quick reference on command line tools and techniques for the people with limited time
The command line quick reference is a github repository by Utsav Barman
1. Introduction
1.1. Scope
- This document will mainly focus on things that are usually done in the command line
- The focus is on tools and techniques for one shot adhoc tasks.
- Most of the things in the document can be applied to shell scripting. However, shell scripting is not the focus of this document
- This document is for Linux/Unix
1.2. Background
- I am documenting things I learnt and am still learning
- I had multiple files with random snippets.
- I decided to organize the mess into a single piece, so that I can efficiently look up things
- I cannot claim to be an expert in command line.
1.3. Purpose
- The document will not help anyone become an expert.
- However, it may help people become effective.
- I have kept the document concise. I skipped over details that I never used much and can be looked up later in the internet.
- I have focused on concrete examples of common or interesting use cases.
1.4. Next steps
- rephrase unclear material
- correct incorrect material
- add missing material
- add further reading material
- Contributions are welcome
2. Basics
2.1. Common commands
command | description |
---|---|
man | get help on commands. e.g.man date |
date | gives the date and time |
cal | shows the calender |
ls | tells which files are present in the current working directory when a -l option is used, the command returns owner, size, date of file, persissions, etc. |
cat | shows contents of a file |
cp | copy a file |
mv | move a file |
diff | lists differences between 2 files |
rm | removes file |
grep | find occurences of strings in one or more files |
pwd | print the present working directory’s name |
cd | changes the current directory |
2.2. Shortcuts
keyboard shortcut | description |
---|---|
Ctrl+A | move the cursor to the start of the line |
Ctrl+E | move the cursor to the end of the line |
Alt+F | move the cursor forward by one word |
Ctrl+F | move the cursor forward by one character |
Alt+B | move the cursor backward by one word |
Ctrl+B | move the cursor backward by one charater |
Ctrl+XX | toggle the cursor’s position between the current position and the previous position |
keyboard shortcut | description |
---|---|
Ctrl+U | cut all the characters to the left of the cursor |
Ctrl+K | cut all the characters to the right of the cursor |
Ctrl+W | cut one word to the left of the cursor |
Ctrl+H | cut one character to the left of the cursor |
Alt+D | cut one word to the right of the cursor |
Ctrl+D | cut one character to the right of the cursor |
Ctrl+Y | paste the cut characters |
Ctrl+_ | undo the last deletion |
Tab | complete arguments or list all available commands |
keyboard shortcut | description |
---|---|
Crtl+R | search the command history |
Ctrl+G | abort |
Ctrl+P/UP | the previous command in the history |
Ctrl+N/DOWN | the next command in the history |
3. Streams, Pipes and Redirects
3.1. Streams
A linux shell’s inputs and outputs are sequence of characters called streams. There are three standard I/O streams:
name | description | file descriptor |
---|---|---|
stdout | displays output from commands | 1 |
stderr | displays error from commands | 2 |
stdin | provides inputs to commands | 0 |
3.2. Redirections
Input and output redirections are done using angular brackets (<>)
Bracket type | description |
---|---|
> | send stream to a file. E.g.ls a > o.txt |
>> | append stream to a file. E.g.ls b >> o.txt |
>& | write into stream. E.g.ls c > o2.txt 2>&1 |
< | receive stream from a file. E.g.wc < o.txt |
<< | embed the text that will be fed to a command within the script Example cat << EOF > output.txt line 1 line 2 line 3 EOF echo done also known as heredoc |
<<< | expand variable and feed into the stdin of the command Example wc <<< $word also known as herestring |
3.2.2.1. Send standard output to sout.txt and standard error to serr.txt
command1 > sout.txt 2> serr.txt
3.2.2.2. Send standard output and standard error streams to the same file sone.txt
command1 > sone.txt 2>&1 OR command1 &> sone.txt
3.2.2.3. Check if a string is present in a shell variable
grep "ABC" <<< $var
3.2.2.4. Ignore both standard input and output
command1 &> /dev/null
/dev/null is a null device file. This will discard anything written to it, and will return EOF on reading.
3.3. Pipe
Piping can redirect the standard output of one command to the input of another command.
command1 | command2 paramater1 | command3 parameter1 parameter2 | command4
Examples:
- Print file contents only once. remove duplicate records:
sort file1 | uniq
- print 5 most frequently used commands
history | awk '{a[$2]++}END{for(i in a){print a[i] " " i}}' | sort -rn | head -5
- print the file types and their frequencies
ls | rev | cut -f1 -d'.' | rev | sort | uniq -c | sort -n
3.4. xargs
- short for extended arguments
- some commands can take arguments from both standard input and as command-line arguments.
