oxo @oxo@qoto.org

Making steps into the right direction with my Rijndael cipher script. Almost done now with the key expansion and the so called g-function. Last step is implementation of the round constants.
When that is done I 'only' have to put all the stuff in a proper loop. If that works I will be happy, but the journey will not end there. A lot of testing and see how to decrypt...
😃

#linux #bash #shell #script #aes #rijndael #encryption

https://codeberg.org/oxo/tool/src/branch/main/dev/aes

Finally found my way through the Galois Fields.
The mix-colums part of the Rijndael key scheduler is done now.
Next up: generating round keys.

#linux #bash #shell #script #aes #rijndael #encryption

https://codeberg.org/oxo/tool/src/branch/main/dev/aes

In my 'baes' :) project I am struggling with Galois Fields and get field-exceeding values properly calculated.

I am home now and suddenly it struck me that probably the solution can be (note-to-self):

After multiplication with the irreducible polynomial, use a mask and a bitwise AND to keep only the lower 7 bits of the final result value.

I.e.:
value=0xA3F
mask=0x7F
result=$(( value & mask ))
So then the result will become: 0b00111111

Hopefully this will work...

#linux #bash #shell #script #aes #rijndael #encryption

https://codeberg.org/oxo/tool/src/branch/main/dev/aes

Shift rows in the pocket!
#linux #bash #shell #script #aes #rijndael #encryption

https://codeberg.org/oxo/tool/src/branch/main/dev/aes

The substitution box, one of the steps in key scheduling for the Rijndael cipher, is working.

The table is in hexadecimal, therefore some sub-optimal conversion is going on. But it works for now.

Still far away from fully working, but I am still making good progress each time I work on this project.

Next up: bitwise row and column shifting.

https://codeberg.org/oxo/tool/src/branch/main/dev/aes

#linux #bash #shell #script #aes #rijndael #encryption

Steadily on my way builing an Rijndael-esque encryptor.
Building upon my #xor shell script; yesterday I made an 256 bits sustitution table.
Today I made the foundation of a key scheduler. Interestingly bash is able to process 4×4, 6×4 or 8×4 matrices. And the code is still quite simple. I am learning a lot and enjoying all the little puzzles to solve!
Follow my progress on #codeberg.
https://codeberg.org/oxo/tool/src/branch/main/dev/aes
#linux #bash #shell #script #aes #rijndael #encryption

Duck.ai is not able to generate a complete substitution matrix from the Rijndael cipher. It starts off promising, but then gets stuck about halfway. I asked it to generate the table from the halfway point and it did. But it made up the values it produced. #stupidai

Cut the #base64 endoding from my #xor shall script. Also adjusted to a non-redundant key length for the last chunk of input into the xor function.
Next hurdle: proper key rotation...
Will upload my progress to the repos asap.
#linix #bash #xor #enccryption

The #base64 command has a wrap option. If it is set to 0 then you basically have created a base64 stream (without any EOL's). Which comes in handy for my #xor project.
Development btw is published when I have internet connection;
https://codeberg.org/oxo/tool/src/branch/main/dev/aes
Enjoy your day!
#linux #commandline #script