W3C home > Mailing lists > Public > public-qa-dev@w3.org > October 2007

Re: Warning: Config::General 2.36 is incompatible with the validator

From: Ville Skyttä <ville.skytta@iki.fi>
Date: Fri, 19 Oct 2007 20:34:27 +0300
To: public-qa-dev@w3.org
Message-Id: <200710192034.27863.ville.skytta@iki.fi>

On Friday 19 October 2007, olivier Thereaux wrote:
> Hi Ville,
> On Oct 18, 2007, at 03:14 , Ville Skyttä wrote:
> > On Wednesday 17 October 2007, Ville Skyttä wrote:
> >> FYI:
> >>
> >> Config::General 2.36 [0] was released today, unfortunately with a
> >> regression that affects the validator in its default configuration.
> >> Details at http://rt.cpan.org/Ticket/Display.html?id=30063
> >
> > A somewhat draconian "fix" against CVS HEAD for this is attached.
> > Comments?
> I gave it a good look yesterday. Looks draconian indeed, but for the
> end user, it seems to boil down to "set your base dir explicitely".
> It's a bit of a pity we won't be using $ENV{W3C_VALIDATOR_HOME} as a
> fallback value any more. Not that I expect it's being widely used...

Well, $ENV{W3C_VALIDATOR_HOME} needs to be set too unless /usr/local/validator 
is used so I'm not sure if explicitly setting Base in validator.conf much 
more problematic.

Anyway, I suppose there's another way around the issue, perhaps setting "local 
$ENV{Base} = $ENV{W3C_VALIDATOR_HOME} || /usr/local/validator" and 
using -InterPolateEnv to C::G would work.  Haven't tested it though.

> Do you think the bug in C::G will be fixed quickly? This will impact
> whether we need to put this patch in the markup validator or not.

Your guess is as good as mine.  But I see the change that broke things was 
done in order to fix a real bug that needed to be fixed so it probably takes 
more than just reverting the change: 
Received on Friday, 19 October 2007 17:34:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:36:27 UTC