My favorite material from Codio's Bash Week 2 on Coursera

Bash Scripting and System Configuration | Codio

Glob Expressions

List .png files beginning with file followed by something that’s NOT a lowercase letter, then a number.

ls file[![:lower:]][[:digit:]].png

Extended globbing

The + character allows us to match one or more occurrences of the given patterns.

ls file3c+(.txt|.png)

We can use a for loop to cycle through all of the filenames and display each filename after removing everything #* that comes before the first.

for filename in *; do echo ${filename#*.}; done | sort -u

grep

Locate lines with 1 or more occurrence of the word my.

grep -E 'my+' jabberwocky

Return lines containing exactly 1 or 2 digits.

grep -E '[[:digit:]]{1,2}' jabberwocky

Let’s use grep to search for lines that contain ll inside this file, remembering to include -E to enable extended regexes.

grep -E '(ll)' jabberwocky

Only return the portion of the line that matches this pattern

grep -E -o '(.u.+u.)' jabberwocky

sed

Let’s print all of the lines in the file greeneggs.txt that contain the pattern Sam using sed

sed -n -E '/Sam/p' greeneggs.txt

We can limit this to only match lines beginning with Sam by anchoring the pattern with a ^ at the beginning.

sed -n -E '/^Sam/p' greeneggs.txt

Let’s change every instance of the word green to the word old using substitution and redirect the result to a file called oldeggs.

sed -E 's/green/old/g' greeneggs.txt > oldeggs

Run the command below in the terminal to match the first three digits in each line and replace them with the same digits, wrapped in parenthesis ()

sed -E 's/[0-9]{3}/(&)/' data.csv

We can use backreferences to modify this approach and add a hyphen between the remaining three and four characters.

Let’s create three backreference groups by wrapping our patterns to match in parenthesis ().

\1: 3-digit Area Code: ([0-9]{3})

\2: 3-digit Exchange Code: ([0-9]{3})

\3: 4-digit Line Number: ([0-9]{4})

sed -E 's/([0-9]{3})([0-9]{3})([0-9]{4})/(\1) \2-\3/' data.csv

Inserts a # at the beginning of lines 3−10 inside the file

sed '3,10s/^/#/' flowers.txt

awk

Find each line in the file that contains the pattern Male, and print the 1st and 4th columns of data for each match.

awk -F"," '/Male/{print $1,$4}' people.csv

For example, we can check the column holding first names $2 and match all first names that begin and end with A with any number of any type of characters in between.

Once we locate these matches, we can print only the first name, last name, and gender columns for these matches.

awk -F"," '$2 ~ /^A.*a$/{print $2,$3,$5}' people.csv

Find all of the lines belonging to Discover card users, and display each card number to the terminal.

Discover, Chris Howe, 6543435725654171, 08/31, CVC: 255 American Express, Kathleen Ford, 341584729447996, 08/31, CID: 9885

awk -F"," '/Discover/{print $3}' credit.csv

#!/binbash

while read line
do

  if [[ $line =~ .*[Dd]iscover.* ]];
  then
    if [[ $line =~ [0-9]{16} ]];
    then
      echo ${BASH_REMATCH[0]}
    fi
  fi
done < credit.csv

Create a command that displays a complete list of valid email addresses in the file signups.txt

Between 1 and 16 numbers, uppercase or lowercase letters, periods, or underscores

An \@ sign

A domain with any number of lowercase letters, (like gmail, hotmail, etc.)

A period .

top level domain of .com, .org, or .net

grep -E '([[:alnum:]]|\.|\_)+(@)([[:alnum:]]|\.|\_)+(\.com|\.org|\.net)$' signups.txt | grep -E '([[:alnum:]]|\.|\_\@){1,16}'

#!/bin/bash

grep -E '(^[A-Za-z0-9_\.]{1,16})@[[:lower:]]+\.(com|org|net)' signups.txt