Can somebody experienced have a look at my bash script and advice how to make it simplier? - bash

Question

my task is to create a script that displays the frequency of random characters in the file. The output should display frequency of a to z (case insensitive) in percent.
I created the script below and I just wonder if there is a way how to make it simplier?
#!/bin/bash
echo Hello, please tell me in which file shall I count the letters:
read file
TOTAL=$( grep -o [[:alpha:]] $file | wc -l )
A=$( grep -io a $file | wc -l )
B=$( grep -io b $file | wc -l )
C=$( grep -io c $file | wc -l )
D=$( grep -io d $file | wc -l )
E=$( grep -io e $file | wc -l )
F=$( grep -io f $file | wc -l )
G=$( grep -io g $file | wc -l )
H=$( grep -io h $file | wc -l )
I=$( grep -io i $file | wc -l )
J=$( grep -io j $file | wc -l )
K=$( grep -io k $file | wc -l )
L=$( grep -io l $file | wc -l )
M=$( grep -io m $file | wc -l )
N=$( grep -io n $file | wc -l )
O=$( grep -io o $file | wc -l )
P=$( grep -io p $file | wc -l )
Q=$( grep -io q $file | wc -l )
R=$( grep -io R $file | wc -l )
S=$( grep -io s $file | wc -l )
T=$( grep -io t $file | wc -l )
U=$( grep -io u $file | wc -l )
V=$( grep -io v $file | wc -l )
W=$( grep -io w $file | wc -l )
X=$( grep -io x $file | wc -l )
Y=$( grep -io y $file | wc -l )
Z=$( grep -io z $file | wc -l )
echo Frequency of 'a': $(($A*100/$TOTAL))%
echo Frequency of 'b': $(($B*100/$TOTAL))%
echo Frequency of 'c': $(($C*100/$TOTAL))%
echo Frequency of 'd': $(($D*100/$TOTAL))%
echo Frequency of 'e': $(($E*100/$TOTAL))%
echo Frequency of 'f': $(($F*100/$TOTAL))%
echo Frequency of 'g': $(($G*100/$TOTAL))%
echo Frequency of 'h': $(($H*100/$TOTAL))%
echo Frequency of 'i': $(($I*100/$TOTAL))%
echo Frequency of 'j': $(($J*100/$TOTAL))%
echo Frequency of 'k': $(($K*100/$TOTAL))%
echo Frequency of 'l': $(($L*100/$TOTAL))%
echo Frequency of 'm': $(($M*100/$TOTAL))%
echo Frequency of 'n': $(($N*100/$TOTAL))%
echo Frequency of 'o': $(($O*100/$TOTAL))%
echo Frequency of 'p': $(($P*100/$TOTAL))%
echo Frequency of 'q': $(($Q*100/$TOTAL))%
echo Frequency of 'r': $(($R*100/$TOTAL))%
echo Frequency of 's': $(($S*100/$TOTAL))%
echo Frequency of 't': $(($T*100/$TOTAL))%
echo Frequency of 'u': $(($U*100/$TOTAL))%
echo Frequency of 'v': $(($V*100/$TOTAL))%
echo Frequency of 'w': $(($W*100/$TOTAL))%
echo Frequency of 'x': $(($X*100/$TOTAL))%
echo Frequency of 'y': $(($Y*100/$TOTAL))%
echo Frequency of 'z': $(($Z*100/$TOTAL))%
I considered using for loop as in below script which replace the first part of above script...but then I got stuck as I do not know if there is any way to work with those outputs further?
#!/bin/bash
echo File:
read file
TOTAL=$( grep -o [[:alpha:]] $file | wc -l )
for letter in {a..z}
do echo grep -io $letter $file | wc -l
done
I also want to ask if there is any way how to have output of my script with two decimal places?
This is my first script so please be merciful :) But I will be grateful for any feedback or advice how to get better.

Answer 1

You were almost there! Here's a solution with 2 variants, depending on the output you want and if you want to use bc.
#!/bin/bash
echo File:
read file
TOTAL=$( grep -o "[[:alpha:]]" "$file" | wc -l )
for letter in {a..z}
do
count=$(grep -io $letter "$file" | wc -l)
echo "Frequency of $letter : $(bc <<< "scale=2; $count*100/$TOTAL")%" # Variant with floats, requires bc
echo "Frequency of $letter : $(($count*100/$TOTAL))%" # Variant with integers
done

Answer 2

You can use the awk command inside your bash script
awk -vFS="" 'BEGIN{OFMT="%.2f"} {for(i=1;i<=NF;i++){ if($i~/[a-zA-Z]/) { w[tolower($i)]++} sum++} }END{for(i in w) print i,(100*w[i]/sum),"%"}'