From me at the-compiler.org  Tue Jun  2 09:12:39 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Tue, 2 Jun 2015 09:12:39 +0200
Subject: [pytest-dev] New pytest plugin: pytest-mccabe
Message-ID: <20150602071239.GB26357@tonks>

Hi,

(first of all, sorry for the cross-posting, but I can't decide on which
list this belongs - at least I resisted sending it to the TIP list
too :P)

I recently started switching from flake8[1] to pytest-pep8[2] and
pytest-flakes[3], but missed a way to run the mccabe[4] code
complexity checker.

So I wrote my first pytest plugin to integrate that with pytest:

https://github.com/The-Compiler/pytest-mccabe

Any feedback is welcome! I plan to write some more tests (for the
config wildcards mainly) and run pylint/pyflakes/pep8 over it, and
then I'll release v0.1 on PyPI.

Thanks to:

- Florian Schulze for pytest-flakes, on which the code is based.
- Ned Batchelder for the original McCabe script.
- Florent Xicluna for mccabe on PyPI.

Florian

[1] https://pypi.python.org/pypi/flake8
[2] https://pypi.python.org/pypi/pytest-pep8
[3] https://pypi.python.org/pypi/pytest-flakes
[4] https://pypi.python.org/pypi/mccabe

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150602/0455dc40/attachment.sig>

From bubenkoff at gmail.com  Tue Jun  2 10:41:12 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Tue, 02 Jun 2015 08:41:12 +0000
Subject: [pytest-dev] New pytest plugin: pytest-mccabe
In-Reply-To: <20150602071239.GB26357@tonks>
References: <20150602071239.GB26357@tonks>
Message-ID: <CACTnz+Fyh9WkW7QLCgU6tfe06D=8ZKNPAymKMNOpd_grpT9nwQ@mail.gmail.com>

there's a pylama project which is a bit buggy at error reporting, but it
solves the code checks as a whole, which i appreciate

https://github.com/klen/pylama

So it general solution for code checks but it's also a pytest plugin which
includes mccabe checks

I use it in several projects

On Tue, Jun 2, 2015 at 9:12 AM Florian Bruhin <me at the-compiler.org> wrote:

> Hi,
>
> (first of all, sorry for the cross-posting, but I can't decide on which
> list this belongs - at least I resisted sending it to the TIP list
> too :P)
>
> I recently started switching from flake8[1] to pytest-pep8[2] and
> pytest-flakes[3], but missed a way to run the mccabe[4] code
> complexity checker.
>
> So I wrote my first pytest plugin to integrate that with pytest:
>
> https://github.com/The-Compiler/pytest-mccabe
>
> Any feedback is welcome! I plan to write some more tests (for the
> config wildcards mainly) and run pylint/pyflakes/pep8 over it, and
> then I'll release v0.1 on PyPI.
>
> Thanks to:
>
> - Florian Schulze for pytest-flakes, on which the code is based.
> - Ned Batchelder for the original McCabe script.
> - Florent Xicluna for mccabe on PyPI.
>
> Florian
>
> [1] https://pypi.python.org/pypi/flake8
> [2] https://pypi.python.org/pypi/pytest-pep8
> [3] https://pypi.python.org/pypi/pytest-flakes
> [4] https://pypi.python.org/pypi/mccabe
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150602/695d722f/attachment.html>

From me at the-compiler.org  Tue Jun  2 10:48:27 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Tue, 2 Jun 2015 10:48:27 +0200
Subject: [pytest-dev] New pytest plugin: pytest-mccabe
In-Reply-To: <CACTnz+Fyh9WkW7QLCgU6tfe06D=8ZKNPAymKMNOpd_grpT9nwQ@mail.gmail.com>
References: <20150602071239.GB26357@tonks>
 <CACTnz+Fyh9WkW7QLCgU6tfe06D=8ZKNPAymKMNOpd_grpT9nwQ@mail.gmail.com>
Message-ID: <20150602084827.GC26357@tonks>

