perl-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Randy Kobes <ra...@theoryx5.uwinnipeg.ca>
Subject Re: [mp2] directive/setupenv.t test on Win32
Date Sat, 11 Dec 2004 16:48:58 GMT
On Sat, 11 Dec 2004, Stas Bekman wrote:

> Randy Kobes wrote:
> > On Win32, the last subtest of t/directive/setupenv.t tests
> > that $env{HOME} doesn't exist. However, as we've discussed
> > before in some other contexts, for some reason Win32 does
> > pass, in particular, the HOME environment variable, if it's
> > set. I have HOME set, and consequently, this last subtest
> > fails.
> >
> > I'm not sure if this different behaviour is due to
> > mp2 or to Apache. In the meantime, would it be acceptable
> > to skip this subtest for Win32:
>
> +1, but please add a little comment, like:
>
> # on win32 sometimes $ENV{HOME} is set (probably by Apache) and
> # sometimes it's not
>
> But may be we should choose a different ENV var which exists on all OSes?
> SHELL? USER?

That's a good idea .... On Win32, neither SHELL nor USER is
used in a standard configuration - PATH seems to be the only
one in common (but that is set). Should do a Special Case
(tm) for Win32, using, eg, OS, only to skip it, or just
leave it as is (with the comment)? I think the problem with
the way the test currently is is that it'll pass on Win32 in
a standard configuration where $HOME isn't set at all,
whereas the test presumes that $HOME has been set but
doesn't exist at this level because it's not been passed
through.

-- 
best regards,
randy


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org


Mime
View raw message