Linux Command Quick Reference

find

find is inherently recursive and is intended to descent through directories looking for files with certain attributes or executing commands. cf) -type f --> plain file; d --> directory p --> fifo or named pipe 1.) Find all files modified between 50 ~ 100 days


	find . -mtime +50 -mtime -100

2.) Find all files permission 777


	find . -type f -perm 0777

3.) Find all read-only file


	find . -perm /u=r

4.) Find all excutable file


	find . -perm /a=x

5.) Find all empty file


	find . -type f -empty

6.) Find all files basedon user


	find . -user "username"

7.) Find all c files then change mode

find . -name "*.c" -exec chmod g+w {} \;

grep

grep [options] pattern [files] -c ::count. With -v, count nonmatching lines -d read ::read directories like ordinary files (default) skip ::skip directories recurse:: recursively ... same as -r examples: 1.)

$ cat a
this is line 1 UNIX UNIX
this is line 2 unix
this is line 3 Unix Unix
this is line 4 hello

2.)

$ grep unix a
this is line 2 unix

$ grep -c -a unix a
1

3.)Ignore upper and lower case

$ grep -i unix a
this is line 1 UNIX UNIX
this is line 2 unix
this is line 3 Unix Unix

4.)Add line number

$ grep -i -n unix a
1:this is line 1 UNIX UNIX
2:this is line 2 unix
3:this is line 3 Unix Unix

$ grep -i -n hello a
4:this is line 4 hello

5.)Print only patter parts

$ grep -o unix a
unix

$ grep -oi unix a
UNIX
UNIX
unix
Unix
Unix

6.)Print lines not containing patterns

$ grep -vin unix a
4:this is line 4 hello

7.)Search file for multiple patterns or for pattern beginning with hyphen

grep -n -e unix -e hello a
2:this is line 2 unix
4:this is line 4 hello

8.)Multiple files

$ cat a
this is line 1 UNIX UNIX
this is line 2 unix
this is line 3 Unix Unix
this is line 4 hello

$ cat b
this is line 1 hello
this is line 2 Unix
this is line 3 UNIX UNIX
this is line 4 unix unix

$ grep -in unix a b
a:1:this is line 1 UNIX UNIX
a:2:this is line 2 unix
a:3:this is line 3 Unix Unix
b:2:this is line 2 Unix
b:3:this is line 3 UNIX UNIX
b:4:this is line 4 unix unix

$ find .
.
./a
./b
$ grep -r unix .
./a:this is line 2 unix
./b:this is line 4 unix unix

Print 3 lines before, 2 lines after term "unix".

$ grep -B 3 -A 2 unix .

## Serching pattern

What it does	Pattern
The string bag	bag
"bag" at the begining of line or string.	^bag
"bag" at the end of line or string.	bag$
"bag" as the only text on line	^bag$
"Bag" or "bag"	[Bb]ag
Second character is vowel	b[aeiou]g
Second character is NOT vowel	b[^aeiou]g
Second character is any character except new line	b.g
Any line containing exactly three characters.	^...$
Any line that begins with a dot	^\.
Any line that doesn't begin with a dot	^[^.]
"bug", "bugs", "bugss", etc	bugs*
A word in quotes	"word"
A word, with or without quotes	"word"
One or more uppercase letters	[A-Z][A-Z]*
An uppercase letter, followed by zero or more characters	[A-Z].*
Zero or more uppercase letters.	[A-Z]*
Any letter.	[a-zA-Z]
Any alphanumeric sequence	[0-9A-Za-z]+

Showing large files

1.) Using head and tail command default : read 10 lines

$ cat a
this is line 1 UNIX UNIX
this is line 2 unix
this is line 3 Unix Unix
this is line 4 hello

$ head -n 2 a
this is line 1 UNIX UNIX
this is line 2 unix

$ head -n +2 a
this is line 1 UNIX UNIX
this is line 2 unix

read last 2 lines

$ tail -n 2 a
this is line 3 Unix Unix
this is line 4 hello

read from 2nd line to last

$ tail -n +2 a
this is line 2 unix
this is line 3 Unix Unix
this is line 4 hello

read 3rd line only

$ tail -n +3 a | head -n 1 
this is line 3 Unix Unix

2.)Using "cut"
[options]
+ -d --> delimiter, special character should be quoted " ". with -f1 --> 1st column
+ -c --> print characters in list columns

$ cat a
this is line 1 UNIX UNIX
this is line 2 unix
this is line 3 Unix Unix
this is line 4 hello

$ cut -c 1 a
t
t
t
t

$ cut -c 1,4 a
ts
ts
ts
ts

$ cut -c 1-4 a
this
this
this
this

$ cut -d" " -f1 a
this
this
this
this

3.)Using sed
Read 2nd and 4th line

$ sed -n -e 2p -e 4p a
this is line 2 unix
this is line 4 hello

Read 2nd to 4th line

$ sed -n 2,4p a
this is line 2 unix
this is line 3 Unix Unix
this is line 4 hello

sed

Syntax

sed [options] 'command' file(s) sed [options] -f scriptfile file(s) From lynda sed 's/a/d'

s: substitution

a: search string

b: replacement string

ex)

$ echo 'upstarem' | sed 's/up/down/'
downstarem

$ cat a | sed 's/is/are/'
thare is line 1 UNIX UNIX
thare is line 2 unix
thare is line 3 Unix Unix
thare is line 4 hello

That is, it is not global default. 'g' should be added. However, remember that each line is a stream. So, 'is' in each stream (line) is changed not second one.

$ cat a | sed 's/is/are/g'
thare are line 1 UNIX UNIX
thare are line 2 unix
thare are line 3 Unix Unix
thare are line 4 hello

$ cat a | sed -e 's/this/that/g' -e 's/is/are/g' -e 's/line/line number/g'
that are line number 1 UNIX UNIX
that are line number 2 unix
that are line number 3 Unix Unix
that are line number 4 hello

options

-n Suppress the default output
-e editing command are followed.

commands

[address [,address]][!]command[arguments]

a\ Append text after a line
c\ Replace text(usually text block)
i\ Insert text before a line.
d Delete lines.
s Make substitutions
y Translate characters

Chris IJ Hwang

Contents

Home

Data Science/Machine Learning related

Quantitative Finance Modeling and Analysis

Visualization

Math Finance

Others