Two questions about the future

[jason pellerin]

1) Given that it has been 12 months since any update to the unittest2
plugins branch, should the plan for nose2 (that it should be just a
collection of unittest2 plugins) be reconsidered?

2) Since having stable/unstable split between google code and
bitbucket is not working well (IMO at least), and many of us are now
(or about to be) using git for most other projects, should we move
code hosting to github?

What do you think?

JP

[James Casbon]

On 29 November 2011 15:17, jason pellerin <jpel...@gmail.com> wrote:
> 1) Given that it has been 12 months since any update to the unittest2
> plugins branch, should the plan for nose2 (that it should be just a
> collection of unittest2 plugins) be reconsidered?

pass.

> 2) Since having stable/unstable split between google code and
> bitbucket is not working well (IMO at least), and many of us are now
> (or about to be) using git for most other projects, should we move
> code hosting to github?

yes, please.

[Thomas Kluyver]

On 29 November 2011 15:17, jason pellerin <jpel...@gmail.com> wrote:

2) Since having stable/unstable split between google code and
bitbucket is not working well (IMO at least), and many of us are now
(or about to be) using git for most other projects, should we move
code hosting to github?

I may have missed previous discussions on the list, but why not simply use one of the existing repositories for both (preferably bitbucket, IMHO)?

Thomas

[jason pellerin]

Github's feature set and the higher visibility of projects there would
be my reasons to pick github over bitbucket.

JP

[Thomas Kluyver]

On 29 November 2011 15:29, jason pellerin <jpel...@gmail.com> wrote:

Github's feature set and the higher visibility of projects there would
be my reasons to pick github over bitbucket.

I like Github as a site, but I find mercurial nicer to use as a VCS. Six of one, half a dozen of the other, although maybe the site is likely to improve quicker.

Thomas

[jason pellerin]

I feel the same way, but as I'm using git more and more at work and
for other projects, I'm finding it harder to switch back and forth. So
I like hg more in theory (for most things) but in practice I'm using
git anyway.

JP

[Matthew Brett]

Hi,

I haven't used bitbucket for development, but I've been very happy
with the github workflow.

Has anyone tried hg-git with github?

Best,

Matthew

[garyd]

On Wednesday, 30 November 2011 02:17:38 UTC+11, jason pellerin wrote:

1) Given that it has been 12 months since any update to the unittest2
plugins branch, should the plan for nose2 (that it should be just a
collection of unittest2 plugins) be reconsidered?

I don't know much about these plans, so forgive me if i say something ignorant. As I see it, the advantage of Nose is that it provides a different (IMHO superior) way to write and run tests in Python. I can see the benefit of re-using (part of) the standard implementation where appropriate (eg. if unittest2 provided a really good plugin/extensibility framework), but in the end some of our core goals (eg. target Python 2.4 and Jython; provide a more flexible and Pythonic API) seem to prevent us from being tied too closely to unittest2 (assuming that it continues to be developed).
 

2) Since having stable/unstable split between google code and
bitbucket is not working well (IMO at least), and many of us are now
(or about to be) using git for most other projects, should we move
code hosting to github?

I would prefer to use mercurial on bitbucket, but i don't actually mind

gary

[Philip Jenvey]

On Nov 29, 2011, at 7:17 AM, jason pellerin wrote:

> 2) Since having stable/unstable split between google code and
> bitbucket is not working well (IMO at least), and many of us are now
> (or about to be) using git for most other projects, should we move
> code hosting to github?

I generally prefer hg but I don't mind if you want to move to github.

+1 on no longer having a stable/unstable split.

--
Philip Jenvey

[Kumar McMillan]

Hey, sorry for the late reply! /me shakes fist at email

On Tue, Nov 29, 2011 at 5:16 PM, garyd <ga...@crucialfruit.com.au> wrote:

On Wednesday, 30 November 2011 02:17:38 UTC+11, jason pellerin wrote:

1) Given that it has been 12 months since any update to the unittest2

I don't know much about these plans, so forgive me if i say something ignorant. As I see it, the advantage of Nose is that it provides a different (IMHO superior) way to write and run tests in Python. I can see the benefit of re-using (part of) the standard implementation where appropriate (eg. if unittest2 provided a really good plugin/extensibility framework), but in the end some of our core goals (eg. target Python 2.4 and Jython; provide a more flexible and Pythonic API) seem to prevent us from being tied too closely to unittest2 (assuming that it continues to be developed).
 

Our idea for nose2 was simply that it would be a thin wrapper around unittest2 so that unittest2 feels like Nose.  This would allow developers to move to unittest2 without even really knowing it since we'd create a compatible command line interface and support existing Nose plugins.  I think this is possible but I'm not up to speed on Jason's current work on it. 

The nice thing about unittest2 is that it supports most of what Nose does out of the box and has extra features too.  It's well maintained by Michael Foord so I think it would lessen the maintenance burden on Nose by switching over.


 

2) Since having stable/unstable split between google code and
bitbucket is not working well (IMO at least), and many of us are now
(or about to be) using git for most other projects, should we move
code hosting to github?

A thousand times YES TO GITHUB.  I am annoyed with Google Code for so many reasons.  Sigh.  Git issues are pretty minimal but I think it would invite much more participation.  I think the cut over using the *existing* repo could actually be pretty easy.  Google Code supports git now so first thing we'd do is convert from hg to git then move all development to github.  This would make branching, code reviews, and pull requests sooooo much easier.  I've been using git for quite a while now and I love its work flow (doing these same things in Mercurial are really hard). For some tldr; I wrote up my preferred work flow w/ git: http://blog.mozilla.com/webdev/2011/11/21/git-using-topic-branches-and-interactive-rebasing-effectively/

-Kumar
 

[jason pellerin]

On Tue, Dec 13, 2011 at 2:15 PM, Kumar McMillan
<kumar.m...@gmail.com> wrote:

> A thousand times YES TO GITHUB.  I am annoyed with Google Code for so many
> reasons.  Sigh.  Git issues are pretty minimal but I think it would invite
> much more participation.  I think the cut over using the *existing* repo
> could actually be pretty easy.  Google Code supports git now so first thing
> we'd do is convert from hg to git then move all development to github.

Ok, sounds like we more or less agree. I just set up the github
project, https://github.com/jpellerin/nose and did an initial import
from google code hg.

I'll convert google code to git soon (not today probably but this
week) so we can easily pull from one to the other, and then we can
figure out how to migrate the open issues, and maybe set up an
organization to own the main repo.

JP

[Kumar McMillan]

On Tue, Dec 13, 2011 at 1:50 PM, jason pellerin <jpel...@gmail.com> wrote:

Ok, sounds like we more or less agree. I just set up the github
project, https://github.com/jpellerin/nose and did an initial import
from google code hg.

\o/
 

I'll convert google code to git soon (not today probably but this
week) so we can easily pull from one to the other, and then we can
figure out how to migrate the open issues, and maybe set up an
organization to own the main repo.

Awesome!
 

JP

--
You received this message because you are subscribed to the Google Groups "nose-dev" group.

To post to this group, send email to nose...@googlegroups.com.
To unsubscribe from this group, send email to nose-dev+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/nose-dev?hl=en.