- however, there are some commands that cannot take input from standard input. They accept inputs only from arguments. For these commands we need to use args.
- xargs converts input from standard input into arguments to a command.
- divides the arguments to a permitted number and runs the command repeatedly over each greoup of arguments.
- using -n option, the number of arguments per command can be specified.
find . | xargs -n1 basename
- in order to assign the std input to a placeholder , use
-I{}
. the placeholder is usually used when we want to place the std input in the middle of a commandls | xargs -I{} echo {} file is found
- xargs, by default, divides the input into arguments using whitespace. When -0 option is used, xargs uses null termination to identify arguments.
Examples:
- print the number lines/words/characters in the files in a directory
ls | wc
- print the file types and their frequencies
find . -type f | xargs basename -a | grep "\." | rev | cut -f1 -d'.' | rev | sort | uniq -c | sort -n
- rename all files in a directory
ls | xargs -I{} mv {} {}.bkp
- copy all .c files to a specific directory
find . -name "*.c" -print0 | xargs -0 -n1 -I{} cp {} some/folder/
The print0 option of the find command prints the file names with the null character as termination character. By using the -0 option with xargs, it starts looking for the null character for reading filenames from the stdin.
3.5. tee
- the command reads from standard input and writes it to a standard output and to one or more files
- this is useful when we want to see the output of a command both on the screen and also want to save the output in a file for later analysis.
Examples:
ls | tee fileList.txt
4. Filename expansion
- a wildcard character is a character that is used to represent one or more characters in a filename or foldername.
- file globbing is the operation that recognizes these wildcard characters and does the expansion.
wildcard | description | example | matches | does not match |
---|---|---|---|---|
* | matches 0 or more characters | ls to* | to, tom, ton, tow, tommy, tommie | tata, tea |
? | matches 1 character | ls to? | tom, tow, ton | to, tommy, tommie, tata, tea |
[abc] | matches any one of the characters in the brackets | ls [bc]at | bat, cat | Bat, Cat, rat |
[a-z] | matches any one of the characters within the range in the brackets | ls day[1-9] | day1, day2 upto day9 | day11, day |
[!abc] | matches any one character that is not in the brackets | ls [!r]at | bat, cat, Bat, Cat, Rat | rat |
[!a-z] | matches any one character that is not in the range in the brackets | ls day[!1-9] | day0, days | day1 upto day9 |
5. Brace expansion
- it is generation of a range of strings to be used on the command line
- useful for generating a list of sequential data or running any command on sequence of data.
- a range or comma-separated list of data can be used for expansion. They are enclosed within curly braces.
- example:
- {0..4} generates a sequence from 0 to 4
- {alice, bob, charlie} generates alice, bob and charlie
- the list can be preceded by an optional string (called preamble or prefix). this is prefixed to each string in the braces
- the list can be followed by an optional string (called postscript or postfix). this is appended to each string in the braces
- syntax –
- string list
{String1, String2}
- range list
{<start>..<end>}
- range list with custom increment
{<start>..<end>..<increment>}
<prefix>{ string or range }<postfix>
- string list
5.1. Examples
- create files for each day of month January
touch {1..31}
- print all even numbers from 1 to 100
echo {2..100..2}
- create input, output, backup and error folders
mkdir /some/path/{input, output, backup, error}
- print the alphabets in reverse order
echo {z..a}
- create a backup of a file
cp file.txt{,.bkp}
- restore from .bkp file
mv file.txt{.bkp,}
- create folder for each day of January, February and March
mkdir -p ~/logs/{January/{1..31},February/{1..28},March/{1..31}}
6. Job control
- command line provides the ability to stop/suspend the execution of a process and resume a suspended process at a later point in time
- each running program is called job
- unique id is assigned to every job
command | description |
---|---|
jobs | list all the jobs that the current shell is running or has suspended |
fg | bring job to foreground |
bg | send job to background |
kill | terminate job |
stop | suspend job |
Ctrl+c | terminate job |
Ctrl+z | suspend job |
- to start a job in the background, we need to add an ampersand(&) at the end of the command
Examples:
- start a job in the background
sleep 1000 &
- bring the job to the foreground
fg %2 # where 2 is the job number
- suspend a job
stop %2 # where 2 is the job number
- resume a job in the background
bg %2 # where 2 is the job number
- terminate a job
kill %2 # where 2 is the job number
7. Process handling
- Get a snapshot of processes running with
ps
command - Get the process id of a command by
ps -ef | grep command
- wait for a process to finish by using
bash wait <process id>
- kill a process by using
kill <process id>
- wait for completion of all child processes
wait
- wait for completion of specific process
wait 1234 # where 1234 is the process id
7.1. difference between ps
and jobs
jobs
tells the list of jobs the current shell is managingps
tells the list of all the processes running in the system
8. Quoting
- Quoting is used to remove special meanings from characters or words
- single quotes – when the single quotes are used, every character within the quotes is preserved and is not evaluated
- double quotes – when the double qoutes are used, the dollar sign, back quotes and blackslashes are evaluated and interpreted.
