On Thu, Sep 29, 2011 at 11:43 PM, Jason Glasgow <jglasgow_at_google.com> wrote:
> Daniel & Ben,
> Thanks for the feedback.  Let me try to address the issues you raise and
> I'll send an updated patch.
> -Jason
>
> On Thu, Sep 29, 2011 at 2:14 AM, Daniel Stenberg <daniel_at_haxx.se> wrote:
>>
>> On Wed, 28 Sep 2011, Ben Greear wrote:
>>
>>> Last time I posted a similar patch, I think it was requested that the
>>> option name be more generic.  I ended up with CURLOPT_DNS_SERVERS, just in
>>> case some day it supports more than just c-ares.
>>
>> Good point!
>>
>> I would also prefer to have the function return CURLE_NOT_BUILT_IN if
>> libcurl wasn't built to support the option (rather than unknown option).
>>
>> --
>>
>>  / daniel.haxx.se

It took me awhile to get back to this patch. Here is a second try
that I think addresses all the concerns short of testing. If this
looks acceptable, I will try to figure out how to properly test this
in the existing test framework.
Received on 2011-11-06