tech-pkg archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: CVS commit: pkgsrc/security/gnupg21
From: Joerg Sonnenberger <joerg%britannica.bec.de@localhost>, Date: Mon, 21 Sep 2015 11:47:52 +0200
> On Mon, Sep 21, 2015 at 06:30:47PM +0900, Ryo ONODERA wrote:
>> Hi,
>>
>> From: Joerg Sonnenberger <joerg%britannica.bec.de@localhost>, Date: Mon, 21 Sep 2015 11:09:41 +0200
>>
>> > On Mon, Sep 21, 2015 at 09:11:42AM +0900, Ryo ONODERA wrote:
>> >> I would like to commit these and remove security/gnupg2/buildlink3.mk
>> >> after freeze.
>> >> Please send me your comments.
>> >
>> > Wrong package name.
>> ???
>
> It's still called gnupg, not gnupg2.
What is your 'it'?
security/gnupg2 and security/gnupg21 share 'gnupg2' ${PKGBASE},
because upstream says security/gnupg2 and security/gnupg21 cannot
coexists in a environment.
>> > The question remains -- why point source builds to
>> > the 2.0.x version, when binary packages and bulk builds are picking up
>> > the higher 2.1.x version instead?
>>
>> I have no idea about your bulk build environment.
>> In general, higher version/revision number is preferred in pkgsrc.
>
> Yes, but for source builds you tell it to use gnupg2, not gnupg21.
You mean that behavior of security/gnupg2 and security/gnupg2 is
not in usual way?
>> > As we don't have tool support for
>> > gnupg (and I don't see a good reason for wanting to have that), just use
>> > ${PREFIX} and not find-prefix.mk.
>>
>> At the moment, find-prefix.mk has less meaning.
>> However using ${PREFIX} directly may became harmful in future.
>
> For packages without builtin.mk support or custom sub-prefix, it is just
> a left-over of package wedges. As such, it just adds noise and build
> time.
I cannot judge whether it is just noise or not.
If it is really noise, please note it in the pkgsrc guide.
Thank you.
--
Ryo ONODERA // ryo_on%yk.rim.or.jp@localhost
PGP fingerprint = 82A2 DC91 76E0 A10A 8ABB FD1B F404 27FA C7D1 15F3
Home |
Main Index |
Thread Index |
Old Index