- escape character – \ is used to preserve the literal value of the following character.
- ANSI C quoting – backslash escaped characters are treated according to ANSI C standard. Format:-
$'string
Examples:
examples | command | result |
---|---|---|
no quote | echo $HOME | /home/user1/ |
escape character | echo \$HOME | $HOME |
single quote | echo '$HOME' | $HOME |
double quote | echo "$HOME" | /home/user1/ |
ANSI C quoting | echo $'There\'s a quote in my string!' | There’s a quote in my string |
9. Basic file management
9.1. list directory ls
command | description |
---|---|
ls | list the contents of current directory |
ls * | list contents of the directory along with the subdirectory |
ls -l | list the contents of a directory along with the owner, permission, date, size |
ls -a | list hidden file |
ls -t | list files in descending order of last modified date |
ls -rt | list files in ascending order of last modified date |
ls -R | list files of the current directory and the subdirectory recursively to the last subdirectory |
ls /path/to/dirextory | list files in the directory mentioned |
9.2. show file contents
command | description |
---|---|
cat demo.txt | show the contents of file. use for relatively small files |
head demo.txt | show the first part of the file |
tail demo.txt | show the last part of the file |
tail -f demo.txt | show the text appended to the file as the file grows |
less demo.txt | show contents of file one screen at a time |
less -p "regular" demo.txt | show contents of the file from the first line with which the pattern matches |
strings -a binaryfile | print the sequence of all the printable characters in the file |
diff file1 file2 | shows difference between 2 files |
comm file1 file2 | compare 2 sorted file |
9.3. file handling
command | description |
---|---|
cp file1 file2 | copy file1 to file2 |
cp file1 file2 directory1 | copy file1 and file2 to directory1 |
cp -R directory1 directory2 | copy contents of directory1 to directory2 |
cp *.txt directory1 | copy all files ending with .txt to directory1 |
command | description |
---|---|
mv file1 file2 | rename file1 to file2 |
mv file1 directory1/ | move file1 to directory1 |
mv *.jpg directory1/ | move all files ending with .jpg to directory1 |
command | description |
---|---|
rm file1 | remove file1 |
rm file1 file2 | remove file1 and file2 |
rm *.png | remove all files ending with .png |
rm -d emptyDirectory | remove an empty directory |
rm -r directory1 | recursively remove all files, subdirectories and directory of directory1 |
9.3.4. file linking
- a symbolic link is a file that points to another file or directory
- there are 2 types of links
- Hard link -> it is an additional name for the existing file. Each file is associated with an unique number. This unique number is called inode. Hardlink associate 2 or more filenames to the same inode and in turn the same file. If the original file is removed, the contents are still available via the hardlink
- Soft link -> it is an indirect pointer to a file or directory. It has only the path of the original file and not its contents
command | description |
---|---|
ln file1 link1 | create a hardlink link1 to file file1 |
ln -s file1 link1 | create a softlink link1 to file file1 |
9.3.5. change directory
command | description |
---|---|
cd | Change to home directory |
cd ~ | Change to home directory |
cd - | Change to the previous directory |
cd .. | Change to the parent directory |
cd / | Change to the root directory |
cd ~/dir1/dir2 | Change to directory relative to home directory |
9.3.6. display disk usage
command | description |
---|---|
du | estimates and displays the disk space used by the files |
du -h * | prints size in human readable’ |
9.3.7. file ownership and permissions
command | description |
---|---|
chmod | change permissions |
chown | change ownership |
10. Special shell variables
- there are special variables that are set internally
- these variables are available to the user
variable | description |
---|---|
$0 | name of the shell |
$$ | the process id of the current shell |
$? | the exit status of the last executed command |
$! | the process id of the last background process |
$_ | the last argument of the previous command |
11. Exit codes
- also known as return code
- it is the code that is returned by command or process after it finishes
- following are few exit codes and their meaning
exit code | description |
---|---|
0 | Success |
1 | general error |
126 | command invoked does not have execute permissions |
127 | command not found |
130 | command terminated using Ctrl+C |
12. grep
- searches for pattern in files and prints each line that matches the input pattern
