Krdytkyu

If I have a file called text.txt
and it contains one line of text, ijbiaubiu
is it possible to move each letter one space forward or back in the alphabet in order to encrypt it manually, with what I believe would be a ROT cipher?

How is it done?

I have tried sed and perl -pi -e but I don't know how to read the contents of the file before applying commands, and these utilities never did the trick.

To clarify, I am not looking for help encrypting my files.
Rather, with help learning how to use the CLI to shift about alphabetical characters in a file manually.

sed

can do that with the transform command y/…/…/, e.g. to replace “a” with “b” and so forth:

sed 'y/abcdefghijklmnopqrstuvwxyz/bcdefghijklmnopqrstuvwxyza/'

or with capitals:

sed 'y/abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/bcdefghijklmnopqrstuvwxyzaBCDEFGHIJKLMNOPQRSTUVWXYZA/'

To simplify this you can use bash Parameter expansion, to be exact substring expansion:

a=abcdefghijklmnopqrstuvwxyz

b=ABCDEFGHIJKLMNOPQRSTUVWXYZ

r=1

sed "y/$a$b/${a:$r}${a::$r}${b:$r}${b::$r}/"

To alter the rotation just set r to the desired rotation, e.g. r=13 for ROT13. It works with negative numbers as well. The same approach can be used for the below perl version.

To read from a file I recommend redirection (here’s why), e.g. <text.txt sed …, see below for an example. You can however also give the file as an argument to sed, e.g. sed '…' text.txt.

Example run

$ <<<"ijbiaubiu" sed 'y/abcdefghijklmnopqrstuvwxyz/bcdefghijklmnopqrstuvwxyza/'

jkcjbvcjv

$ <<<"ijbiaubiu" sed 'y/abcdefghijklmnopqrstuvwxyz/zabcdefghijklmnopqrstuvwxy/'

hiahztaht

$ echo "ijbiaubiu" >rot_file

$ <rot_file sed 'y/abcdefghijklmnopqrstuvwxyz/bcdefghijklmnopqrstuvwxyza/'

jkcjbvcjv

perl

has an easy-to-use transliterator tool called tr, e.g. to replace “a” with “b” and so forth:

perl -npe 'tr/abcdefghijklmnopqrstuvwxyz/bcdefghijklmnopqrstuvwxyza/'

-n is to iterate over filename arguments and -p to print lines afterwards, same as adding ;print to the expression would do.

Example run

$ <<<"ijbiaubiu" perl -npe 'tr/abcdefghijklmnopqrstuvwxyz/bcdefghijklmnopqrstuvwxyza/'

jkcjbvcjv

You can do it using tr command. A search for "ROT cypher Bash" leads me to caesar-cipher.sh.

• Note that the commands are case sensitive.

Caesar cipher encoding:

$ echo "THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG" | tr '[A-Z]' '[X-ZA-W]'

QEB NRFZH YOLTK CLU GRJMP LSBO QEB IXWV ALD

Caesar cipher decoding:

$ echo "QEB NRFZH YOLTK CLU GRJMP LSBO QEB IXWV ALD" | tr '[X-ZA-W]' '[A-Z]'

THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG

It can also be adjusted to ROT13 instead:

$ echo "THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG" | tr '[A-Z]' '[N-ZA-M]'

GUR DHVPX OEBJA SBK WHZCF BIRE GUR YNML QBT



$ echo "GUR DHVPX OEBJA SBK WHZCF BIRE GUR YNML QBT" | tr '[N-ZA-M]' '[A-Z]'

THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG

In your case it would be:

tr '[A-Z]' '[X-ZA-W]' < file

Also from another source to decode (reverse the order of patterns to encode):

ROT-3 = tr 'd-za-cD-ZA-C' 'a-zA-Z'

ROT-4 = tr 'e-za-dE-ZA-D' 'a-zA-Z'

ROT-5 = tr 'f-za-eF-ZA-E' 'a-zA-Z'

ROT-6 = tr 'g-za-fG-ZA-F' 'a-zA-Z'

ROT-7 = tr 'h-za-gH-ZA-G' 'a-zA-Z'

ROT-8 = tr 'i-za-hI-ZA-H' 'a-zA-Z'

ROT-9 = tr 'j-za-iJ-ZA-I' 'a-zA-Z'

ROT-10 = tr 'k-za-jK-ZA-J' 'a-zA-Z'

ROT-11 = tr 'l-za-kL-ZA-K' 'a-zA-Z'

ROT-12 = tr 'm-za-lM-ZA-L' 'a-zA-Z'

ROT-13 = tr 'n-za-mN-ZA-M' 'a-zA-Z'

ROT-14 = tr 'o-za-nO-ZA-N' 'a-zA-Z'

ROT-15 = tr 'p-za-oP-ZA-O' 'a-zA-Z'

ROT-16 = tr 'q-za-pQ-ZA-P' 'a-zA-Z'

ROT-17 = tr 'r-za-qR-ZA-Q' 'a-zA-Z'

ROT-18 = tr 's-za-rS-ZA-R' 'a-zA-Z'

ROT-19 = tr 't-za-sT-ZA-S' 'a-zA-Z'

ROT-20 = tr 'u-za-tU-ZA-T' 'a-zA-Z'

ROT-21 = tr 'v-za-uV-ZA-U' 'a-zA-Z'

ROT-22 = tr 'w-za-vW-ZA-V' 'a-zA-Z'

ROT-23 = tr 'x-za-wX-ZA-W' 'a-zA-Z'

ROT-24 = tr 'y-za-xY-ZA-X' 'a-zA-Z'

ROT-25 = tr 'z-za-yZ-ZA-Y' 'a-zA-Z'

$ cat text.txt 

ijbiaubiu

A 1-rotation with Python, after setting ROT = 1:

$ python -c "with open('text.txt') as f: ROT = 1; print(''.join([chr((ord(char)- ord('a') + ROT)%26+ord('a')) for char in f.read().strip()]))"

jkcjbvcjv

Or setting ROT = -1:

$ python -c "with open('text.txt') as f: ROT = -1; print(''.join([chr((ord(char)- ord('a') + ROT)%26+ord('a')) for char in f.read().strip()]))"

hiahztaht

What it's doing is:

1. reading letters in from the file,


2. getting their integer (or ordinal) values,


3. subtracting the ordinal value of lowercase a to get a number from 0 - 25,


4. adding the rotation value,


5. taking % 26 so after z you wrap around to a if necessary,


6. adding the ordinal value of lowercase a back and


7. converting the ordinal value to a string value.

If you know Python, it might be easier to read if it's not a 1-liner:

ROT = 1



def rotate_char(char, rotate_amount):

    return chr((ord(char) - ord('a') + rotate_amount)%26 + ord('a'))



with open('text.txt') as f:

    line = f.read().strip()

    print(''.join([rotate_char(char, ROT) for char in line]))