Re: asdf-install installs more than you ask for

On 2/3/06, [email protected] <[email protected]> wrote:
> >>>>> "TFB" == Thomas F Burdick <[email protected]> writes:
>
>     TFB> On 2/3/06, Gary King <[email protected]> wrote:
>     >> I'm not sure I understand. To make this more concrete, suppose I have
>     >>
>     >> * packaged up LIFT so that it contains two system definitions
>     >> (lift.asd and lift-test.asd),
>     >> * packaged up MOPTILITIES so that it containers two system
>     >> definitions (moptilites.asd and moptilities-test.asd)
>
>     TFB> Then don't do that, use asdf:test-op for your tests instead.
>
> That may not be easy to do.  E.g., I often use a local-only test
> system that is not polished enough for me to release to the public.
> In that case I like to make a separate ASDF testing system that can be
> loaded, (with a dependency on the base system).  That way people who
> use my code don't have to be able to use my testing code.  That seems
> reasonable to me.

You don't want to release it to the public, but you make an asdf
system that contains it, in a public release?  It *really* seems to me
that you should be using test-op.  It seems like a better win for
everyone if you made new test-file and test-module component types
that are ignored by load-op, but used by test-op; then, instead of
changing asdf-install, submit the new, improved test support to asdf.

> In general, it's not clear to me that a blanket restriction on the
> contents of asdf-install systems is the best solution to this problem.

You're just proposing to replace one blanket restriction with another.


-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid3432&bid#0486&dat1642