duplicity-team team mailing list archive

[Michael Terry <michael.terry@xxxxxxxxxxxxx>]

> it's just the sheer amount of modified code that worries me a bit.
> as i wrote in the other email

Fair.  I tried to thoroughly test, but I'm human.  :)

> 1. old backend code e.g. released by somebody privately will not work
> anymore when dropped into duplicity/backends

This is a good point.  I'm curious how many people do that; I can easily imagine it happening.

As for your comments, Ken:

> So, if we wanted, we could move the modernizations over to 0.7 and
> fix 0.6.24 to work on older Pythons.

As much as I hate giving any thoughts to py2.4/2.5, this is probably a nice thing to do.  We clearly have interested users still on RHEL5, from my last survey on the mailing list.

If I have time in the next week, I can do this.  Else, if you tackle this, note that you shouldn't do the testing on a recently-released distro.  Because recent versions of tox/setuptools/virtualenv all dropped support for <2.6.  Maybe try Ubuntu 12.04?  I haven't tried an older distro myself yet, but ideally you'd be able to add "py24,py25" to the tox.ini file and just run tox to confirm that we work with older pythons.  If that doesn't work, you'd have to go to an earlier release of your distro of choice.
-- 
https://code.launchpad.net/~mterry/duplicity/backend-unification/+merge/216764
Your team duplicity-team is requested to review the proposed merge of lp:~mterry/duplicity/backend-unification into lp:duplicity.