 | -wx |
4 | r-- |
5 | r-x |
6 | rw- |
7 | rwx |
Another way to change permission is the symbolic notation:
u
userg
groupo
worlda
all (u+g+o)+
add permission-
remove permission=
set permission
id
show the user identity and his groupschmod
change modeumask
set default permissionssu
run as another user (for that session of the shell)su -c
command execute a single command as another user (the same ofsudo
)sudo
run as another user (for single command)chgrp
change groupchown
change owner.chown ownername [:[array of groups]] arguments
permits to set user and groups in one shotpasswd
*
means any char
?
one char
[expression]
set of chars [!expression]
the !
means exclude
Expression can be ranges or classes of chars. Classes are:
[:alnum:]
a-z A-Z 0-9[:alpha:]
a-z A-Z[:digit:]
0-9[:lower:]
a-z[:upper:]
A-Z
mkdir
create directory
ln
create symbolic link
cp
copy
-a
copy owner and permissions too-i
interactive copy (otherwise it s overwrites silently)-r
recursive copy-u
update (copy or overwrite only new or younger)-v
verbose
mv
move or rename
-i
interactive (otherwise it s overwrites silently)-r
recursive-u
update (only new or younger)
rm
remove files or directories
-i
interactive (otherwise it removes without confirmation)-r
recursive-f
force-v
verbose
Remember! There is no UNDO for rm
!
cat
concats params in output destinationsort
sort lines in outputuniq
remove duplicates from output-d
only duplicatesgrep
any string or regexp apply that as filter-i
ignore case-v
reverse matchingwc
count words, lines and bytes of the outputhead
print from the beginning of the output-n
specify number of lines (default 10)
tail
print from the end of the output-n
specify number of lines (default 10)
tee
get from specified input and write into specified output>
write from the beginning>>
append here2>
write to standar error output&>
destination Redirects file (STDERR) and (STDOUT) to destination&>/dev/null
Redirects file (STDERR) and (STDOUT) to `/dev/null´. This basically means to ignore output.
$((expression))
where expression is any arithmetic calculation
{range_start..range_end}
prints all the combinations (it works with reversed ranges too)
An uppercased name preceded by $
is a system variable
Is possible to use double quotes in a command to ignore whitespaces.
What is contained between single quotes is considered plain text
\
to escape a single caracter
\b
is a backspace
\n
is a newline
\r
is the carriage return
\t
is a tab
ctrl+l
clear screenctrl+t
transpose actual char with previous onealt+t
transponse actual word with previous onealt+l
lowercase from cursor to endalt+u
uppercase from cursor to endctrl+k
erase linectrl+u
erase from cursor to the beginning of the linealt+d
erase from the cursor to the end of the linealt+backspace
erase last word- double tab display all the possible completions
-
ps
snapshot of running processes.R
means process running,S
sleeping,D
unstoppable sleep,T
stopped,Z
zombie,N
low priority,<
high priority process -
aux
shows USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND -
top
display running tasks in real time and 5 lines containing the summary- uptime, number of users and average load for queued processes
- task situations
- CPU state
- amout of KU in use
- SWAP (virtual memory) in use
-
jobs
display running jobs -
bg
place a job in background -
fgl
place a job in foreground -
kill
send a signal to a process1 HUP
hangup (the dial connection)2 INT
interrupt (asctrl+c
)9 KILL
kill the process without performing any cleanups15 TERM
terminate process18 CONT
resume stopped process19 STOP
pause process3 QUIT
11 SEGV
segmentation (memory) violation20 TSTP
terminal stop28 WINCH
window resize/change
-
killall
Send a signal to all processes running any of the specified commands. If no signal name is specified, SIGTERM is sent. -
lshw
List hardware -
-html
Outputs the device tree as an HTML page. -
-xml
Outputs the device tree as an XML tree. -
-json
Outputs the device tree as a JSON object -
-short
Outputs the device tree showing hardware paths -
-businfo
Outputs the device list showing bus information, detailing SCSI, USB, IDE and PCI addresses. -
-C
type of hardware, for examplememory
-
shutdown
shuts down the system -
pstree
show processes as parent child tree -
vmstat
Report virtual memory statistics: processes, memory, paging, block IO, traps, and cpu activity. -
xload
opens a windo with system stats -
tload
draws in the termina a graph with system stats
set
set shell optionsexport
put in the environmentalias
printenv
shows all the system variables Some interesting variables:EDITOR
,HOME
,PAGER
,SHELL
,LANG
,PATH
(where executables are searched),PS1
(prompt string),TERM
,PWD
,TZ
,USER
The system configuration is established by the following files in this order:/etc/procfile
, ~/.bash_profile
, ~/bash_login
, ~/profiler
. If the user is not logged in the shell starts loading /etc/.bashrc
and ~/.bashrc
.
