echo "Bash" > search & find

What You will Learn

  • How to find a folder or a file
  • Search for a pattern in a string or text
  • manipulate a text using the command line

Find A File or A Folder

find MyFolder

find . -name "myFile.txt"


  1. find
  2. a file path
  3. an expression

Find A File or A Folder

* - any number of characters

? - one character

[] - any of the characters inside the brackets

find . -name "*.html" # anything that ends with .html
find . -name "*.???" # anything that ends with a three letter file extension like .txt or .css
find . -name "[fts]*" # anything that starts with the letter f t or s
find . -name "*main*" # anything that has the text main somewhere in the filename

Find A File or A Folder

https://www.digitalocean.com/community/tutorials/how-to-use-find-and-locate-to-search-for-files-on-linux

sudo apt install mlocate

Search inside a text

echo "Lisa
Mark
Elie
Beth
Tim
Elizabeth
Tom
Matt
Liza
Janey
Jane
Shana" >> people.txt

cat people.txt | grep Tim

Search inside a text

grep -i "elie" people.txt # case insensitive 

grep -iw "beth" people.txt # case insensitive + full world

grep -v "Beth" people.txt # everything but

grep -c "Jane" people.txt # count

grep -ni "Jane" people.txt # print with line numbers

Search inside a text

grep -wc "...." people.txt

grep -wc "T.*" people.txt

grep -wc "[LME].*" people.txt

grep -wc "[^T].*" people.txt

The real power of grep shines when using regular expressions

PRACTICE

echo "Lisa
Mark
Elie
Beth
Tim
Elizabeth
Tom
Matt
Liza
Janey
Jane
Shana" >> people.txt
  1. I want all people with names starting by T
  2. All people with  names of  5 or more characters
  3. All people not having a e in their name

PRACTICE - solution

grep "T" people.txt

grep "....." people.txt

grep -v "e" people.txt

# Thanks @delhaj ❤️

Regular Expression

https://regexone.com/

cut

echo "Issam,Pokemon, 5
Marta,Extreme sports, 2
Dhirar,SublimeText, 1
JC,Stargate, 7
Wenting,Disney world, 6
Shuting,Bag Raiders, 3
Stephanie,Wolves, 4
Sebastien,Naruto, 8
Trung,CSGO, 10
Rahul,GOT, 9" > best_people_ever.txt

cat best_people_ever.txt | cut -d ',' -f 1 # Issam, Marta...

cat best_people_ever.txt | cut -d ',' -f 2 # Pokemon, Extreme sports...

Used to cut lines based on a delimiter or a number of characters

sort

echo "Issam,Pokemon, 5
Marta,Extreme sports, 2
Dhirar,SublimeText, 1
JC,Stargate, 7
Wenting,Disney world, 6
Shuting,Bag Raiders, 3
Stephanie,Wolves, 4
Sebastien,Naruto, 8
Trung,CSGO, 10
Rahul,GOT, 9" > best_people_ever.txt

cat best_people_ever.txt | sort -t ',' -k 1 # sort names

cat best_people_ever.txt | sort -t ',' -k 3 # sort by number ⚠️

Used to sort lines based

head & tail

echo "Issam,Pokemon, 5
Marta,Extreme sports, 2
Dhirar,SublimeText, 1
JC,Stargate, 7
Wenting,Disney world, 6
Shuting,Bag Raiders, 3
Stephanie,Wolves, 4
Sebastien,Naruto, 8
Trung,CSGO, 10
Rahul,GOT, 9" > best_people_ever.txt

