One option smart needs to conquer the world

[Jeff Johnson]

On Feb 22, 2006, at 1:19 PM, Gustavo Niemeyer wrote:

>> Hmmm, there's actually a better implementation of downgrades  
>> possible.
>>
>> What stresses users with downgrades is that vital or important
>> information might be lost forever.
>>
>> Repackaging the files that were present and saving makes downgrades
>> less risky.
>
> Yes, true.  I should have implemented support for repackaging/rolling
> back ages ago.  Not having time to implement those things is the main
> reason I haven't yet, which is a shame.
>

Couple lines of code -- either C or python -- to add  
RPMTRANS_FLAG_REPACKAGE, not hard.

The core issue is Chauvinism (which yum's pompous grandeur is preventing
certain script kiddies from noticing) is

     Who gets to set RPMTRANS_FLAG_REPACKAGE?

I claim that the end-user, not packagers or implementors, should be  
able to set
a global (and eventually enforcing, when I get around to it) flag to  
control the
repackaging mechanism in rpmlib.

Hence the zero-lines-of-depsolver-code implementation in rpm (which  
yum overrides, bad yum):

     echo "%_repackage_all_erasures        1" >> /etc/rpm/macros

BTW, this is the default erasure configuration in rpm-4.4.5-1 (at  
least my version).

Hey I flipped a coin, only end-users know heads or tails ;-)

73 de Jeff