grep -<options> <pattern> <filenames>
- grep can be used to search in a single file or in multiple files
12.1. Useful grep
options
option | description |
---|---|
-i | ignore case |
-n | display line numbers along with lines |
-v | display lines that do not match the pattern |
-c | count the number of matching lines |
-r | search in all files under each directory |
-l | diplay the filename of the file which has the matching pattern |
-o | print only the matched string. The whole line with the matched string is not printed |
-I | ignore binary files |
-A<n> | include n lines after match |
-B<n> | include n lines before match |
-C<n> | include n lines before and after the match |
12.1.1. Examples
- lets say a demo file(demo.txt) has following content
- THIS IS UPPER CASE LINE
- this is lower case line
- This is regular line
- This line is also regular
- Line number four
- Line #5
- Search for a string in a file
grep "this" demo.txt
- Search for a string in a file, without matching case
grep -i "this" demo.txt
- Search for a string in a file and get both the line number and output
grep -n "this" demo.txt
- Get the number of lines matching the searched string
grep -c "this" demo.txt
- Get the filename in which the searched string is found
grep -l "this" demo.txt
- Get 2 lines after the matching line
grep -A2 "This" demo.txt
- Get 2 line before the matching line
grep -B2 "This" demo.txt
- Get 2 lines before and after the matching line
grep -C2 "This" demo.txt
- Search recursively in all files in subdirectories
grep -inrI 'some text' /path/to/dir
12.2. Regular expression in grep
- a regular expression is a sequence of characters that specifies the search pattern in text.
- different characters have special meaning in regular expressions
character | description |
---|---|
[abc] | matches any one of the characters in the square brackets |
[a-d] | matches any one of the characters in the range specified in the square brackets |
^start | matches the pattern only if the pattern is at the start of the line |
end$ | matches the pattern only if the pattern is at the end of the line |
[^abc] | matches any one character that is NOT present in the square brackets |
[^a-d] | matches any one character that is NOT present in the range |
. | matches any one character |
* | mathces 0 or more occurences of the preceding character |
.* | matches zero or more of any character |
12.3. Examples
- Match any one character
grep "[Tt]his" demo.txt
- Search for line starting with the search pattern
grep "^last" demo.txt
- Search for line ending with the search patter
grep "regular$" demo.txt
- Search for line with a character in specified range
grep "[0-9]" demo.txt
- Search for line without a character in specified range
grep "[^0-9]" demo.txt
- Search for a line where the middle characters are not known
grep "line.*regular" demo.txt
Note: grep can have regular expressions in the search pattern part, and can have wildcards in the files to search section.
- last line
13. find
- the
find
command is used to search and locate the list of files and directories - the list returned will satisfy the conditions used in find command.
- syntax is
find [starting point] [expression]
-exec [command]
can be used to run command on the files located byfind
- by default all the results are newline terminated. -print0 option can be used to make the results null terminated
13.1. Examples
- find files with specific name
find . -name demo.txt
- find files with a specific pattern
find ./Codes -name *.cpp
- find directories with specific name
find . -name Codes -type d
- find files with permission as 777 and change permissions to 644
find . -type f -perm 0777 -print -exec chmod 644 {} \;
- find and remove files
find . -type f -name "*.bkp" -exec rm -f {};
14. sed
filter and transform text
14.1. Overview
- looks for pattern and edits them
- works on both files and stdin
- original files is not updated
- results are put into standard output
- syntax is
sed 'instructions' file
- instruction is of format
'[address]command/regex/replace/modifier'
- example – ‘5,15s/abc/ABC/g’ -> this will substitute abc with ABC in lines 5 to 15.
- modifier g specifies that all occurences of abc in line will be replaced with ABC
- example – ‘5,15s/abc/ABC/g’ -> this will substitute abc with ABC in lines 5 to 15.
- when address is used, the lines belonging to the address are examined/modified. ->
sed '1,100 s/A/a/' file.txt
. this restricts substition to first 100 lines. - By default all the lines are printed. When ‘-n’ flag is used, this behavior is suppressed. Nothing will be printed unless an explicit request to print is found.
