anuvab/leetspeak

A fairly robust LeetSpeak to English converter using probability of word usage

leetspeak

Years ago I started working on Garot Bot 2.0. Well, that project was enormous and never completed. However, as part of the project I wanted to have a LeetSpeak to English translator. This multi-threaded Python 3 converter/translator deciphers your messages using probability of word usage taken from the Project Gutenberg from all the possibilities it can generate. I have yet to see any translator that works in as many cases as this one.

Using

1. Install python3.
2. Run: python3 leet.py --demo to test or python3 leet.py [-e] "Your Message" to decode (by default) or encode a message.

Notes

• It will run slower the first time while it generates the .pickle files.
• This program decodes better than encodes since encoding just randomly choses characters, which does not look at all like normal leet.
• The program ignores all punctuation. How does it know if that period is part of the leet or is a period?
• I'm not fond of the coding style I used back then. This was one of my first major Python projects.

Example

Some things work.

 [garrett leetspeak]$ python3 leet.py -e "How are you doing today mother?"
 )-(ω\_1_/ /\|2ə \j()/_/ [)¤ai/|/gee -l-¤l]/-\-/ /|/|{}-l-[-]€[z?

 [garrett leetspeak]$ python3 leet.py ")-(ω\_1_/ /\|2ə \j()/_/ [)¤ai/|/gee -l-¤l]/-\-/ /|/|{}-l-[-]€[z?"
 how are you doing today mother

And some things don't. Note that Github was not a common word a few hundred years ago.

 [garrett leetspeak]$ python3 leet.py -e "This is a test for Github."
 '][']-[eyeš 3y35 /-\ +e$-1- |=0® (_+l†}-{yuu|3.

 [garrett leetspeak]$ python3 leet.py "'][']-[eyeš 3y35 /-\ +e$-1- |=0® (_+l†}-{yuu|3."
 this is a tehiubhig for

License

The words lists are from /usr/share/dict, which I believe is from aspell, under GPL2. I would say where stopwords_en.txt came from, but I think that's a combination of many sites and my own thoughts. As for slang, it's from No Slang, and I don't see a license. The gutenberg_small.txt file is obviously from Gutenberg; although, I have since lost the script to generated it (hence why I just provide the file here). The phonetics.py file is from AdvaS Advanced Search and is GPL. All of my code, as always, is under the ISC license.