perl-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stas Bekman <>
Subject Re: [TODO] completed
Date Sat, 23 Apr 2005 16:33:21 GMT
Randy Kobes wrote:
> On Thu, 21 Apr 2005, Philip M. Gollucci wrote:
>>knock out 1 of the todo/bugs_apr_ext :)
>>FYI: APR::NetLib does not exist.
>>See attached.
> That'd be nice to clean this up ... I'm wondering
> though about having, in particular, apr-ext tests for
> that simply load the modules. In looking at the corrsponding
> responses in t/response/TestAPR/, it appears in their
> current form that all 3 of these require mod_perl to do
> their work. Might having a corresponding apr-ext test lead
> some to believe that these function outside of mod_perl, and
> so in this context their presence could be considered
> somewhat misleading?

That's correct for the socket modules, as there is no way to get the apr 
socket at the moment but via mod_perl. but what's wrong with brigade API? 
you can create one yourself:

   $bb = APR::Brigade->new($pool, APR::BucketAlloc->new($pool));

> Also, I seem to recall that we had a test just for
> APR::Const, but dropped it, as APR::Const was used in a
> number of the other tests anyway. The test itself I seem to
> remember was just comparing some APR::Const::* to a
> numerical value, which may be difficult to maintain, in case
> APR at the C level changes the numerical value.

Right and APR::Const is indeed loaded by other apr-ext tests anyway.

Stas Bekman            JAm_pH ------> Just Another mod_perl Hacker     mod_perl Guide --->

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message