- A pattern can be used as a address ->
sed -n '/start/,/stop/ p' file.txt
- When ‘!’ is used, the command is run outside of the address ->
sed -n '/match/ !p' file.txt
command | description |
---|---|
d | delete |
p | |
s | substitute |
q | quit |
a | insert a line after pattern |
i | insert a line before pattern |
c | change a line |
y | transform |
14.2. Examples
- remove blank lines
sed '/^$/d' file.txt
- remove all lines with search string
sed '/Search/d' file.txt
- remove all instances of search string
sed 's/Search//g' file.txt
- print lines with the search string
sed '/Search/ p' file.txt
- substitue a string with another
sed 's/oldString/newString/g' file.txt
- remove trailing spaces
sed 's/ *$//' file.txt
- remove leading spaces
sed `s/^ *//' file.txt
- add spaces to start of everyline
sed 's/^/ /' file.txt
- print the first 10 line
sed '10 q' file.txt
sed -n '1,10 p' file.txt
- append line after a pattern
sed '/pattern/ a add line here' file.txt
- insert line before a pattern
sed '/pattern/ i add line here' file.txt
- change a line with a pattern
sed '/pattern/ c line changed here' file.txt
- change a->p, b->q, c->r
sed 'y/abc/pqr/' file.txt
14.3. Grouping
- sed allows capturing specific parts of text into groups.
- these groups can be manipulated
- group is enclosed within parentheses expression “(” and “)” in the search string
- each group is assigned a number. The first group is assigned \1 and so on.
- \1 can be both in pattern string and replacement string
14.3.1. Grouping Examples:
- switch first and second columns
sed 's/\([a-z]*\) \([a-z]*\)/\2 \1/' file.txt
- print lines which have consecutive duplicate words
sed -n '/\([a-z][a-z]*\) \1/p' file.txt
- remove consecutive duplicate words in a line
sed 's/\([a-z][a-z]*\) \1/\1/' file.txt
14.4. Hold Buffer
- When sed read text, each line is placed into a temporary space.
- When a new line is read, the old text is replaced by the new line in the temporary space.
- This temporary space is called pattern space.
- Hold buffer is like a long term storage. text can be copied to and from pattern space.
command | description |
---|---|
x | exchange hold space and pattern space |
h | copy pattern buffer into hold space |
H | append pattern buffer into hold space |
g | copy hold space to pattern space |
G | append hold buffer into pattern buffer |
14.4.1. Example
- print one line after and before the pattern match
sed -n '/999/ !{x;d};/999/ {x;p;x;p;n;p}' file.txt
- add space after every line
sed 'G' file.txt
- insert blank line above every line which matches pattern
sed '/start/ {x;p;x}' file.txt
- Insert blank line after every line which matches pattern
sed '/start/ {G}' file.txt
- Insert blank line before and after every line which matches pattern
sed '/start/ {x;p;x;G}' file.txt
15. awk
- command line utility to find, process and transform text files
- the basic syntax is
pattern { action }
- the pattern is compared with every input line. pattern can be any regular expression
- when the pattern matches, the action is performed
- when no pattern is provided, the action is applied to every line
- example ->
/^HTTP/ {print}
- every line starting with HTTP is printed
- there are 2 other important patterns
- BEGIN – specifies actions to be performed, before any line is read
- END – specifies actions to be performed, after all lines are read
- Example ->
awk 'BEGIN{print "start"} {print} END{print "end"}' file.txt
- first start is printed, then all lines of file.txt are printed and then end is printed
- awk interprets each line as a record of fields
- one or more consecutive spaces or tabs are considered as a single delimiter between fields
- $1, $2, etc. represents the first field, second field, and so on.
- $0 represents the entire input line
- awk has 2 data types – strings and integers
- awk internally converts variable according to context
- awk supports associative arrays. example
var[key] = value
- on integers, basic arithmatic operations (+-*/%) are supported. autoincrement(++) and decrement(–) is also supported.
15.1. Actions
- following are few actions that can be performed
action | description |
---|---|
{ print $0; } | print records |
{ exit; } | ends program |
{ next; } | skips current line |
{a=$1; b=”X”} | variable assignment |
{ c[$1] = $2 } | array varaible assignment |
{if (condition) { action } else if (condition) { action } else { action }} | if else conditions |
{ for (i=1; i < x; i++) { action } } | for loop |
{ for (item in c) { action } } | for loop iterating over a list |
15.2. Special variables
variable | desc |
---|---|
FS | Input field separator. can be modified |
RS | Input record separator. default value is newline. can be modified by user |
OFS | Output field separator. can be modified |
ORS | Output record separator. default value is newline. can be modified by user |
NF | Number of fields in the current line (record). cannot be updated by user |
NR | Number of lines processed so far. cannot be updated by user |
Note: -F option can be used to update the input field separator -> awk -F":"'{ print $1 }' file.txt
15.3. Examples
- split up “,” (comma) separated fields and print the third field ($3)
awk -F"," '{print $2}' file.txt
- print the 3rd field of a csv if the second field ($2) exists and is not empty
awk -F"," '{if ($2)print $3}' file.txt
- print the last field in each line
awk -F"," '{ print $NF }' file.txt
- print the line after the line matching search pattern
awk '/pattern/ { i=1; next; } {if(i) {i--; print;}}' file.txt
- print the line and the 2 lines after the line matching search pattern
awk '/regexp/ {i=3;} { if(i) {i--; print;}}' file.txt
- print the lines from a file starting at the line matching start until the line matching stop
awk '/start/,/stop/' file.txt
- count lines (wc -l)
awk 'END{print NR}' file.txt
- print matching lines (grep)
awk '/pattern/'
- print non matching lines (grep -v)
awk '!/pattern/'
- remove duplicate consecutive lines (uniq)
awk 'a !~ $0 {print}; {a=$0}' file.txt
- print first 10 lines of file (head)
awk 'NR < 11' file.txt
- print last 10 lines of file (tail)
awk '{vect[NR]=$0;} END{for(i=NR-9;i<=NR;i++) {print vect[i];}}' file.txt
- print the total number of bytes used by files
ls -l | awk '{ x += $5 } END { print "Total bytes: " x }'
- read a csv(comma separated) file and print the first and third field in semicolon separated format
awk 'BEGIN{FS=",";OFS=";"}{print $1, $3}' file.txt
16. Arithmetic expansion
- arithmetic expression can be evaluated and the result can then be used.
