[OE-core] [PATCH] libc-common: Apply Debian-naming to libc debug symbols too

[Martin Jansa]

On Mon, Mar 19, 2012 at 11:52:36AM +0000, Mike Crowe wrote:
> Op 16 mrt. 2012, om 17:32 heeft Mike Crowe het volgende geschreven:
> > > The magic in rootfs_ipk.bbclass expects to generate debug package
> > > names by just appending '-dbg'.
> 
> It turns out that the above was a local change so that statement was
> incorrect. Nevertheless the change probably makes sense anyway.
> 
> > >                                 This breaks for eglibc because whilst
> > > the binary package has been renamed to libc6 the debug package is
> > > still called libc-dbg. This patch ensures that the debug package is
> > > renamed too.
> > > ---
> > > meta/classes/libc-common.bbclass |    1 +
> > > 1 files changed, 1 insertions(+), 0 deletions(-)
> > > 
> > > diff --git a/meta/classes/libc-common.bbclass b/meta/classes/libc-common.bbclass
> > > index 962f205..d0c1ab4 100644
> > > --- a/meta/classes/libc-common.bbclass
> > > +++ b/meta/classes/libc-common.bbclass
> > > @@ -27,4 +27,5 @@ python populate_packages_prepend () {
> > > 		bpn = d.getVar('BPN', True)
> > > 		d.setVar('PKG_'+bpn, 'libc6')
> > > 		d.setVar('PKG_'+bpn+'-dev', 'libc6-dev')
> > > +		d.setVar('PKG_'+bpn+'-dbg', 'libc6-dbg')
> > > }
>  
> On Fri, Mar 16, 2012 at 08:39:00PM +0100, Koen Kooi wrote:
> > Missing PR bump
> 
> On eglibc?
> 
> >                 and broken upgrade path.
> 
> I'm having trouble coming up with a way to solve that. I think the
> libc6-dbg package needs to RPROVIDE, RREPLACE and RCONFLICT with
> libc-dbg but if I add the following:
> 
>   d.setVar('RPROVIDES_' + bpn + '-dbg', bpn + '-dbg')
>   d.setVar('RCONFLICTS_' + bpn + '-dbg', bpn + '-dbg')
>   d.setVar('RREPLACES_' + bpn + '-dbg', bpn + '-dbg')

try to hardcode "libc" instead of bpn in value if you want

Provides: libc-dbg
Replaces: libc-dbg
Conflicts: libc-dbg


> 
> Then the package names undergo translation and I get:
> 
>   Provides: libc6-dbg
>   Replaces: libc6-dbg
>   Conflicts: libc6-dbg
> 
> Which is clearly wrong. :(
> 
> How would you recommend I provide backward compatibility for this
> change (assuming it is still seen as desirable?)
> 
> Thanks for your comments.
> 
> Mike.
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core at lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.openembedded.org/pipermail/openembedded-core/attachments/20120319/3b6b6cc3/attachment-0002.sig>