* Anatoly Bubenkov <bubenkoff at gmail.com> [2015-06-02 08:41:12 +0000]:
> there's a pylama project which is a bit buggy at error reporting, but it
> solves the code checks as a whole, which i appreciate
> 
> https://github.com/klen/pylama
> 
> So it general solution for code checks but it's also a pytest plugin which
> includes mccabe checks
> 
> I use it in several projects

I've seen that, but at least if I can believe the documentation it
can't run pylint for python 3:

    Pylint isn't supported in Python 3.

Also I really dislike its approach of bundling the checkers instead of
depending on them ;)

So far, pytest-{pep8,flakes,mccabe} do all I want :)

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150602/910cdbee/attachment.sig>

From bubenkoff at gmail.com  Tue Jun  2 10:49:26 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Tue, 02 Jun 2015 08:49:26 +0000
Subject: [pytest-dev] New pytest plugin: pytest-mccabe
In-Reply-To: <20150602084827.GC26357@tonks>
References: <20150602071239.GB26357@tonks>
 <CACTnz+Fyh9WkW7QLCgU6tfe06D=8ZKNPAymKMNOpd_grpT9nwQ@mail.gmail.com>
 <20150602084827.GC26357@tonks>
Message-ID: <CACTnz+Gp-_5-cWSqdCShnVO5705QmeYYM2NHHV-pW+ZVvcoJnw@mail.gmail.com>

docs lie, it works for python3 :)

On Tue, Jun 2, 2015 at 10:48 AM Florian Bruhin <me at the-compiler.org> wrote:

> * Anatoly Bubenkov <bubenkoff at gmail.com> [2015-06-02 08:41:12 +0000]:
> > there's a pylama project which is a bit buggy at error reporting, but it
> > solves the code checks as a whole, which i appreciate
> >
> > https://github.com/klen/pylama
> >
> > So it general solution for code checks but it's also a pytest plugin
> which
> > includes mccabe checks
> >
> > I use it in several projects
>
> I've seen that, but at least if I can believe the documentation it
> can't run pylint for python 3:
>
>     Pylint isn't supported in Python 3.
>
> Also I really dislike its approach of bundling the checkers instead of
> depending on them ;)
>
> So far, pytest-{pep8,flakes,mccabe} do all I want :)
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150602/a5b0014a/attachment.html>

From holger at merlinux.eu  Tue Jun  2 11:30:49 2015
From: holger at merlinux.eu (holger krekel)
Date: Tue, 2 Jun 2015 09:30:49 +0000
Subject: [pytest-dev] trainer needed for pytest training at EP2015
Message-ID: <20150602093049.GP29732@merlinux.eu>

Hello,

as i am giving a keynote at EuroPython 2015 in Bilbao and want to do a
little workshop in connection with it, I'd rather not do the pytest
training which also got accepted.  

Is anybody here interested to give the 3h pytest training using my
materials?  I'd then ask the EP organizers to wave the entrance fee
as it certainly requires some preparation still.

best,
holger
 

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From me at the-compiler.org  Wed Jun  3 11:31:48 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Wed, 3 Jun 2015 11:31:48 +0200
Subject: [pytest-dev] trainer needed for pytest training at EP2015
In-Reply-To: <20150602093049.GP29732@merlinux.eu>
References: <20150602093049.GP29732@merlinux.eu>
Message-ID: <20150603093148.GP26357@tonks>

Hi,

* holger krekel <holger at merlinux.eu> [2015-06-02 09:30:49 +0000]:
> Hello,
> 
> as i am giving a keynote at EuroPython 2015 in Bilbao and want to do a
> little workshop in connection with it, I'd rather not do the pytest
> training which also got accepted.  
> 
> Is anybody here interested to give the 3h pytest training using my
> materials?  I'd then ask the EP organizers to wave the entrance fee
> as it certainly requires some preparation still.

