[Mageia-dev] RPM5 AND MAGEIA

[Anne nicolas]

2011/3/8 Per Øyvind Karlsen <peroyvind at mandriva.org>:
> 2011/3/3 Buchan Milne <bgmilne at staff.telkomsa.net>:
>>
>> ----- "devzero2000" <devzero2000 at rpm5.org> wrote:
>>> Apart from the rest - of which i will ask for sponsorship when it
>>> will
>>> be - I wanted to know if there are plans to move to rpm5 by Mageia,
>>> such as Mandriva has been doing lately.
>>
>>
>>> Rpm5 already has a builtbot
>>> with Magela and rpm5. I can, if you can think useful or have plan for
>>> this, lay the necessary modification to enter into rpm5 Mageia, with
>>> the features of Mandriva cooker - fingerprint, syslog, etc. without
>>> trademark ecc- and produce a first rpm rpm5 for mageia , which also
>>> contains the functionality required by the passage to the "RPM
>>> ACID " feauture (berkeley db conversion)
>>
>> But, can you:
>> -ensure that all valid packages that build under rpm-4.x (e.g. in Mandriva 2010.x) will build under rpm5?
>> -ensure that all valid packages that install under rpm-4.x will install under rpm-4.x?
> No and no (I'm assuming you mean "install under rpm5 will install
> under rpm-4.x").
> Such guarantees has never been provided with any other rpm versions
> either and would effectively prevent the possibility of doing any
> serious development
> and improvement on rpm itself and packaging.
>
> There's a reason for having backports and why we don't even try aiming
> at such goals either.
>
> If able to give any such guarantees with rpm.org on Mageia you gotta be
> either stupid, insane or a damn liar! ;p
>
> The guarantees and priorities is as always:
> * legacy compatibility for older packages
> (opposed to future compatibility gets kinda hard with the the whole
> time travelling issue and limitations attached to it making future
> hard to reliably
> define;)
> *  backportability of current packages
> packages needs to be adapted to follow current policies, practice, functionality
> etc. in the current distribution, while efforts in ensuring
> possibility of backports
> needs to be invested in the packaging and adopting along the way rather than
> keep adapting rpm to stay compatible with the packaging which gets rather
> backwards.
>
> Very few changes results in breakage for backports, and where it happens it's
> easy enough to add conditional behaviour, nothing new forcing any real changes
> in long-established practices here..
> Much of the same breakages and issues you hit, you'll hit just as well in newer
> versions from rpm.org as well..
>>
>> There is no document specifying what has changed, or even when highlighting changes, no-one (@rpm5.org, or @mandriva.com) has bothered to list them so that contributors can save time instead of troubleshooting breakage.
>>
>> Some issues that have impacted me so far:
>> -changed behaviour of %exclude
> Ambiguity on %exclude usage is a clear bug, %exclude which is solely
> intended for
> excluding files from a specific package (rather than from being packaged at all.
> removing files at end of %install already fit this purpose
> sufficiently, which should
> make it obvious to most people with understanding of doing technical designs in
> general that wiring already existing functionality into an existing
> function with
> different functionality wouldn't make sense. Also this bug was fixed
> since in later
> releases such as 4.4.6 & 4.4.8 shipped before the rpm.org change, and should
> rather be treated as a regression.) predates the unpackaged files check and
> should *not* be used for other purposes.
> Fixing this is in packaging is *very* trivial and fully backwards
> compatible, not
> fixing this OTOH breaks compatibility.
>> -new reserved macros (%sql)
> all new macros introduced has the potential of conflicting with others
> and should
> always be fixed, it being reserved is more a benefit IMO as it prevents such
> incidents to go unnoticed (using very generic naming for macros is a bad
> practice in general anyways)..
> fixing this does not break any compatibility either ;)
>> -possible race condition between %__os_install_post and processing of %files (.lzma man pages reported missing where they are in fact .xz)
> your own packaging mistake independent of rpm version, explained on
> cooker and fixed for you already ;)
>>
>> (and of course, the unavailability of the build system - during one of the periods I had the most time to work on packages - due to the rpm5 "upgrade")
>>
>> rpm5 has wasted more than half of the time I could afford to contribute to Mandriva. It seems Mandriva has resources to waste, I don't think we have.
> you gotta put short-term and long-term effects up against eachother. breakages
> were already expected long before starting the upgrade, and the
> majority of these
> were actually rather in various tools etc. related to rpm rather than
> in rpm itself.
> The existing situation made it hard to maintain and do development of
> rpm in distribution,
> packaging and on a the various tools due to being left with since-long
> unmaintained
> tools used (ie. the older version of the perl bindings that only mandriva
> uses and that has been rewritten from scratch since and actively maintained
> upstream as well) and having to keep work around it and moving further
> and further away from "standard" rpm packaging by keep introducing any new
> functionality, scripts, macros etc. as distro specific and harder to collaborate
> with others on..
>
> You gotta break a few eggs..
> Issues hit in Mandriva gets fixed along the way in both cooker and upstream
> in parallel, making extremely few of them of any big concerns for other to
> worry about later.
> Maintenance and development of various tools, packaging etc. and dealing with
> your existing and future issues experienced is something you'll be left to deal
> with alone though..
> Considering the *major* amount of time and work invested in r&d historically
> always being on Mandriva's end with almost all developers employed to
> work on it full time. The harsh reality of trying to keep this up with only a
> few of these working on it during their limited spare time should be obvious..
> You're entitled to the freedom of not showing any interest in sharing efforts on
> any of these things (and for yourself to blame;), at least you're made aware of
> competence, skills, interest and resources that's been offered and is still
> available to you. :)
>
>
>>
>> (At present, I am not sure if I will continue to maintain packages in Mandriva, the ones where I need newer packages on non-Mandriva at work which I currently maintain in Mandriva and then rebuild I will maintain for the present, but ones I don't need for work may languish ...)
>
> (Sorry for slow reponse and late reply..:/)

Thanks for your inputs. Decision was taken some weeks ago and we will
follow it for now. You may have very good reasons on your side, please
respect ours.


-- 
Anne
http://www.mageia.org