- syntax –
$(( expression ))
16.1. Example
- print sum of 2 numbers
echo $((1+3))
17. Command substitution
- In command substitution, the output of a command replaces the command
- the output of a command can be used an argument to another command
- syntax is
`command`
and$(command)
- using backticks (`) is discouraged and has been deprecated
$(commmand)
supports nesting i.e.$(command1 $(command2))
17.1. Examples
- assign the output of a command to a variable
thedate=`date`
- use the output of command as a parameter of another command
vi $(grep -l 123 *)
- assign contents of file to a variable
filedata=$(cat file.txt)
filedata=$(<file.txt)
second option is faster
18. Process substitution
- the input or output of a command can appear as a file. This is known as process substitution
- this technique is useful when we want to use the output of multiple commands as the input to a command
- process substitution can also be used to capture output and redirect it to the input of a process
- template –
<(command)
and>(command)
18.1. Examples:
- sort and compare two files
diff <(sort file1) <(sort file2)
- compare 2 folders
diff <(ls $first_directory) <(ls $second_directory)
19. Subshell
- a subshell is a child process launched by a shell
- whenever a shell script is run, a subshell is created and the script is run in the subshell
- variables defined in parent shell can be accessed if
export
is used while defining the variable - subshells can also be created using parentheses
(command1; command2; command3)
- subshells are a convenient way to group commands.
- can be used to temporarily move to a different directory
#do something in current directory (cd some/other/directory; other-command) #back in the original directory
- to run a command or script in the current shell, without creating a subshell, use ‘.’ as in
. script.sh
20. Command grouping
- sometimes we need to run multiple commands and redirect all the output to a single file. Command grouping helps in this.
- without command grouping we need to redirect each command output individually to a file
20.1. using parentheses
- Grouping list of commands can be done using parentheses ()
- a subshell is created
- example –
(date; uptime) > file.txt
20.2. using curly braces
- Grouping commands can be done using curly braces {}
- causes the list to be executed in the current shell context
- no subshell is created
- example –
{date; uptime;} > file.txt
21. Text editing with cut
, paste
and join
21.1. cut
cut
command cuts out sections from each line and writes result to standard output- syntax is
cut OPTION [FILE]
option | description |
---|---|
-c | character range which will be selected |
-d | delimiter which will separate each fields in line in file |
-f | fields which will be printed |
when -c option is used, ranges can be specified. Each range can be one of:
range type | description |
---|---|
N | Nth character |
N- | fron Nth character to end of line |
N-M | from Nth character to Mth character |
-M | from first to Mth character |
21.1.1. Examples
- print the first and third columns of a csv file
cut -f1,3 -d"," file.txt
- print the first 3 characters of each line
cut -c -3 file.txt
21.2. paste
- merges lines of files
- by default, the lines from each files are delimited by tab
- when ‘-‘ is used instead of filename, the command reads from standard input
- syntax
paste [OPTION] [FILE] [FILE]
option | description |
---|---|
-d | used to specify the delimiter |
-s | paste one file at a time |
21.2.1. Examples
lets take 2 files – number.txt and name.txt
- cat number.txt 1
- 2
- 3
- 4
- cat name.txt Alice
- Bob
- Charlie
- David
- merge 2 files, first file will give the first column and second file will give the second column
paste number.txt name.txt
- merge 2 files, delimited by ‘,’
paste -d"," number.txt name.txt
- merge 2 files, sequentially, i.e first only first file is printed and then only the second file
paste -s number.txt name.txt
21.3. join
- join lines of two files on a common field
- syntax
join [OPTIONS] FILE1 FILE2
21.3.1. Examples
lets take 2 files – number.txt and name.txt
- cat number.txt 1 100
- 2 101
- 3 102