I think you're talking about [1], right?

I got into pytest as part of the "Adopt pytest month", so I'm rather
new to it - but I certainly know how to use it, write fixtures, and
recently wrote a plugin[2] (and contributed to others[3][4]) for it.

Looking at that overview, I'm confident I'd be able to do that, even
though I'm only using pytest for 1-2 months.

This would probably make the difference between coming and not coming
to EuroPython for me - I never was there before, and with the entrance
fee, I dismissed it as being too expensive for me[*], but looking at
the flight/hotel prices, I could afford that part.

So if there isn't someone else who'd take it and you think I could do
it, let's talk!

(Holger: If you prefer you can also query me on IRC, The-Compiler on
Freenode - tried to ping you there, but without any success)

Florian

[1] https://ep2015.europython.eu/conference/talks/pytest-simple-rapid-and-fun-testing-with-python
[2] https://github.com/The-Compiler/pytest-mccabe
[3] https://github.com/davehunt/pytest-html/pulls?q=is%3Apr+author%3AThe-Compiler
[4] https://github.com/pytest-dev/pytest-qt/pulls?q=is%3Apr+author%3AThe-Compiler
[*] I'm aware of the financial aid they're offering - but in theory I
    do have the money, I just don't want to spend that much as I'm
    saving for my upcoming studies - but at the same time, that makes
    me feel like others can use that support more and I shouldn't
    "exploit" it.

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/7838c6e5/attachment.sig>

From andreas at pelme.se  Wed Jun  3 11:43:31 2015
From: andreas at pelme.se (Andreas Pelme)
Date: Wed, 3 Jun 2015 11:43:31 +0200
Subject: [pytest-dev] trainer needed for pytest training at EP2015
In-Reply-To: <20150603093148.GP26357@tonks>
References: <20150602093049.GP29732@merlinux.eu> <20150603093148.GP26357@tonks>
Message-ID: <E656ADAE-F5FC-4050-AA6B-BDD1C2428ECE@pelme.se>

Hi Florian,

> On 3 jun 2015, at 11:31, Florian Bruhin <me at the-compiler.org> wrote:
> 
> So if there isn't someone else who'd take it and you think I could do
> it, let's talk!

I helped out last year and it was a lot of fun. I have not finalized my plans for EP and am not yet sure I will be able to make it. If I?m there, I will for sure be able help out and assist as much as I can. I?m happy if you take it on and if it works out paying your ticket that would be great. I will probably not be able to help with any preparations before EP, but during EP I can help out as much as needed. I should finalize my plans for EP in the coming week and will let you know whether I will make it or not.

Holger, Floris and Brianna: I?m really excited that you?re keynoting, giving a talk on the pytest hooks and the Adopt pytest month! The EP session list looks great :)

Cheers,
Andreas

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 204 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/32a8cb0b/attachment.sig>

From holger at merlinux.eu  Wed Jun  3 16:20:52 2015
From: holger at merlinux.eu (holger krekel)
Date: Wed, 3 Jun 2015 14:20:52 +0000
Subject: [pytest-dev] moving to git / issues?
Message-ID: <20150603142052.GV29732@merlinux.eu>

Hi all,

i we want to move pytest to github what would we do with the
issues?  Can someone experiment with migrating the issues to a github
repo (at some user location, not pytest-dev)?  A quick search revealed
https://github.com/vbabiy/bitbucket_issue_migration which might help.

best,

holger

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From bubenkoff at gmail.com  Wed Jun  3 16:22:31 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Wed, 03 Jun 2015 14:22:31 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150603142052.GV29732@merlinux.eu>
References: <20150603142052.GV29732@merlinux.eu>
Message-ID: <CACTnz+EzYEFufspZ_KD4HxQ7AgdBzCiuw3pLQ0Lh-LOEiWbc+g@mail.gmail.com>

wow, i didn't expect such a topic from Holger!
I'll check what's available as well

