Re: [proxyTools-users] first time user of proxyTools (was:Re: Help for an absolutely new user)

[wayne <wa...@ny...>]

> From: Hat <ha...@ny...>
> To: pro...@li...
> Subject: Re: [proxyTools-users] first time user of proxyTools (was:Re: Help for an absolutely new user)
> 
> Hello wayne,

Hi Hat,

> On 7 Nov 2002, wayne <wa...@ny...> wrote:
> >> From: Mat Brown <mat...@ya...>
> >> Subject: Re: [proxyTools-users] first time user of proxyTools (was:Re:
> >> Help for an absolutely new user)
> 
> [...]
> 
> >
> >I don't know how to advise on the CVS stuff. It requires no extra 
> >knowledge, but you're more likely to see bugs, crashes etc. 
> >As long as you're willing to keep a working version somewhere as 
> >a backup, you can take advantage of it (the fixes, and all the new 
> >features as they happen). 
> >You get to be able to tell me about bugs more often.
> >You get to call yourself a beta tester :-)
> >
> >The biggest problem seems to be for people to be able to use WinCVS 
> >properly :-(
> 
> I don't recommend WinCVS. Instead if users don't have Cygwin installed to
> use CVS, then I recommend jCVS (Java CVS). It is easy to use.  Here is a
> link from the archive to a tutorial I once wrote on jCVS: 
> 
> http://groups.yahoo.com/group/proxy-elites/message/4097

I should have mentoned Cygwin; I couldn't live without it now.

> >Or if yu don't have much time, and want to get alive/dead and speed 
> >results just use -t 0
> 
> Well, I guess there is some bug in the latest version of SP:
> 
> C:\LP\proxyTools>perl statproxy.pl -v
> This is statProxy 4.120 (proxyTools.zip)
> 
> Author: wa...@ny...
> 
> This program is free software; you can redistribute it and/or
> modify it under the same terms as Perl itself.
> 
> C:\LP\proxyTools>perl statproxy.pl -t 0 -l uncommon.txt
> Loading proxies...
> Extracting proxy strings, expanding/skipping ports, validating, resolving,
> dedup
> ing...
> 90 proxies to test (after processing)
> Running test: 0 130.225.8.117:8000
> 
> As you see the last line, the tested proxy is in the same line of the
> statement  "Running test: 0", and no results are shown per tested proxy!

Not a bug. The new statProxy work slike this.
The design is now parallel testing instead of sequential (only 
partially implemented as of now, but the design had to change 
completely).
Essentially, it's looping over all the proxies and then changing 
the test. The results can't be printed on-the-fly as before, and 
are printed in the standard format at the end.
If you watch that line above, the proxy under test will be changed.
Test 0 is the slowest test too (it measures the speed, so the whole 
page must be waited for). Try -t d, if you want to see the new 
code in action :-). That just tests them all for connectability.
Depending on your system (Windows is bad), it should be able to 
test your 90 proxies in one timeout (20 secs, IIRC). Maybe two 
timeouts on Windows 95/98/ME. Many more proxies in one timeout on 
any Unix system.

This code is new, and only now (last few versions) does it appear 
to be settling down on all systems. I am developing on Win2k, 
NetBSD and Linux, so I'm still keen to hear about the ones I can't 
test on (Win95/98/ME and XP, and other Unices).

You can bump up the debug level from the command line (-x 3 say) 
to see what's really happening.

> Regards,
> Hat

--
wa...@ny...
http://proxytools.sourceforge.net/