I wanted a global way to filter profanity w/out having to modify every model, view, or form. While middleware takes overhead, this technique is intended mainly for sites w/few postbacks. Hopefully this snippet will lead to more/better techniques in the comments below. Usage (settings.py): MIDDLEWARE_CLASSES = ( 'PROJECT_NAME.FILE_NAME.ProfanityFilterMiddleware', )
A better way of dealing w/profanity - disemvowel it! From [Wikipedia](http://en.wikipedia.org/wiki/Disemvoweling), "disemvoweling is a technique used to censor unwanted postings such as spam, internet trolling, rudeness or criticism and yet maintain some transparency, both of the act and the underlying word." Credit: Boing Boing Example: This original sentence: In the fields of Internet discussion and forum moderation, disemvoweling (also spelled disemvowelling) is the removal of vowels from text. would be disemvowelled to look like this: n th flds f ntrnt dscssn nd frm mdrtn, dsmvwlng (ls splld dsmvwllng) s th rmvl f vwls frm txt. Usage: body_input = form.cleaned_data["body"] body_input = disemvowel_profanity(body_input)