[oe] [PATCH] clamav: allow clamav-data installation
Roman I Khimov
roman at khimov.ru
Sun Aug 22 15:58:04 UTC 2010
В сообщении от Суббота 21 августа 2010 22:55:18 автор Khem Raj написал:
> On Sat, Aug 21, 2010 at 11:19 AM, Roman I Khimov <roman at khimov.ru> wrote:
> > Technically, there are three possible installation combinations, each has
> > its valid use:
> > * clamav-data
> > * clamav-freshclam
> > * clamav-data & clamav-freshclam
> >
> > Before the patch only second one was possible (probably with the recent
> > BAD_RECOMMENDATIONS fixes to opkg and OE also the first one). After the
> > patch only first and third will be possible. Do we really care about
> > second case?
>
> Can clamav-freshclam function without clamav-data?
Yes.
> > And if yes, how to allow all of three to coexist?
>
> may be have separate recipe for data.
What about:
- RDEPENDS_${PN} = "${PN}-data"
- RDEPENDS_${PN}-daemon = "${PN}-data ${PN}-conf"
- RDEPENDS_${PN}-freshclam = "${PN}-conf"
- RRECOMMENDS_${PN} = "${PN}-freshclam"
- RRECOMMENDS_${PN}-daemon = "${PN}-freshclam"
- RPROVIDES_${PN}-freshclam = "${PN}-data"
+ RDEPENDS_${PN}-daemon = "${PN}-conf"
+ RDEPENDS_${PN}-freshclam = "${PN}-conf"
+ RRECOMMENDS_${PN} = "${PN}-data ${PN}-freshclam"
+ RRECOMMENDS_${PN}-daemon = "${PN}-data ${PN}-freshclam"
This way the default is third (both data and freshclam installed) and using
BAD_RECOMMENDATIONS one can disable one or another to implement first or
second.
The only downside is that you can technically put both to BAD_RECOMMENDATIONS
and clamav won't work without its databases (and you won't have freshclam to
get them). But then you have to be very persistent to get that effect.
--
http://roman.khimov.ru
mailto: roman at khimov.ru
gpg --keyserver hkp://subkeys.pgp.net --recv-keys 0xE5E055C3
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.openembedded.org/pipermail/openembedded-devel/attachments/20100822/c5c332a3/attachment-0002.sig>
More information about the Openembedded-devel
mailing list