I'd prefer we avoid pointless flexibility.
Recommend we pick one, and since underscore requires less quoting and
can be used as part of a method name, recommend we go with that.
Adam K
On Sun, Nov 1, 2009 at 7:07 AM, David Golden <xda...@gmail.com> wrote:
> On Mon, Oct 12, 2009 at 10:08 AM, David Golden <xda...@gmail.com> wrote:
>> On Mon, Oct 12, 2009 at 5:33 AM, Zefram <zef...@fysh.org> wrote:
>>> David Golden wrote:
>>>>For this reason I'd like to formally define the acceptance of 'x-' prefixed
>>>>key names.
>>>
>>> Good concept, but I think "x_" would be a better prefix than "x-".
>>> We've already established usage of "_" to separate words, making the
>>> top-level hash keys satisfy identifier syntax. (Except for "meta-spec",
>>> which is an ugly exception.)
>>
>> I don't have a strong feeling about "x_" vs "x-". The former is more
>> consistent with the naming pattern of the spec and the latter is more
>> consistent with how extension fields are named in a lot of RFCs.
>
> Since no one has strong feeling about it, the new rule shall be qr{^x[_-]}i.
>
> Patch here:
>
> http://github.com/dagolden/cpan-meta-spec/tree/34-private-keys
>
> -- David
>