- 4 103
- 5 104
- cat name.txt 1 Alice
- 2 Bob
- 3 Charlie
- 4 David
join 2 files based on the first columnjoin number.txt name.txt
22. Aliases
- aliases are short names for long commands
- when we need to execute long commands multiple times, it is advisable to create aliases
- syntax –
alias [-p] [name[=value]]
- we can create permanent alias by storing them in configuration files
- to temporarily bypass an alias, use \ ->
\ll
- creating a alias
alias name='values'
- removing alias
unalias name
- print all defined alias
alias -p
22.1. useful aliases
alias gh='history|grep' alias c=clear alias cx='chmod +x' alias ..='cd ..' alias sl=ls alias left='ls -t -1' alias count='find . -type f | wc -l' alias f='find . |grep '
23. Functions
- set of commands that accomplish a specific task
- can be used numerous times
- helps avoid writing the same code repeatedly
- can use loops and conditions within them
- arguments can be passed to functions
- by using
export -f functionname
, we can make the functions available to shell scripts- export can be added to configuration files like .bashrc
- To keep things modular, create a new file called ~/.bash_functions and then have your .bashrc load it
- syntax
function_name () { commands }
function_name () { commands; }
23.1. useful functions
mcd() { mkdir -p "$1"; cd "$1";} cdl() { cd "$1"; ls;}
24. sort
- sort lines text files
option | description |
---|---|
-r | sort in reverse order |
-n | sort in numberical order |
-k | sort based on nth column |
-u | sort and remove duplicates |
25. uniq
- report or omit repeated lines
- the input file must be sorted
option | description |
---|---|
-c | show how many times a line is repeated |
-d | prints only the repeated lines only once |
-u | prints only the unique lines |
-i | case insensitive comparison |
26. Conditions
26.1. If else
- if-then-else is supported in command line
- Syntax
if [ condition1 ]; then command1; elif [ condition2 ]; then command2; else command3; fi
if [ condition1 ]; then command1; elif [ condition2 ]; then command2; else command3; fi
- there are different types of conditions
- file based condiitons
conditions | description |
---|---|
-e | check if file exists |
-r | check if file exists and is readable |
-w | check if file exists and is writable |
-d | check if file exists and is a directory |
- string based conditions
conditions | description |
---|---|
== | check if both the strings are equal |
!= | check if both the strings are not equal |
> | check if the first string is lexicographically greater than the second |
< | check if the first string is lexicographically smaller than the second |
-n | check if the string has a length more than 0 |
-z | check if the string is an empty string |
- number based conditions
conditions | description |
---|---|
-eq | check if the numbers are equal |
-ne | check if the numbers are not equal |
-gt | check if the first number is greater than the second |
-ge | check if the first number is greater than or equal to the second |
-lt | check if the first number is less than the second |
-le | check if the first number is less than or equal to the second |
- 0 is considered true and numbers greater than 0 are considered false.
- this is because in Unix/Linix, when a process ends successfuly, it returns 0
26.2. Short circuiting
- an alternative way of using conditions is by using logical AND (&&) and logical OR(||)
- evaluation of a logical expression is stopped, as soon as the outcome has been determined. This is known as short-circuiting.
- in case of Logical AND, as soon as sub-expression becomes false, the whole expression evaluates to false
- in case of expr1 && expr2, if expr1 evaluates to false, then the whole expression will evaluate to false. So, expr2 is not evaluated at all.
- && can be used to ensure that command2 is run only if command1 ends successfully. example ->
command1 && command2
- in case of Logical OR, as soon as sub-expression becomes true, the whole expression evaluates to true
- in case of expr1 || expr2, if expr1 evaluates to true, then the whole expression will evaluate to true. So, expr2 is not evaluated at all.
