[asdf-devel] Tests now leave files behind

[Robert Goldman]

On 3/15/10 Mar 15 -10:26 PM, Faré wrote:
> The simplest way to do it would be something like
> 
> export ASDF_OUTPUT_TRANSLATIONS=/:$PWD/../tmp/fasls/$command
> 
> or maybe you'll prefer that at the end of defun load-asdf:
> 
> (asdf::initialize-output-translations
>  `(:asdf-output-translations
>    (:root (,(truename (asdf:merge-pathnames* #p"../tmp/fasls/
> *default-pathname-defaults*)) :implementation))
>    :ignore-inherited-configuration))

Sigh.  And this reminds me that I never managed to get the CLEAN-OP into
ASDF.  I had it all queued up to go, was just waiting to get some tests
written, and then the code got all rearranged, and I never slogged
through the diff process to figure out how to salvage the CLEAN-OP....

But we should probably have one for this.

Alternatively, should we use something like gentemp to make the fasl
directory and then blow it away afterwards?

I just the other day wrote some filesystem related tests and, yes, they
came back to bite us when I didn't clean some of the files up after
myself.... :-/

cheers,
r

> 
> [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ]
> No woman ever falls in love with a man unless she has
> a better opinion of him than he deserves.       — Edgar Watson Howe
> 
> 
> On 15 March 2010 22:19, Robert Goldman <rpgoldman at sift.info> wrote:
>> On 3/15/10 Mar 15 -10:11 PM, Faré wrote:
>>> I think that we should turn relocation on, and make it happen in a
>>> known temporary directory (e.g. tmp/fasls/) that we can remove
>>> afterwards.
>>>
>>> janderson's test also creates files, that should be made to exist
>>> under same tmp/ directory.
>>>
>>> Will you do that? Otherwise, I'll do it later this or next week.
>>
>> I'll see if I can figure it out.  It would probably be good for me to
>> figure out how the output-translations work.
>>
>> Is this a job for "translate ALL the binaries" under lisp control or
>> should we put one of those local relocation specifier files into the
>> tests directory, and do this that way?
>>
>> thanks,
>> r
>>
>>>
>>> [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ]
>>> First, they said Java was good for app-lets; then they recoiled, and said
>>> it was good for serv-lets; actually, what it's good for is toy-lets.
>>>
>>>
>>>
>>> On 15 March 2010 22:07, Robert Goldman <rpgoldman at sift.info> wrote:
>>>> I find that the tests don't clean up after themselves.  The tests now
>>>> leave behind .fasl files in ~/.cache/common-lisp/**/...
>>>>
>>>> This is not a directory structure I ever used before, nor is it one I
>>>> ever intended to use.  I /believe/ that the tests used to tidy up the
>>>> fasls they created.
>>>>
>>>> I think either the ASDF fasl relocation facility should be turned off,
>>>> or we need to provide a new clean up facility.  I would prefer the
>>>> former (except for tests that explicitly test the facility, preferably
>>>> into /tmp rather than into the user's homedir).
>>>>
>>>> Best,
>>>> r
>>>>
>>>> _______________________________________________
>>>> asdf-devel mailing list
>>>> asdf-devel at common-lisp.net
>>>> http://common-lisp.net/cgi-bin/mailman/listinfo/asdf-devel
>>>>
>>
>>