To reload the environment source .bashrc
.
echo $PS1
to show the actual prompt config (generally stored in .bashrc
).
In the output is possible to find \w
for the working directory and \h
for the full hostname.
Each color number is included between [033[
and ]
and followed by a m
.
Black | 0;30 | Dark Gray | 1;30 |
Blue | 0;34 | Light Blue | 1;34 |
Green | 0;32 | Light Green | 1;32 |
Cyan | 0;36 | Light Cyan | 1;36 |
Red | 0;31 | Light Red | 1;31 |
Purple | 0;35 | Light Purple | 1;35 |
Brown | 0;33 | Yellow | 1;33 |
Light Gray | 0;37 | White | 1;37 |
Packages of different systems are, obvioulsly, not compatibles; for example deb
and rpm
.
All systems use 2 kind of tools: low and high level management tools.
Low level tools are for install and remove packages into or from the system.
High level tools are to perform metadata searches and dependency resolution.
For Ubuntu the low level one is dpkg
and the high level are. apt-get
and aptitute
.
dpkg --install
packagefile install the given package without performing the dependecy resolution.
dpkg --list
list all the installed packages.
dpkg --status
packagename show the package status
dpkg -r
packagename removes the package
dpkg --search
packagename tells who installed that package
apt-cache search
packagename (or string) looks for a repo for that package (or containing one with the given string).
apt-get install
packagename install the package and performs the dependecy resolution.
apt-get remove
removes the package
apt-get update
and apt-get upgrade
perform a global update pr upgrade
apt-cache show
packagename show infos about the given package
/etc/fstab
contains ext3
hard disk partitions. It shows:
- file system
- mount point
- file system type
- options Mount options of access to the device/partition
- dump frequency of backups; if is set to 0 dum is disabled
- pass Controls the order in which fsck checks the device/partition for errors at boot time. The root device should be 1. Other partitions should be 2, or 0 to disable checking.
ls /dev
lists all the connected devices
mount
if executed without params shows actuals mounted filesystems-t
devicename (or isoname)mounting point
mounts the filesystem. Examplesudo mount -o loop /<path>/<filename>.iso /mnt/iso
to mount an ISO image as it was an external cd.unmount
fdisk /dev/
devicename formats that devicefsck
Check and repairmkfs
create a file systemdd
write info devicewodim
burn cdmd5sum
calculate MD5 checksum
ping
traceroute
servername shows the first 30 hopt to reach the specified server. * * * means DNS lookup failure.-n
display IPs instead of servernames-w
waiting timenetstat
shows netweork settings and statistics-k
display the kernel routing tableftp
wget
url to download it-O
name to change the name of the file when downloaded-i
textfile containing a list of url to download them all in one shotwget --ftp-user=USERNAME --ftp-password=PASSWORD URL
to download from an FTP with authentication--user-agent=AGENT URL
to fake an user aged as mozilla or chrome-r
recursively-H
convert links to local version-level=NUMBER
level of recursion. 0 means infinitessh
user@host to log into a remote hostssh host command
runs a comand on a remote hostscp
scp [email protected]:foobar.txt /some/local/directory
Copy the file "foobar.txt" from a remote host to the local hostscp foobar.txt [email protected]:/some/remote/directory
Copy the file "foobar.txt" from the local host to a remote hostscp -r foo [email protected]:/some/remote/directory/bar
Copy the directory "foo" from the local host to a remote host's directory "bar"scp [email protected]:/some/remote/directory/foobar.txt [email protected]:/some/remote/directory/
Copy the file "foobar.txt" from remote host "rh1.edu" to remote host "rh2.edu"scp foo.txt bar.txt [email protected]:~
Copying the files "foo.txt" and "bar.txt" from the local host to your home directory on the remote hostscp -P 2264 foobar.txt [email protected]:/some/remote/directory
Copy the file "foobar.txt" from the local host to a remote host using port 2264scp [email protected]:/some/remote/directory/\{a,b,c\} .