# are these two lines below equivalent ?

cat best_people_ever.txt | sort -t ',' -k 3 -g | head -n 2

cat best_people_ever.txt | sort -t ',' -k 3 -g -r | tail -n 2

head -> output first lines of a file

tail -> output last lines of a file

Practice

echo "Issam,Pokemon, 5
Marta,Extreme sports, 2
Dhirar,SublimeText, 1
JC,Stargate, 7
Wenting,Disney world, 6
Shuting,Bag Raiders, 3
Stephanie,Wolves, 4
Sebastien,Naruto, 8
Trung,CSGO, 10
Rahul,GOT, 9" > best_people_ever.txt
  1. What music group Shuting listens to every single day ?
  2. I want all people's activity whose name length is higher than 6 sorted by their index (reversed)

SED

SED

Stream EDitor or Sed works as follows:

SED

Often used to search and replace or substitute, the basic form is the following

sed -e [line number or regular expression]s/regular-expression/replacement/[flags] text
echo "I have three dogs and two cats" > animals.txt
sed -e 's/dog/cat/g' -e 's/cat/elephant/g' animals.txt --debug
sed -e 's@dog@cat@g' -e 's@cat@elephant@g' animals.txt # equivalent

n    replace nth instance of pattern with replacement
g    replace all instances of pattern with replacement
 

SED

echo "line 1 (one)
line 2 (two)
line 3 (three)" > file.txt

sed -e '1,2d' file.txt

sed -e '3d' file.txt

sed -e '1,2s/line/LINE/' -e '/^line/'d file.txt

sed -e '1,2s/line/LINE/' -e '/^line/!'d file.txt


sed -e [line number or regular expression]d text

SED

echo "hello
this text is wiped out
Wiped out
hello (also wiped out)
WiPed out TOO!
goodbye
(1) This text is not deleted
(2) neither is this ... ( goodbye )
(3) neither is this
hello
but this is
and so is this
and unless we find another g**dbye
every line to the end of the file gets deleted" > text.txt

sed -e '/hello/,/goodbye/d' text.txt
sed -e [address1, address2]d text

SED

#!/bin/sh
X='word1\|word2\|word3\|word4\|word5'
sed -e "
/$X/!d
/$X/{
	s/\($X\).*/&/
	s/.*\($X\)/&/
	q
	}" $1

Sed can do much more with the use of subroutines

& refers to the first match (sometimes people use \1)

q forces the end of the evaluation

PRACTICE

echo "Pour jouer au tennis de tableen 2020, il faut:
1 table de tennis de table
2 raquettes
3 balles de ping pong
et de la bonne motivation !" > ping_pong.txt
-> 1 table de tennis de table !!!
-> 2 raquettes !!!
-> 3 balles de ping pong !!!

I want this !

sed -e '/^[0-9].*/!d' -e 's/^[0-9].*/-> & !!!/g' ping_pong.txt

awk

And you were thinking sed was powerful ?

awk

Awk is almost a programming language in itself. It can do any kind of text manipulation you can think of.

awk

echo "hey, how are you ?" | awk 'BEGIN{printf"---|Start|--\n"} {print} END{printf"---|End|---\n"}'
echo "hey, how are you ?
good and you" | awk 'BEGIN{printf"---|Start|--\n"} {sub(/you/ ,"we"); print} END{printf"---|End|---\n"}'

https://www.tutorialspoint.com/awk/index.htm

xargs

xargs command is used to build an execution pipeline from standard input.

echo 'one two three' | xargs mkdir
ls # one two three
find *.txt | xargs grep '1'

Practice

echo "Issam-Hammi-pizza
Rahul-Gakhar-vegetables
Wenting-Bao-steaks
Sebastien-Dubuc-budda bowl
Dhirar-Elhaj-tacos" > food.txt
  1. replace all - by : using sed
  2. return all first names and their favorite dishes separated by a space

Practice

echo "1>>>>2
2>>>>3
3>>>>4
4>>>>5" > numbers.txt
  1. Using cut print out just the numbers 2 3 4 5. Then use xargs to print them all on 1 line
  2. find all text files in your current directory and print all numbers

Bash - session 3

By Issam Hammi

Made with Slides.com

Bash - session 3

Third session focused on text manipulation

  • 540

More from Issam Hammi