On Wed, May 06, 2009 at 12:15:38PM -0400, David Golden wrote:
On Wed, May 6, 2009 at 12:05 PM, David Cantrell <da...@cantrell.org.uk
> wrote:
That's true, but this isn't just about testing libs - it's any
lib I
might delete from the distribution.
You want Devel::Hide.
I think the original question was about catching a module deleted by
mistake -- otherwise, Devel::Hide rocks.
Exactly. If I accidentally delete Foo/Bar.pm in my Foo-Bar
distribution, then that recipe would hide any Foo::Bar that's already
installed, but still try to load it from blib/* when I try to build
the
distro. And then, because it was accidentally deleted, complain
loudly.
Ok. Thanks for the mention of Devel::Hide. There is also
Test::Without::Module and Module::Mask, which all look pretty similar.
But I still have two problems with using any of these directly:
1) I have to specify exactly which modules to hide. Since I don't know
in advance which modules I might delete, I want to hide all the
modules that CHI provides (which will be tedious and error-prone to
list), or at least hide everything matching /^CHI::/.
2) I have to compute the right exception paths. Doing "use lib qw(blib/
lib blib/arch)" as you suggest would only work if the tests were run
from the main directory. e.g. If I'm in the t/ directory and do "perl -
I../lib foo.t", as I do sometimes, it ought to work as well.
Would this make sense as its own module, or as an enhancement of one
of the above? e.g.
foo.t:
use Test::More;
use Test::UseOnlyMyVersionsOfTheseModules qr(/^CHI/);
but with a shorter name. :)
Jon
