>>> 1. IIUC, the forbiddenUrl file will accept either urls or regex
>>> "reductions" of rangers of urls. It seems that I can either insert a
>>> list of hosts from any of a number of web sources, or can reduce that
>>> list of 100+/- addresses into a few regex's:
>>> Which would be the better way in terms of processing speed, memory use,
>>> general overhead?
>> Both are slow.
> So I'm guessing that the table of urls is not memory resident? Is there
> some tweak that I could make to the source code to speed it up (e.g.
> binary search; memory residence?)
Oh, my. That's not Firefox, it's slow, but not *that* slow.
Everything is memory resident. It's slow in the sense that it's easy
to notice in benchmarks, not in the sense that a human being will
notice.
Before Polipo sends a request to a server, it searches first its table
of domain names, then compares the URL with the regexps. The first
step is in O(n) (it's a mere linear search), the second is done with
your system's regexp library, but should usually be in O(n) too.
Polipo will easily scale to 1,000 patterns in your forbidden file (mine
is 250 lines long). It probably won't scale to 100,000 patterns.
> My goal is to use TOR more often, and to spare it the impact of
> transferring a bunch of advertising.
Don't worry, the cost of walking the forbidden table will be lost in
the noise.
> FWICT, both Opera and FireFox have internal adblocking mechanisms
I haven't checked -- but I bet a pint or two that Polipo's mechanism
is fater than whatever is in Firefox.
Juliusz
-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Polipo-users mailing list
Polipo-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/polipo-users