On Wed, Jun 3, 2015 at 4:21 PM holger krekel <holger at merlinux.eu> wrote:

> Hi all,
>
> i we want to move pytest to github what would we do with the
> issues?  Can someone experiment with migrating the issues to a github
> repo (at some user location, not pytest-dev)?  A quick search revealed
> https://github.com/vbabiy/bitbucket_issue_migration which might help.
>
> best,
>
> holger
>
> --
> about me:    http://holgerkrekel.net/about-me/
> contracting: http://merlinux.eu
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/6b36a847/attachment.html>

From holger at merlinux.eu  Wed Jun  3 16:28:59 2015
From: holger at merlinux.eu (holger krekel)
Date: Wed, 3 Jun 2015 14:28:59 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CACTnz+EzYEFufspZ_KD4HxQ7AgdBzCiuw3pLQ0Lh-LOEiWbc+g@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu>
 <CACTnz+EzYEFufspZ_KD4HxQ7AgdBzCiuw3pLQ0Lh-LOEiWbc+g@mail.gmail.com>
Message-ID: <20150603142854.GW29732@merlinux.eu>

On Wed, Jun 03, 2015 at 14:22 +0000, Anatoly Bubenkov wrote:
> wow, i didn't expect such a topic from Holger!
> I'll check what's available as well

I was kind of -0 on the idea so far, maybe +0 now.
Ronny and Floris also seem to be >=+0 if i get it right.
So if someone solves the issues issue it might happen :)

holger

> On Wed, Jun 3, 2015 at 4:21 PM holger krekel <holger at merlinux.eu> wrote:
> 
> > Hi all,
> >
> > i we want to move pytest to github what would we do with the
> > issues?  Can someone experiment with migrating the issues to a github
> > repo (at some user location, not pytest-dev)?  A quick search revealed
> > https://github.com/vbabiy/bitbucket_issue_migration which might help.
> >
> > best,
> >
> > holger
> >
> > --
> > about me:    http://holgerkrekel.net/about-me/
> > contracting: http://merlinux.eu
> > _______________________________________________
> > pytest-dev mailing list
> > pytest-dev at python.org
> > https://mail.python.org/mailman/listinfo/pytest-dev
> >

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From nicoddemus at gmail.com  Wed Jun  3 16:34:19 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Wed, 03 Jun 2015 14:34:19 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CACTnz+EzYEFufspZ_KD4HxQ7AgdBzCiuw3pLQ0Lh-LOEiWbc+g@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu>
 <CACTnz+EzYEFufspZ_KD4HxQ7AgdBzCiuw3pLQ0Lh-LOEiWbc+g@mail.gmail.com>
Message-ID: <CA+RQFfcL+8=7HQvVj1d+v6iZH3=BzzGROOa0HZQqk2zU4BWuvA@mail.gmail.com>

On Wed, Jun 3, 2015 at 11:22 AM Anatoly Bubenkov bubenkoff at gmail.com
<http://mailto:bubenkoff at gmail.com> wrote:

wow, i didn't expect such a topic from Holger!
> I'll check what's available as well
>
Nice! :)

Anatoly, let me know if I can help in any way.

Cheers,


> On Wed, Jun 3, 2015 at 4:21 PM holger krekel <holger at merlinux.eu> wrote:
>
>> Hi all,
>>
>> i we want to move pytest to github what would we do with the
>> issues?  Can someone experiment with migrating the issues to a github
>> repo (at some user location, not pytest-dev)?  A quick search revealed
>> https://github.com/vbabiy/bitbucket_issue_migration which might help.
>>
>> best,
>>
>> holger
>>
>> --
>> about me:    http://holgerkrekel.net/about-me/
>> contracting: http://merlinux.eu
>> _______________________________________________
>> pytest-dev mailing list
>> pytest-dev at python.org
>> https://mail.python.org/mailman/listinfo/pytest-dev
>>
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/a7116300/attachment.html>

