> Attached patch makes building in the dynclass ghetto a bit less
> inhospitable...
Thanks, applied.
leo
Are we comfortable adding the dynclasses to the default build target?
I want to at some point, if only to make really sure that we don't
break them. (As stuff that gets built and tested by default stays up
to date, while the rest... tends not to)
--
Dan
--------------------------------------it's like this-------------------
Dan Sugalski even samurai
d...@sidhe.org have teddy bears and even
teddy bears get drunk
> At 10:07 AM +0100 11/4/04, Leopold Toetsch wrote:
>
>> Sam Ruby <ru...@intertwingly.net> wrote:
>>
>>> Attached patch makes building in the dynclass ghetto a bit less
>>> inhospitable...
>>
>> Thanks, applied.
>
> Are we comfortable adding the dynclasses to the default build target? I
> want to at some point, if only to make really sure that we don't break
> them. (As stuff that gets built and tested by default stays up to date,
> while the rest... tends not to)
The best way to find out is to add it.
I would prefer that this be added sooner rather than later, and now is
as good a time as any.
Within a day or so, I should have an initial set of python PMCs that I
would like to see committed, as well as the beginnings of a test suite.
- Sam Ruby
Lets do that then.
> > I want to at some point, if only to make really sure that we don't
>> break them. (As stuff that gets built and tested by default stays up
>> to date, while the rest... tends not to)
>
>Good point. We need tests:
>1) basic operation: is dynclass loading working (current samples,
> modified not to use absolute type numbers)
>
>2) Where should dynamic {tcl, perl, python, ...}_classes tests go?
>
> t/xt/$lang # xt for extension and test sorting?
If the individual dynclasses have pir or pasm that can test them,
then it can go in t/dynclass/<pmcname>.t. If it needs anything
fancier... we can punt for now. Language specific PMCs can be tested
in the language's test suite, I expect.
Dan Sugalski wrote:
> At 3:51 PM +0100 11/4/04, Leopold Toetsch wrote:
>
>> Dan Sugalski <d...@sidhe.org> wrote:
>>
>>> Are we comfortable adding the dynclasses to the default build target?
>>
>>
>> I think it can go in.
Done.