[Python-Dev] Cherry-pick between Python 3.4 RC2 and final? (original) (raw)

Nick Coghlan ncoghlan at gmail.com
Tue Mar 4 07:23:12 CET 2014

On 4 March 2014 14:20, Larry Hastings <larry at hastings.org> wrote:

On 03/03/2014 07:58 PM, Nick Coghlan wrote:

All of our development guides for testing against trunk are designed around running from a Mercurial checkout - it would really be whole lot easier for everyone else that is trying to test the release if you could just do a push from your release clone to a server side clone on hg.python.org (the link to create one is on http://hg.python.org/cpython/, and then it's just a hg push to publish a mirror of the exact state of your current clone).

I've pulled enough shenanigans over here (discarding and recreating the 3.4 branch, rebasing) that I'm really glad I haven't made the revisions public. And I'm not done, either, yet another really old revision has cropped up for cherry-picking.

The clones doesn't need to be updatable, we just need an easy way to do "hg clone " to get a release branch to test against.

It doesn't matter if hg pull doesn't work later - we just need you to make it as easy as possible for us to help you test things, because if there's not going to be an rc3 then we're running out of time to make sure SQL Alchemy and Flask actually work properly on Python 3.4 final (and I've been told the pkgutil bug that affected Flask may have affected Vim as well).

Let us deal with the Mercurial problems - you can rebase and cherry pick however the heck you want. But at the moment you're making it hard for people to test the release, and that is scaring me - our test coverage is a long way from 100%, so we need people running third party test suites on the pre-release to spot coverage gaps the way the Alembic and Flask test suites did.

Regards, Nick.

-- Nick Coghlan | ncoghlan at gmail.com | Brisbane, Australia

More information about the Python-Dev mailing list