On Wednesday 04 of April 2012 09:42:59 Cliff Perry wrote:
> On 04/04/2012 04:47 AM, Tomas Lestach wrote:
> > On Friday 30 of March 2012 17:06:17 Johannes Renner wrote:
> >> On 03/07/2012 11:57 AM, Johannes Renner wrote:
> >>> On 03/06/2012 05:42 PM, Johannes Renner wrote:
> >>>> Hey,
> >>>> 
> >>>> I want to propose a change concerning the names of cloned errata.
> >>>> Currently, names of cloned errata are generated like this:
> >>>> 
> >>>> String clonedAdvisoryName = "CL" +
> >>>> published.getAdvisoryName().substring(3); String clonedAdvisory = "CL"
> >>>> +
> >>>> published.getAdvisory().substring(3);
> >>>> 
> >>>> The reason for doing it like this might have been a small sized field
> >>>> in
> >>>> the database, but AFAIK we already upstreamed a patch to enlarge the
> >>>> respective field. Since cloned errata currently end up with rather
> >>>> strange names, I would like to propose to do the naming like this
> >>>> instead:
> >>>> 
> >>>> String clonedAdvisoryName = "CL-" + published.getAdvisoryName();
> >>>> String clonedAdvisory = "CL-" + published.getAdvisory();
> >>>> 
> >>>> Even the classification used for the Red Hat patches (E, S and B) gets
> >>>> lost
> >>>> when removing the first 3 characters, since names start with RH(E|S|B).
> >>>> So
> >>>> please tell me if I am missing any reason to keep the old naming
> >>>> scheme.
> >>>> 
> >>>> Attached please find my proposed patch for spacewalk master.
> >>> 
> >>> Of course this algorithm is duplicated in the perl parts of Spacewalk
> >>> ;-)
> >>> Patch is attached.
> >> 
> >> Just reactivating/raising this thread on the list since there is no
> >> comments yet, and I'm curious if you will accept the proposed changes or
> >> not.
> >> 
> >> Thanks,
> >> Johannes
> > 
> > Hello Johannes,
> > 
> > I am sorry, it took so long to respond.
> > 
> > You're absolutelly correct, that the erratum type information gets lost,
> > when replacing first 3 characters.
> > On the other hand I do not like the idea, the erratum clone name would be
> > 3
> > chars longer than the original name by default . There's a risk the clone
> > process would fail, because of clone name would go over the max errata
> > name
> > limit.
> > Thats' why I committed only the first idea of your proposed patch as:
> > 96f4e2cdeebc50d35b8a0e959506452eadb0aa26
> > 
> > Thank you.
> > 
> > Best Regards,
> 
> I think the CLSA CLBA format is more ideal and restores us to old behavior.
> 
> http://docs.redhat.com/docs/en-US/Red_Hat_Network_Satellite/5.4/html/Channel
> _Management_Guide/sect-Channel_Management_Guide-Custom_Channel_and_Package_M
> anagement-Manage_Software_Channel_Details.html
> 
> "By default, cloned errata inherit the label of the original Red Hat
> advisory with the "RH" prefix replaced with "CL". For example,
> RHSA-2003:324 becomes CLSA-2003:324. Subsequent clones of the same
> advisory have their second letters sequenced to denote their order, such
> as "CM" and "CN". "

Cool, this is the missing part ... because currently the 2nd clone would get 
the same name as the 1st one, what would fail the clone process.
I'll most probably implement this part.

Thanks Cliff!

Tomas
-- 
Tomas Lestach
RHN Satellite Engineering

> 
> I think though we lost the behavior of generating additional clones of
> the same errata as CM, CN over time, but to me is not a big deal.
> 
> Regards,
> Cliff
> 
> _______________________________________________
> Spacewalk-devel mailing list
> Spacewalk-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/spacewalk-devel
_______________________________________________
Spacewalk-devel mailing list
Spacewalk-devel@redhat.com
https://www.redhat.com/mailman/listinfo/spacewalk-devel

Reply via email to