* Finer granularity for RSUGGEST
@ 2008-09-22 20:55 Henning Heinold
2008-09-23 9:31 ` Robert Schuster
0 siblings, 1 reply; 4+ messages in thread
From: Henning Heinold @ 2008-09-22 20:55 UTC (permalink / raw)
To: openembedded-devel
[-- Attachment #1: Type: text/plain, Size: 622 bytes --]
Hi,
after fiddling with a filesystem kernel modul and his tools I noticed
in package_ipk.bbclass that if we set RSUGGESTS Variable the suggest field
in the ipk is set for all packages even -doc -dbg -dev and self defined
packages. I personal
think is not a good choice and way. In my example the kernel module only needs
to suggest the tools package. And with the old method the tools package suggest
it self.
I have prepared a patch for ipk and deb that let us set:
${PN}_RSUGGESTS = mooo
I hope the patch do not break the old behavior.
If I get positiv feedback I will check it in.
Bye Henning
[-- Attachment #2: package.patch --]
[-- Type: text/x-diff, Size: 2195 bytes --]
#
# old_revision [1cf701ebedabbcfca1fb1b88f2077258b1dfd000]
#
# patch "classes/package_deb.bbclass"
# from [a780e6f7dff8c1d720645d2b4f66b6f8e05d576d]
# to [962d33da2b1d4cb489270ad65397c5ce10adf8e7]
#
# patch "classes/package_ipk.bbclass"
# from [74c1d60c5b62a72c229c78bd65e5a0a0cf6486f1]
# to [acf85c751b2bba622bd398108d70a8bc3854354f]
#
============================================================
--- classes/package_deb.bbclass a780e6f7dff8c1d720645d2b4f66b6f8e05d576d
+++ classes/package_deb.bbclass 962d33da2b1d4cb489270ad65397c5ce10adf8e7
@@ -193,7 +193,7 @@ python do_package_deb () {
rdepends = [dep for dep in rdepends if not '*' in dep]
rrecommends = explode_deps(unicode(bb.data.getVar("RRECOMMENDS", localdata, 1) or ""))
rrecommends = [rec for rec in rrecommends if not '*' in rec]
- rsuggests = (unicode(bb.data.getVar("RSUGGESTS", localdata, 1) or "")).split()
+ rsuggests = (unicode(bb.data.getVar("RSUGGESTS", localdata, 1) or bb.data.getVar(pkgname + '_RSUGGESTS', localdata, 1) or "")).split()
rprovides = (unicode(bb.data.getVar("RPROVIDES", localdata, 1) or "")).split()
rreplaces = (unicode(bb.data.getVar("RREPLACES", localdata, 1) or "")).split()
rconflicts = (unicode(bb.data.getVar("RCONFLICTS", localdata, 1) or "")).split()
============================================================
--- classes/package_ipk.bbclass 74c1d60c5b62a72c229c78bd65e5a0a0cf6486f1
+++ classes/package_ipk.bbclass acf85c751b2bba622bd398108d70a8bc3854354f
@@ -229,7 +229,8 @@ python do_package_ipk () {
rdepends = explode_deps(bb.data.getVar("RDEPENDS", localdata, 1) or "")
rrecommends = explode_deps(bb.data.getVar("RRECOMMENDS", localdata, 1) or "")
- rsuggests = (bb.data.getVar("RSUGGESTS", localdata, 1) or "").split()
+ rsuggests = (bb.data.getVar("RSUGGESTS", localdata, 1) or bb.data.getVar(pkgname + '_RSUGGESTS', localdata, 1) or "").split()
+
rprovides = (bb.data.getVar("RPROVIDES", localdata, 1) or "").split()
rreplaces = (bb.data.getVar("RREPLACES", localdata, 1) or "").split()
rconflicts = (bb.data.getVar("RCONFLICTS", localdata, 1) or "").split()
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: Finer granularity for RSUGGEST
2008-09-22 20:55 Finer granularity for RSUGGEST Henning Heinold
@ 2008-09-23 9:31 ` Robert Schuster
2008-09-23 11:03 ` Koen Kooi
2008-09-25 19:35 ` Henning Heinold
0 siblings, 2 replies; 4+ messages in thread
From: Robert Schuster @ 2008-09-23 9:31 UTC (permalink / raw)
To: openembedded-devel
[-- Attachment #1: Type: text/plain, Size: 1058 bytes --]
Henning Heinold schrieb:
> Hi,
>
> after fiddling with a filesystem kernel modul and his tools I noticed
> in package_ipk.bbclass that if we set RSUGGESTS Variable the suggest field
> in the ipk is set for all packages even -doc -dbg -dev and self defined
> packages. I personal
> think is not a good choice and way. In my example the kernel module only needs
> to suggest the tools package. And with the old method the tools package suggest
> it self.
