It's time to ask that most dreaded and anticipated of questions: Are we
ready to release 2.2.0?

We've pushed out out second release candidate. We've had it available for
testing. Only one flaw has been found, and that can be integrated into the
release easily (and since it's a security flaw, it will be).

What say you, the community? Are you satisfied with what will be released
if we were to release right now? I'd like to make the release official, and
happen on Thursday, so speak before then!

--
Michael J. Pedersen
My Online Resume: http://www.icelus.org/ -- Google+ http://plus.ly/pedersen
Google Talk: m.peder...@icelus.org -- Twitter: pedersentg

Do we have any plan in how to manage dependencies required for TG2.2?

Until now we have been able to rely on the private index,
with stable release we will probably need a sprox0.8 and crank0.6.3
release on pypi.

I tried to get in touch with Chris by mail a few days ago, hope he is
able to give a look on time.

We should probably also do a version check of the various dependencies
we are placing on the private index.
Packages like Extremes, TurboJSON, Peak.Rules and so probably are not
used anymore and things like tgming can be upgraded to newer versions
(actually the 0.0.2 we have into the private index shouldn't work
anymore with 2.2)

Should we just remove all the packages which are not dependencies of
the TurboGears or tg.devtools packages from the private index? They
are usually installed from pypi and not from the index itself, so it
should be safe to just remove them and let people keep using pypi.

i have been building a project with tg2.2 and i haven't encountered any
bugs. the only problem i had with was tgext.crud and it was related to
tw2.forms which i was installing from pypi instead of our private index. in
short, i haven't encountered any bugs. if all the tests are passing let's
release it. :)

On Tue, Jul 17, 2012 at 5:43 AM, Alessandro Molina <

Unfortunately, our options are kind of slim here. Since I don't have the
ability to release a new sprox or a new crank, we can only check out the
code, make a private release, and upload it to our eggbasket.

I don't want to hold this release back any longer. We need it to happen, so
we may just have to bite the bullet on this one.

This part concerns me. As part of my process, I use yolk to find all the
active packages and get them upgraded to the highest level version. Your
statement here produces two questions:

1. Why would yolk have missed tgming?
2. Why do our tests currently pass?

As for the others, they are there because they were added during 2.1
testing. And, you're right, they should be removed. I'll spend some time
this week paring them down. If our tests don't make them a requirement,
*and* I have no other reason to have them in, I'll pull them out
(tgext.admin, for instance, will stay, even though we don't require it).

i have been building a project with tg2.2 and i haven't encountered any

Anybody else? Our issues right now are looking pretty minor and well
understood. Do we have a reason to hold back?

--
Michael J. Pedersen
My Online Resume: http://www.icelus.org/ -- Google+ http://plus.ly/pedersen
Google Talk: m.peder...@icelus.org -- Twitter: pedersentg

Chris just answered to my mail reporting an issue with some tests on
sprox, probably something broke due to another update to TW2. I'll
give a look at those and he will probably make the release soon after.