Re: Stopping Sporge tutorial using NewsProxy (where do we get nfilter.dat)?

de Me.Here 02/01/2008 12:35

[OK] Peter J Ross wrote:
> In news.software.readers on Thu, 31 Jan 2008 11:48:42 -0600, M.L.
> <me@privacy.invalid> wrote:
>
> [I previously wrote:]
>>> Without XHDR (or HEAD implemented in the filtering software),
>>> anti-sporge filters aren't likely to work very well, though you can
>>> detect a few patterns, such as an absence of lower-case letters in the
>>> right hand side of a Message-ID or email address, which are
>>> characteristic of *some* sporge posts and very few non-sporge posts.
>> Nice tip.
>
> Then please give me the credit by including an attribution line for
> the quoted text. This thread is about dealing with Usenet annoyances,
> so it's ironic that you have a mildly annoying habit yourself.
>
>> Would this filter satisfy that criterion?
>> * drop Message-ID:.*[^a-z].*
>
> No. That would drop any article whose Message-ID contained any
> character other than lower-case a-z, including "@" - IOW, it would
> drop all articles.
>
> Try
> * drop Message-ID:.*@[^a-z]+>.*
>
> This should match a string of characters, none of which is a
> lower-case letter, between "@" and ">".
>
> The final ".*" takes account of any trailing white space after the
> Message-ID's closing angle-bracket: " *", "[ ]*" or "s*" would do so
> more efficiently, but I don't know which, if any, would work with
> NewsProxy, since I don't know what kind of regular expressions it
> uses. (And I can't install it myself, since I'd have to install
> Windows first.)
>

I've seen some servers using an IP in the Message-ID line. Not sure if
it was a sporged one or not though... If it turns out to be a valid
use, the REgEx would need to take that into account.

Would something like the following work:

* drop Message-ID:.*@[^a-z0-9]+>.*

--
I filter all Google posts - sorry, too much spam