>
> I have prepared a patch for ipk and deb that let us set:
>
> ${PN}_RSUGGESTS = mooo
I do not think this is a good idea. You can have the same effect just
now by doing
RSUGGEST_${PN} = "mooo"
Currently if you set a variable that affects the binary package you know
it will end up in all of the binary packages declared in PACKAGES. E.g.
MAINTAINER, DESCRIPTION, LICENSE and so on.
You can easily prevent this by doing:
PACKAGES = "${PN} ${PN}-data"
LICENSE = "GPLv2"
LICENSE_${PN}-data = "CC-BY-SA"
Or am I missing an important point here?
Regards
Robert
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 268 bytes --]
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: Finer granularity for RSUGGEST
2008-09-23 9:31 ` Robert Schuster
@ 2008-09-23 11:03 ` Koen Kooi
2008-09-25 19:35 ` Henning Heinold
1 sibling, 0 replies; 4+ messages in thread
From: Koen Kooi @ 2008-09-23 11:03 UTC (permalink / raw)
To: openembedded-devel
Robert Schuster wrote:
> Henning Heinold schrieb:
>> Hi,
>>
>> after fiddling with a filesystem kernel modul and his tools I noticed
>> in package_ipk.bbclass that if we set RSUGGESTS Variable the suggest field
>> in the ipk is set for all packages even -doc -dbg -dev and self defined
>> packages. I personal
>> think is not a good choice and way. In my example the kernel module only needs
>> to suggest the tools package. And with the old method the tools package suggest
>> it self.
>>
>> I have prepared a patch for ipk and deb that let us set:
>>
>> ${PN}_RSUGGESTS = mooo
> I do not think this is a good idea. You can have the same effect just
> now by doing
>
> RSUGGEST_${PN} = "mooo"
>
> Currently if you set a variable that affects the binary package you know
> it will end up in all of the binary packages declared in PACKAGES. E.g.
> MAINTAINER, DESCRIPTION, LICENSE and so on.
>
> You can easily prevent this by doing:
>
> PACKAGES = "${PN} ${PN}-data"
Not related to Hennings point, put overwriting PACKAGES the way pictured
about will break out nice -dev and -dbg dependency changing (among other
things). Only overwrite PACKAGES if you fully understand the
consequences, which I think no OE developer that isn't called 'RP" does :)
regards,
Koen
>
> LICENSE = "GPLv2"
> LICENSE_${PN}-data = "CC-BY-SA"
>
> Or am I missing an important point here?
>
> Regards
> Robert
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: Finer granularity for RSUGGEST
2008-09-23 9:31 ` Robert Schuster
2008-09-23 11:03 ` Koen Kooi
@ 2008-09-25 19:35 ` Henning Heinold
1 sibling, 0 replies; 4+ messages in thread
From: Henning Heinold @ 2008-09-25 19:35 UTC (permalink / raw)
To: openembedded-devel
On Tue, Sep 23, 2008 at 11:31:09AM +0200, Robert Schuster wrote:
> RSUGGEST_${PN} = "mooo"
>
> Regards
> Robert
>
Ah,
thanks this is what I did not know. So forget the patch and maybee time
to add this to the oe manual.
Bye Henning
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2008-09-25 19:39 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-09-22 20:55 Finer granularity for RSUGGEST Henning Heinold
2008-09-23 9:31 ` Robert Schuster
2008-09-23 11:03 ` Koen Kooi
2008-09-25 19:35 ` Henning Heinold
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.