Copy multiple files from the remote host to your current directory on the local hostscp [email protected]:~/\{foo.txt,bar.txt\}.
locate
Find by name into system db. The internal system db is updated daily but it can be manually updated withupdatedb
find
accepts some options-name
-size
-empty
-groupname
-newer
or-cnewer
filename as time reference-nogroup
No group corresponds to file's numeric group ID.-nouser
No user corresponds to file's numeric user ID.-type
Supported. POSIX specifiesb
,c
,d
,l
,p
,f
ands
. GNU find also supportsD
, representing a Door, where the OS provides these.
And can perform some action on the resulting search
-
-delete
-
-ls
-
-print
-
-quit
exit immediatly without doing anything else -
-exec
execute the command without asking. it accepts an array of params -
-ok
Like-exec
but ask the user first. If the user agrees, run the command. -
xargs
command take the input and use it as param for the specified command -
--null
or-0
Input items are terminated by a null character instead of by whitespace; the quotes and backslash are not special (every character is taken literally). -
touch
Update the access and modification times of each file to the current time. A file that does not exist is created empty, unless -c is supplied. -
-c
do not create anything -
stat
filename or dirname or filesystem display properties
gzip
filename compress and replace-d
is the same ofunzip
gunzip
filename uncompress and replace-l
For each compressed file list compressed size, uncompressed size, ratio, uncompressed_name-t
Check the compressed file integrity-r
recursively-#
(0-9),--fast
--best
velocity VS more compressionbzip
The same ofgzip
but using a different algorithmtar
Makes a single file out of multiple files or extract from an archive. Example:tar -zxvf data.tar.gz -C location
-c
create-x
extract-r
append to the end of an archive-t
list the content-f
specify the archive name-z
Uncompress the resulting archive with gzip command.-C
Change directory to the specified locationzip
join and compress Filesunzip
location filenamersync [OPTION] SOURCE [DEST]
-a
create an unique archive-r
recursive-u
skip files that are newer on the receiver-t
preserve times-p
preserve permissions-g
preserve group-o
preserve owner (super-user only)-n
perform a trial run with no changes made--delete
delete files which no loger exists on source--exclude
regexp exclude matches from syncronization
grep
means global regular expression print
-i
ignore case-v
invert match (aka exclude)-c
count the number of matches-l
print the name of the file containing the match instead of the match itself-L
filenames that not match-n
print the line number where the match occurr-h
suppress filenames in the output^
string starting with$
string ending with
[chars]
Specified chars[^chars]
Exclude specified chars[start-end]
likea-k
or0-9
[:alnum:]
a-Z
and0-9
[:word:]
Alphanumeric and_
[:alpha:]
a-Z
[:blank:]
Spaces and tabs[:cntrl:]
Ascii chars from 0 to 127[:digit:]
0-9
[:graph:]
Any visible char[:lower:]
Lowercase letters[:punct:]
Punctuaction chars[:print:]
graph
range and space[:upper:]
Uppercase chars[:xdigit:]
Chars used in exadecimal numbers
?