- || can be used to ensure that command2 is run only if command1 fails. example ->
command1 || command2
26.2.1. example
- create folder if it does not exist
[ -d ./some/path/folder ] || mkdir /some/path/folder
- create create file only if folder exists
cd /some/path/folder && touch file.txt
27. Loops
27.1. while loops
- the loop runs as long as the given condition is true
- syntax while [ condition ]; do commands; done
27.1.1. example
- print all the folders with .c files
find . -name *.c | {while read -r filename; do dirname $filename; done;} | sort | uniq # dirname returns the directory name
27.2. for loops
- the loop iterates over a list of values or preset number of times
- syntax for <variable name> in <a list of items>;do <some command> $<variable name>;done;
27.2.1. example
- copy files from one folder to another
for file in ./code/*.txt; do cp $file /home/code/backup; done
28. ssh
- ssh (SSH client) is a program for logging into a remote machine and for executing commands on a remote machine
- syntax
ssh user@host
- running a single command on remote server
ssh user@host command_to_run
- logging into server with different port
ssh -p portnum user@host
- ssh connection using host in the middle
ssh -t reachable_host ssh unreachable_host
29. curl
- used to transfer data from one server to another
- syntax –
curl [options] [URL]
options | description |
---|---|
-o | save the output in a file with name as specified after the option |
-O | save the output in a file with the same name as in the url |
-C – | resume a download |
-I | fetch headers of a url |
-L | follow redirects |
29.1. Examples
- retrieve a webpage
curl example.com
- save a webpage
curl example.com -o example.html
- resume a download
curl -C - -O https://releases.ubuntu.com/21.10/ubuntu-21.10-desktop-amd64.iso
- fetch headers only
curl -I example.com
- fetch weather
curl wttr.in/london
29.2. handling REST calls
29.2.1. GET
- GET is used to fetch resource
- the GET method is the default method
- example
curl https://reqres.in/api/users/2
29.2.2. POST
- POST is used to create resource in a server
- To send a curl POST request we use the option -X POST
- example
curl -X POST -H "Content-Type: application/json" -d '{"email": "eve.holt@reqres.in","password": "pistol"}' https://reqres.in/api/register
29.2.3. PUT
- PUT is used to update resource
- To send a curl PUT request we use the option -X PUT
- example
curl -X PUT -H "Content-Type: application/json" -d '{"name": "morpheus","job": "zion resident"}' https://reqres.in/api/users/2
29.2.4. DELETE
- DELETE is used to remove resource
- To send a curl DELETE request we use the option -X DELETE
- example
curl -X DELETE https://reqres.in/api/users/2
30. wget
- command line utility to download file
options | description |
---|---|
-O | download file under different name |
-c | resume a download |
30.1. Examples
- download a file
wget https://releases.ubuntu.com/21.10/ubuntu-21.10-desktop-amd64.iso
- resume a download
wget -c https://releases.ubuntu.com/21.10/ubuntu-21.10-desktop-amd64.iso
31. One liners
- print the files and directories in tree structure
find . | sed -e "s/[^-][^\/]*\// |/g" -e "s/|\([^ ]\)/|-\1/"
- print the directories in tree structure
find . -type d | sed -e "s/[^-][^\/]*\// |/g" -e "s/|\([^ ]\)/| - \1/"
- quick access to ascii table
man ascii
- find common lines in 2 files
cat file1 file2 | uniq -d
- find lines in file1 that is not present in file2
cat file1 file2 file2 | uniq -u
- find most frequently used commands
history | cut -c8- | sort | uniq -c | sort -rn | head
- recursively remove only directories with no files
find . -depth -type d -exec rmdir {} \;
- count the number of total active network interfaces
wc -w <<<$(netstat -i | cut -d" " -f1 | egrep -v "^Kernel|Iface|lo")
- display all the files/directories sorted by size
du -ah . 2>/dev/null | sort -hr | less
- make multiple copies of file
tee filecopy{1..10}.txt < file.txt
32. Further reading
33. Change History
- presented in reversed chronological order i.e. the latest change is at the top
Name | Date | Change Description |
---|---|---|
Utsav Barman | 26 Jan 2022 | added: [arithmetic expansion, grep recursively search, command substitution example] |
Utsav Barman | 26 Jan 2022 | added: [ANSI C quoting] |
Utsav Barman | 26 Jan 2022 | added: [brace expansion, find piped to xargs, herestrings, awk example] |
Utsav Barman | 26 Jan 2022 | added: [command grouping] |
Utsav Barman | 25 Jan 2022 | added: [chown, chmod and du] |
Utsav Barman | 25 Jan 2022 | added: [& to start a background job, added curl and wget] |
Utsav Barman | 25 Jan 2022 | read should be used with option -r ; used appropriate variable names in for loop example; |
Utsav Barman | 25 Jan 2022 | Updated variable DATE to thedate. Uppercase variables are conventionally used for environment variables. Added $() for command substitution. backticks are discouraged; replaced -a with -e |
Utsav Barman | 25 Jan 2022 | fixed typo for cd to root directory; added section for special shell variables and exit codes; |
Utsav Barman | 25 Jan 2022 | Added wait, globbing, grep -o; fixed typos |
Utsav Barman | 24 Jan 2022 | Initial draft |
Leave a Reply