Anthony:
*>As I mentioned earlier, replace BADWORDS.match(value) with
BADWORDS.search(value).
*. Sorry I missed that, you made it perfectly clear...
*> add word boundaries r'\b' to the RE,
>.. Though he said he explicitly did not want to do that.
*. Sorry I wasn't clear on that.... but the word boundary will be useful to
me in future.
Thanks all
~Rob
I will return just a True, or False with this, then add a validator to the
username for registration,
then do some testing, and post complete solution, moving code to the proper
place, with conditional etc.
*Note*:
Google doesn't allow bad words for usernames,
but rather than lecture you on bad words, they just say
Username:* frig*you123 is already taken...
and you can keep trying, like
* frig*you123vvzz.yeearbob
and Google will just tell you that name is taken,
which I don't think it is :)
*Here is my code, now working...
*
# this import is required in web2py
import base64, re
#let's assume:
# username can't contain spaces, just a-z and periods
# 'frig' is a very bad word
# 'sadf' is a racial slur
# so even if a person's name as frig, or asdf in it
# we will not let them use that.
# asdf - is a bad username
# asdfyou - is a bad username
# youasdf - is NOT a bad username, but IT SHOULD BE.
badlist = ['frig', 'asdf', 'etc']
BADWORDS = re.compile(r'|'.join(badlist))
class IS_BAD:
def __init__(self,error_message='bad word is contained in username'):
self.error_message = error_message
def __call__(self,value):
if BADWORDS.search(value):
return (value,self.error_message)
return (value, None)
--