More than 1+
at least 1{n,m}
at least n to maximum m
cat
concatenate-A
display non printing chars too-n
display line numbers-s
suppress multiple blank linessort
sort lines. Is possible specify an offset to a column. Examplesort -k 3-7
means sort by the seventh char of the third field-b
ignore leading blanks-f
ignore case-m
merge-n
numeric sort-o
output to a file-r
reverse-t
specify column separatoruniq
Report or omit repeated lines. With no options, matching lines are merged to the first occurrence.-c
prefix lines by the number of occurrences-d
only print duplicate lines, one for each group-f
Number avoid comparing the first N fields-i
ignore case-s
Number avoid comparing the first N characterscut
remove sections from file. Examplecut -d ':' -f filename
-c
select only these characters-d
delimiter Use specified delimiter instead of TAB for field delimiter. It sould go between'
-f
field list select only these fields--complement
reverse extractionpaste
Copy content from origin to destinationjoin
Join lines of two files on a common field.comm
Compare 2 sorted files. With no options, produce three-column output. Column one contains lines unique to FILE1, column two contains lines unique to FILE2, and column three contains lines common to both files. Files must be sorted. Examplecomm <(sort firstfile) <(sort secondfile)
-1
Suppress column 1 (lines unique to FILE1)-2
Suppress column 2 (lines unique to FILE2)-3
Suppress column 3 (lines that appear in both files)--check-order
Check that the input is correctly sorted, even if all input lines are pairablediff
Compare 2 sorted files showing differences-c
Show context (display the first file before and after the second one)-u
Compact difference (git diff style). Without, equal lines are grouped-a
Treat all as text-r
Recursively compare any subdirectories foundpatch
receives a file containingdiff
output and apply those difference to a given target.tr
Convert input changing a set of chars with another one-s
replace each input sequence of a repeated character that is listed in SET1 with a single occurrence of that charactersed
Stream Editaspell
Check spelling interactively and replace errors,--dont-backup
to avoid the creation of a.bak
file as backup--add-html-check=<list>
,--rem-html-check=<list>
Add or remove a list of HTML attributes to always check.--add-html-skip=<list>
,--rem-html-skip=<list>
Add or remove a list of HTML attributes to always skip while spell checking
nl
numebr lines of files-b
style use style for numbering body lines. Styles can be:-a
All-t
Non blank-n
None-p
Number only lines matching regexp-f
Number footer-h
Number header-i
number Step increment-n
Number format-ln
Left justified-rn
Right justified-rz
Right justified with leading zeros-p
Do not reset page number-s
Add a string after the number-v
First line number on each logical page-w
number Use the given number for line numbersfold
Break lines at specified width-w
Specify width (default is 80)-s
Respect words spacesfmt
Text formatterpri
Paginate for printingprintf
Print a string following the specified format%d
as decimal%f
as float%o
integer as octal%s
as string%x
as hexadecimal%y
as hexadecimal in upcase%%
print a single%
groff
and family are a group of text editors as LATEX
Every script must start with #!/bin/bash
Every script must have read and execute permission
Is a good practice to use the long form for commands in scripts
variable=value
must be written without spaces around the =
.
$variable
echoes the variable value.
Variables can't start start with numbers, spaces and puntuation signs are not allowed.
declare -r UPPERCASED_NAME=VALUE
creates a constant.
{enclose variable name}
to avoid confusion in case of composed names.
function name {code}
is a funcion declaration. Function are called by name and they must be declared before their call.
Local variabled inside a function are declared with LOCAL
.
A script executed with the -x
flag will output all the executed lines.
set -x
(and set +x
) can be used to trace only a section of the script
TIME
command returns the amount of time spent executing the given command.
{}
and ()
can be used to group a set of commands.
< ()
sends the output of the group to its parent.
trap [commands] [signals]
Creates a listener to the system signals to catch them (for example keyboard abort sequences) and execute the specified command. Example trap "echo 'killing signal'" SIGINT SIGTERM
.
sleep
seconds stops the execution for the amount of given seconds
wait
PID Waits for a process to finish it execution.
IF [$var=value];THEN
ELSIF
ELSE
FI
When a command executes, it returns 0
if it success or any other value up to 255
if it fails.
