[Catalyst] Is it normal to have lots of trouble installing catalyst?

Perrin Harkins perrin at elem.com
Wed Feb 8 22:09:17 CET 2006


On Wed, 2006-02-08 at 14:39 -0600, Brandon Black wrote:
> But if there are no effective tests without it, then removing it from
> the dependencies and making it an optional/skippable test means that
> the majority of users installing it will not do any effective testing.

Right.  Do you want to have your cake, or do you want to eat it?  Doing
a good job of testing everything without external dependencies is not
feasible, but that shouldn't mean end users have to go through it.

> Making the bulk of your tests skip completely
> because of a missing external package is about the same as just not
> doing a "make test" before "make install" to begin with in these
> cases, or always doing forced installs.

Which is what people say they are doing, if you hadn't noticed.

Personally, I don't run make test on every machine for a module I have
used before.  I have tests for my application, and if they pass, I don't
need to run any of the other module tests.

CPAN authors do not all agree about the purpose of tests and who should
run them.  Some test suites are clearly intended for use by developers
of that module, not as a set of acceptance tests for end users to run.
I think that's a totally legitimate way to use testing.  The only thing
that needs to happen is for those people to make sure their tests don't
mess things up for CPAN.pm users by skipping the ones that require extra
tools to be installed.

- Perrin




More information about the Catalyst mailing list