increasing the python requirement

[Toshio Kuratomi]

On Tue, Jan 04, 2011 at 07:33:21PM +0100, Vincent Ladeuil wrote:
> >>>>> Jelmer Vernooij <jelmer at vernstok.nl> writes:
> 
>     > If bzr 2.4 introduced any format changes then not supporting Python 2.4
>     > would be a much bigger deal IMO, because users would have to upgrade to
>     > 2.4 to interact with repositories in the newer format. 
> 
> Not anymore since you mentioned it :)
> 
> Iff we introduced a new format, we'll also have to support it for
> python-2.4 [1]
> 
> We have plenty of time to discuss with the users involved. This includes
> mentioning that interoperability inside a project is based on the format
> used[2]. Add a smooth upgrade operation (still need work) and people have
> plenty of time to migrate.
> 
> Even if we have to do such focused backports, I think the corresponding
> effort is still worth the benefits of having to support only 2 different
> versions of python instead of 4.
> 
If you do backports of new formats to the version of bzr that supports
python-2.4 that strikes a great balance as far as I'm concerned.

I still have to keep bzr running on RHEL5 for people who use it there until
RHEL5 EOLs and the big thing is *not* performance optimizations, new
software that targets new APIs, and such... end users understand that those
types of things may require a newer version of python/bzr.  It really is just --
"I'm running RHEL5 and I need to interoperate with the repository format on
launchpad to checkout the latest version of gwibber" or "I'm running bzr
smart server on a RHEL5 server and a few of the clients pushing repos get
errors because they default to the new format."

-Toshio
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <https://lists.ubuntu.com/archives/bazaar/attachments/20110104/40289b3a/attachment.pgp>