[imp-hackers] Common test suite

Juan Jose Garcia-Ripoll juanjose.garciaripoll at googlemail.com
Tue Jun 30 15:21:14 UTC 2009


On Tue, Jun 30, 2009 at 3:49 PM, Sam Steingold<sds at gnu.org> wrote:
> good. does everyone who wants to work on this has an account on sourceforge?
> any opposition to moving there?
> I have been using their services since 2000 and have been reasonably happy.

Both c-l and sourceforge have their quirks. I know because I have
moved the ECL project on both directions. In particular SF has
improved in the last months, but before moving I just would like to
ensure that this is done for good reasons....

> actually, clocc on sourceforge already has an "ansi-test" module
> (based, I think, on the clisp internal regression test suite from
> several years ago).  we can merge the gcl ansi test suite into it.

and not for merging it into a different project or just for the sake
of taking it over. I do not see much of a difference merging the ansi
test suite into clocc project or having it in the ABCL source tree.

> alternatively, we can create a brand spanking new SF project ansi-lisp-test.
> (ansi-cl-test? cl-test?)

_If_ this has to be moved, let it be to a separate project with its
own purpose in life. I would then vote for cl-test-suite or
ansi-cl-tests, as opposed to cl-test, which would suggest something
like a test framework.

>> but what exactly is this policy problem? Too short
>> user name? Or are they somehow refusing to create an account at all?
>
> the policy requires "first initial + last name", i.e., ssteingold for me.

Would it help if several of us long time users of c-l.net wrote to
them about this case...? Well, I already took the liberty to do so :-)
Administrators usually suggest names but can be sensitive to people's
dislikes of their choices.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com




More information about the implementation-hackers mailing list