Jon Peatfield wrote:
> On Fri, 26 May 2006, Connie Sieh wrote:
> 
>> I have included the "Upstream Vendor" release notes for Update 6 and
>> Update 7.
> 
> 
> I just noticed a minor thing with the 3.0.7 release (which my scripts 
> didn't pick up before because I'd not turned up the verbosity level high 
> enough)...
> 
> 3.0.5 contains nedit-5.4-1 but in 3.0.7 this seems to have reverted to 
> 5.3-4.1 (as the vendor shipped), though 30x/SRPMS/ contains the newer
> nedit-5.4-1.src.rpm package (well obviously)...
> 
>  From the release-notes:
> 
>> NEDIT
>>
>>        Nedit was found to have a bug preventing it from opening files.
>>        Rebuilding from source did not clear up the problem.  It also
>>        showed a comment that STRONGLY suggested just getting the binaries
>>        from nedit.org.  We followed this advice, and we packaged the
>>        binaries from nedit.org, into an rpm
>>
>>        nedit-5.4-1.i386.rpm
> 
> 
> but later it says:
> 
>> *As of Update 7
> 
> ...
> 
>> *       nedit-5.3-4.1.i386.rpm
> 
> ...
> 
> So I'm *assuming* that this is an accident rather than a deliberate 
> reversion.
> 
> Apologies for not spotting this before.  This is the only package (that 
> we install) that shows a lower revision for 3.0.7 than in 3.0.5
> 
> If it is intentional the comment about using nedit-5.4-1 should probably 
> be corrected.
> 

Hi Jon,
Well, better finding this out late than never.
The reason we did nedit 5.4, was because of a fairly irritating bug in 
the version RedHat distributed, which made it close to unusuable.  My 
quick and dirty fix was to package the binaries from nedit.org into an 
rpm, so if you look at that source rpm, they really are just binaries in 
a tar ball, with a web pointer to where the source is.
Anyway, nedit.org didn't have a x86_64 version, and the bug wasn't 
happening in that version anyway, so we went ahead with different version.
Well ... to make matters short, in one of our scripts that verifies 
everything, we managed to downgrade it and didn't see it until you 
pointed it out.  We now have it as one of our exceptions.
As much as I hate to do it, because it's not really a security errata, I 
think the 5.4 version is going to show up in the errata area for i386 5.0.7.
Once again, thanks for pointing this out.
Troy
-- 
__________________________________________________
Troy Dawson  [log in to unmask]  (630)840-6468
Fermilab  ComputingDivision/CSS  CSI Group
__________________________________________________