Attempting to convert [MonkeyFighter](https://github.com/TeamOneTickets/Monkey-Fighter) to dynamically handle all grawlix replacements in order to un-censor words.

MonkeyFighter.php

    /**
     * Symbols that could be used to replace letters
     *
     * @var string
     */
    protected $_grawlix = '[!@#$%&*-]';


    /**
     * Associative array of replacements to make
     *
     * @var array
     * @lol http://www.youtube.com/watch?v=vbZhpf3sQxQ
     */
    protected $_replacements = array(
        '/S' . $this->_grawlix . '{2}T/'                => 'SHIT',
        '/([Ss])' . $this->_grawlix . '{2}t/'           => "$1" . 'hit',

        '/P' . $this->_grawlix . '{2}s/'                => 'PISS',
        '/([Ss])' . $this->_grawlix . '{2}s/'           => "$1" . 'iss',

        '/F' . $this->_grawlix . '{2}K/'                => 'FUCK',
        '/([Ff])' . $this->_grawlix . '{2}k/'           => "$1" . 'uck',

        '/F' . $this->_grawlix . '{3}ING/'              => 'FUCK',
        '/([Ff])' . $this->_grawlix . '{3}ing/'         => "$1" . 'ucking',

        '/C' . $this->_grawlix . '{2}T/'                => 'CUNT',
        '/([Cc])' . $this->_grawlix . '{2}T/'           => "$1" . 'unt',

        '/C' . $this->_grawlix . '{2}K/'                => 'COCK',
        '/([Cc])' . $this->_grawlix . '{2}k/'           => "$1" . 'ock',

        '/P' . $this->_grawlix . '{4}Y/'                => 'PUSSY',
        '/([Pp])' . $this->_grawlix . '{4}y/'           => "$1" . 'ussy',

    );

I just don’t see this working out as I don't see a way to dynamically decide how many grawlixes may be present in a censored word.

For example, without specifying a lot of possibilities how can we make 'P---y', 'P--sy', 'Pu--y', 'P----' all end up as 'Pussy'?

This seems to work pretty well

protected $_replacements = array(
    '/([Ss])[hit!@#$%&*-]{3}/'                      => '$1hit',
    '/S[HIT!@#$%&*-]{3}/'                           => 'SHIT',

    '/([Pp])[usy!@#$%&*-]{4}/'                      => '$1ussy',
    '/P[USY!@#$%&*-]{4}/'                           => 'PUSSY',

    '/([Pp])[is!@#$%&*-]{3}/'                       => '$1iss',
    '/P[IS!@#$%&*-]{3}/'                            => 'PISS',

    '/([Mm])[otherfuck!@#$%&*-]{11}/'               => '$1otherfucker',
    '/F[OTHERFUCK!@#$%&*-]{11}/'                    => 'MOTHERFUCKER',

    '/([Cc])[unt!@#$%&*-]{3}/'                      => '$1unt',
    '/C[UNT!@#$%&*-]{3}/'                           => 'CUNT',

    '/([Cc])[ock!@#$%&*-]{3}/'                      => '$1ock',
    '/C[OCK!@#$%&*-]{3}/'                           => 'COCK',

    '/([Dd])[amn!@#$%&*-]{3}/'                      => '$1amn',
    '/D[AMN!@#$%&*-]{3}/'                           => 'DAMN',

    '/([Aa])[s!@#$%&*-]{2}/'                        => '$1ss',
    '/A[S!@#$%&*-]{2}/'                             => 'ASS',

    '/([Bb])[itch!@#$%&*-]{4}/'                     => '$1amn',
    '/B[ITCH!@#$%&*-]{4}/'                          => 'BITCH',

    '/([Tt])[ity!@#$%&*-]{4}/'                      => '$1itty',
    '/T[ITY!@#$%&*-]{4}/'                           => 'TITTY',

    '/([Tt])[it!@#$%&*-]{2}(s)?/'                   => '$1it$2',
    '/T[IT!@#$%&*-]{2}(S)?/'                        => 'TIT$1',

    '/([Nn])[iger!@#$%&*-]{5}/'                     => '$1igger',
    '/N[IGER!@#$%&*-]{5}/'                          => 'NIGGER',

    '/([Nn])[iga!@#$%&*-]{4}(h)?/'                  => '$1igga$2',
    '/N[IGA!@#$%&*-]{4}(H)?/'                       => 'NIGGA$1',

    '/([Ff])[agot!@#$%&*-]{5}/'                     => '$1aggot',
    '/F[AGOT!@#$%&*-]{5}/'                          => 'FAGGOT',

    '/([Ff])[uck!@#$%&*-]{3}/'                      => '$1uck',
    '/F[UCK!@#$%&*-]{3}/'                           => 'FUCK',

    '/([Ff])[ag!@#$%&*-]{2}/'                       => '$1ag',
    '/F[AG!@#$%&*-]{2}/'                            => 'FAG$1',

My issue with that is that you don't know the capitalization of the final product.

I just landed and don't have net access except for on my phone, but I can send you a little test I did on the flight.

Basically, my idea was that given a list of final band names, you pass a grawlixed string and it returns the one it thinks it is. It would convert all of "the f**_ing champs", "the ***_ing champs" and "the ******* champs" all properly.

It does it via conversion of the grawlixed string to a regex and comparing down the list. It's ruby, but its only like 30 lines of code.

https://gist.github.com/spikegrobstein/5813363

just remembered that I needed to post this. had limited free time on my trip to get on here.

Let me know what you think of this concept... basically I think you need a file that contains known degrawlixed strings since I have seen cases of censored band names being more ambiguous.

There are some limitations in my example which would need to be addressed in a final library. For one, I there would need to be hard-coded known censored names. For instance for "God Below" and "God Forbid" which are frequently grawlixed as "G- Forbid" or "G. Forbid" or "G-- Forbid". Maybe the degrawlix.db file could be a slightly better format than just lists of final names, but rather have support for regex for special cases or specific hard-coded transformations. I'm just kinda spitballing here.

Also, I like your implementation of a grawlix character better than mine. I was thinking about the best way to do that on the plane, but I would have used yours had I had access on the plane, but was just trying to get it working.