Re: check for regression-tests.action?

Fabian Keil <[email protected]> Tue, 18 Nov 2014 16:08:18 +0100
Newsgroups	gmane.comp.web.privoxy.devel
Message-ID	<[email protected]>
Ian Silvester <[email protected]> wrote:

> 
> On 14-11-17 2:31 PM, Fabian Keil wrote:
> > Ian Silvester <[email protected]> wrote:
> >
> >> On 14-11-17 1:35 PM, Fabian Keil wrote:
> >>> Lee <[email protected]> wrote:

> >>>> /source/privoxy-3.0.22-stable/current/tools
> >>>> $ ./privoxy-regression-test.pl --max-level 200 --privoxy-address 127.0.0.1:8118
> >>>> 2014-11-17 10:45:05: Asking Privoxy for the number of action files available ...
> >>>> 2014-11-17 10:45:05: Gathering regression tests from 4 action file(s)
> >>>> delivered by Privoxy 3.0.22.
> >>>> 2014-11-17 10:45:06: Executing regression tests ...
> >>>> 2014-11-17 10:45:45: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
> >>>> 2014-11-17 10:45:45: Failure for test 514. Header 'Keep-Alive: 20' and
> >>>> tag 'Parse Keep-Alive header'
> >>>> 2014-11-17 10:45:45: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
> >>>> 2014-11-17 10:45:45: Failure for test 515. Header 'Keep-Alive:
> >>>> timeout=20, max=3' and tag 'Parse Keep-Alive header'
> >>>> 2014-11-17 10:45:45: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
> >>>> 2014-11-17 10:45:45: Failure for test 516. Header 'Keep-Alive:
> >>>> blah=fasel, timeout=20, max=3' and tag 'Parse Keep-Alive header'
> >>>> 2014-11-17 10:46:02: Executed 687 regression tests. Skipped 27. 684
> >>>> successes, 3 failures.
> >>> I believe the problem here is that the tests are executed even though
> >>> they aren't supposed to pass. They are executed if keep-alive-timeout is
> >>> set to any digit but actually require a keep-alive-timeout of 20 or more
> >>> to pass.

This should be fixed in CVS now.

> >> - there is a (minor) error in the config file - it should quote the
> >> default value of 5 rather than None, correct?
> > No.
> >
> > While 5 is the value set in the config file, "default value" refers
> > to the internal default value Privoxy uses if the directive isn't used
> > at all. Often the values specified in the config file match the default
> > values, but sometimes they don't.
> >
> > Maybe the "Default value" and "Effect if unset" sections should be
> > renamed to clarify this.
> Thanks for the clarification Fabian.

You're welcome. It looks like my explanation currently isn't consistently
applied, though. Maybe the meaning changed over time.

> I think that "Effect if unset" is clear (and the config preamble helps), 
> but "Default" is not clear. I think an explanation should be added to 
> section II at the top of the file to explain the difference between the 
> "Default" comment and the default value specified in a vanilla config file.

I'm not sure an explanation in section II will be sufficient.
Currently it already contains the following paragraph:

| Note that commenting out an option and leaving it at its default
| are two completely different things! Most options behave very
| differently when unset. See the "Effect if unset" explanation
| in each option's description for details.

In this paragraph "default" actually refers to the config file default
and looking through the various sections my impression is that in many
cases the "Default value" actually documents just that.

So maybe we should standardise on documenting a "Default (config file) value"
(again) and mentioning Privoxy's internal default in the "Effect if unset"
section. This should also match your previous expectations and probably is
more intuitive than my previous explanation.

Fabian

------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk

_______________________________________________
Ijbswa-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
signature.asc (application/pgp-signature, 819 B)