On Wed, Oct 15, 2008 at 3:51 AM, Alex Rousskov
<rousskov_at_measurement-factory.com> wrote:
> On Wed, 2008-10-15 at 12:35 +1100, Robert Collins wrote:
>
>> > > IMO, if a user explicitly requested feature Foo and Foo cannot be
>> > > supported, we should fail rather than ignore the user request.
>>
>> I completely agree with this. The usual behaviour I see that works well
>> is:
>>
>> no-parameter supplied:
>> Feature enables if possible
>>
>> --without | --disable:
>> Feature does not enable and does not probe for dependencies
>>
>> --with | --enable:
>> Feature enables and probes for dependencies, any missing dependencies
>> are fatal.
>
> The current code supports the last two. Kinkie suggested the first one.
Yes (and sorry for the one-liner).
--
/kinkie
Received on Wed Oct 15 2008 - 05:02:57 MDT
This archive was generated by hypermail 2.2.0 : Wed Oct 15 2008 - 12:00:05 MDT