On 6/12/06, Ben Osman <[EMAIL PROTECTED]> wrote:
Atsushi Shibata wrote:
> Hi :-).
>
> How about customizing "portal_skins/coreblog2/tbping" ?

> Put floowing code after if block, so you can avoid trackback including
> "buzz words" string.
>
> if excerpt.lower().find("buzz words") == -1:
>     return
>
> If you use Python 2.3, you can also write like this.
>
> if "buzz words" in excerpt.lower():
>     return
>
> I think you should learn more about Python ;-).
>
> Have fun :-)
>

I've been getting spam too - so i used Atsushi's suggestion and extended
it. Ideally should load the spamwords list from another file.

You should add more words to the list (do a search for something like
trackback spam word list)

In "portal_skins/coreblog2/tbping" add following code after
"excerpt = cbtool.convert_charcode (excerpt)":

spamWords = [
'casino',
'prilosec',
'viagra',
]

words = excerpt.split(' ')

for word in words:
   if word.lower() in spamWords:
     return

This (plus mod_security for apache) works pretty effectively,  Unfortunately, there seems to be a never-ended supply of spammable words (I don't even know what "retin" is ;).  I'm back to requiring authentication for trackbacks, effecitvely turning them off.  I think moderation is the answer, combined with preventitive measures.

Just me .02,

nick

--
Nick Owen
WiKID Systems, Inc.
404.962.8983 (desk)
404.542.9453 (cell)
http://www.wikidsystems.com
At last, two-factor authentication, without the hassle factor
Now open source: http://sourceforge.net/projects/wikid-twofactor/
_______________________________________________
COREblog-en mailing list
[email protected]
http://postaria.com/mailman/listinfo/coreblog-en
Unsubscription writing to [EMAIL PROTECTED]

Reply via email to