The exit
command return the execution status
To execute an expression, can be used the command test
or the more powerful [[
expression ]]
. For integers the command is ((
expression ))
.
Parenthesis can be used to group expression but they must be escaped.
-n
string length > 0-z
string length = 0==
equal!=
not equal<
or>
given 2 strings, check if the first string orders before (or after) the second one
eq
equalne
not equalle
less than or equallt
less thange
greater than or equalgt
greater
Logical operators can be used between commands too.
-a
for string is written as[[&&]]
and for integers((&&))
. It means and.-o
for string is written as[[||]]
and for integers((||))
. It means or.!
for string is written as[[!]]
and for integers((!))
. It means not.
read
varname (withour $)
read
varname1, varname2, varname3, varname4 ...
If less values are provided, the rest will be set as empty string. Space is used to separate input assignments. The default receiver is the variable REPLY
.
The read command can't be piped because pipes create subshells, each with its own REPLY variable.
The read
command supprts the following options:
-a
assign input to an array-d
delimiter declare a string delimiter which indicates the end of the input. (the default is the new line char)-e
the input can be edited before submitting-n
number read only X chars-p
display a prompt-v
raw mode. it means that\
is not an escape char but is traten as simple text-s
silent mode-t
seconds timeout-u
read from file
The IFS
(internal field separator) command can be used to change the separator between columns (which by default is space). Example IFS=":"
The commands <<
and <<<
stay for here document and here string, everything follows these commands is considered all input until read delimiter is found, no matter what it contains or how large it could be.
$0...$9
by default, ${10}...
from numbers with more than one digit.
WHILE [expression]; DO
code
DONE
UNTIL [expression]; DO
code
DONE
They can be used to read files too.
WHILE [expression]; DO
code
DONE < file.txt
CASE input IN
option1)
;;
option....)
;;
ESAC
;;
means _break. The option)
can be a values or a regexp or a set (like [:alpha:])
FOR var IN [words]; DO
code
DONE
FOR ((i=0;i<5;i=i+1));DO
code
DONE
variable=value
Assignmentlet var2=var1 operation
Set var2 with the result of the operation over var1.echo $variable
Echoes the variable value$variable
and${variable:+default}
Execute the variable values as command, if it is null does nothing${variable:-default}
Execute the variable values as command, if it is null execute the default value as command{$variable:=default}
Execute the variable values as command, if it is null set the variable to the default value and executes it.{$variable:?default}
Execute the variable values as command, if it is null return the default value as error messageecho ${!prefix*}
return a list of variables beginning with the give prefix.
${#string}
String length${string:position:length}
Extracts substring -${string#pattern}
Removes the leading shortest match from string -${string##pattern}
Removes the leading largest match from string -${string%pattern}
Removes the starting shortest match from string -${string%%pattern}
Removes the starting largest match from string -${value_to_find/string/replacing_value}
replace value to find with replacing value. The search accepts://
global replace/#
replace at the beginning/%
replace at the end
Numbers are in decimal base by default. Numbers beginning with 0
are considered octals, numbers beginning with 0x
are considered hexadecimals.
Assignments are the common ones: =
, +=
, -=
, *=
, /=
, %=
, ++
, --
bc
Is the bash decimal calculator. It could be used with the here doc notation. Example: bc -l <<< '10/3'
, without -l
the output is rounded to the nearest integer. Alternatively, scale=
number can be used to specify the number of decimals: bc <<< 'scale=2; 10/3'
There is no maximum limit to the size of an array, nor any requirement that member variables be indexed or assigned contiguously. Arrays are zero-based: the first element is indexed with the number 0.Return to top
declare -a ARRAYNAME
Instantiate an array.ARRAY[index]=value
Assign the value to the specified index.*
and@
Mean all the keys of the array.#ARRAY
Return the array length#array[@]
Returns all the keys of the given array#array[key]
Return the value of the given key+=
Push a new value into the arrayunset
deletes the array ot the element at the given key