[asdf-devel] tests failing

[Faré]

On the latest SBCL compiled myself w/o CLC, my tests all pass on SBCL.
The problem is that the tests expect that there should be no filename
redirection, but CLC does it.

The attached patch will disable CLC if present and make the tests
pass. It will also solve test-module-pathnames.script for clisp.

test-retry-loading-component-1.script remains broken on my debian's
(old?) clisp 2.48, which looks like a bug in that old clisp (it can
find the restart, but not invoke it???)

[ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ]
Most economic fallacies derive... from the tendency to assume that there
is a fixed pie, that one party can gain only at the expense of another.
        -- Milton Friedman



2009/10/22 Gary King <gwking at metabang.com>:
> Weird.
>
> On my Mac, it's http://common-lisp.net/project/asdf/test-results.html
>
> SBCL passes all; the others only fail on one.
>
> I'll check it out.
>
>
> On Oct 22, 2009, at 12:28 PM, Faré wrote:
>
>> I ran the tests for asdf. Out of 17, 5 fail on sbcl and 7 on clisp. Oops.
>>
>> On SBCL, the failures seem related to the pathname redirection of ABL /
>> CLC.
>> test1.script test2.script test3.script test-force.script
>> test-static-and-serial.script
>>
>> On CLISP, the two additional failures are
>> test-module-pathnames.script
>> where clisp reads "foo.cl" as having directory nil instead of
>> (:relative), which I think is legal, and
>> test-retry-loading-component-1.script
>> which seems to have to do with a restart-case retry in do-one-dep, but
>> that's beyond my depth.
>>
>> The good news is most of these failures are only failures of the test
>> themselves.
>> The bad news is someone probably has to fix them and commit a fix.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: asdf.diff
Type: text/x-diff
Size: 3864 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/asdf-devel/attachments/20091023/81c6b21a/attachment.diff>