From me at the-compiler.org  Wed Jun  3 16:54:14 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Wed, 3 Jun 2015 16:54:14 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150603142052.GV29732@merlinux.eu>
References: <20150603142052.GV29732@merlinux.eu>
Message-ID: <20150603145414.GU26357@tonks>

* holger krekel <holger at merlinux.eu> [2015-06-03 14:20:52 +0000]:
> i we want to move pytest to github what would we do with the
> issues?  Can someone experiment with migrating the issues to a github
> repo (at some user location, not pytest-dev)?  A quick search revealed
> https://github.com/vbabiy/bitbucket_issue_migration which might help.

Glad to hear!

I started to play a bit here:
https://github.com/The-Compiler/pytest-unofficial

So far, not that much success:

    [...]
    Created: py.test does not cooperate with twisted's trial TestCase [2 comments]
    Created 762 issues
    Traceback (most recent call last):
      File "migrate.py", line 304, in <module>
        push_issue(gh_username, gh_repository, issue, body, comments)
      File "migrate.py", line 266, in push_issue
        format_comment(comment),
      File "migrate.py", line 122, in format_comment
        comment['user'].encode('utf-8')
    UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 91: ordinal not in range(128)

Time to fix that script then! ;)

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/c83051bb/attachment-0001.sig>

From nicoddemus at gmail.com  Wed Jun  3 17:37:24 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Wed, 03 Jun 2015 15:37:24 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150603145414.GU26357@tonks>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
Message-ID: <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>

I forked the script and fixed that particular issue:
https://github.com/nicoddemus/bitbucket_issue_migration/

Right now I'm hitting another error around 8th issue:

Created 8 of 762 issues
Traceback (most recent call last):
  File "migrate.py", line 302, in <module>
    print "Comments", [comment['body'] for comment in comments]
  File "migrate.py", line 265, in push_issue
    gh_repository
  File
"X:\temp\bitbucket_issue_migration\.env27\lib\site-packages\pygithub3\services\issues\comments.py",
line 55, in create
    return self._post(request)
  File
"X:\temp\bitbucket_issue_migration\.env27\lib\site-packages\pygithub3\services\base.py",
line 139, in _post
    response = self._client.post(request, data=input_data, **kwargs)
  File
"X:\temp\bitbucket_issue_migration\.env27\lib\site-packages\pygithub3\core\client.py",
line 89, in post
    response = self.request('post', request, **kwargs)
  File
"X:\temp\bitbucket_issue_migration\.env27\lib\site-packages\pygithub3\core\client.py",
line 71, in wrapper
    return func(self, verb, request, **kwargs)
  File
"X:\temp\bitbucket_issue_migration\.env27\lib\site-packages\pygithub3\core\client.py",
line 80, in request
    GithubError(response).process()
  File
"X:\temp\bitbucket_issue_migration\.env27\lib\site-packages\pygithub3\core\errors.py",
line 36, in process
    self.response.raise_for_status()
  File
"D:\Shared\dist\12.0-all\requests-2.5.1\lib\site-packages\requests\models.py",
line 831, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 403 Client Error: Forbidden

I will continue investigating, but any help here is welcome.

Btw, what you guys think about concentrating the issue migration effort in
a single place? I can add Florian and Anatoly and anyone else interested as
collaborator of the fork, and we can open a PR to upstream once the script
is ready.




On Wed, Jun 3, 2015 at 11:54 AM Florian Bruhin <me at the-compiler.org> wrote:

> * holger krekel <holger at merlinux.eu> [2015-06-03 14:20:52 +0000]:
> > i we want to move pytest to github what would we do with the
> > issues?  Can someone experiment with migrating the issues to a github
> > repo (at some user location, not pytest-dev)?  A quick search revealed
> > https://github.com/vbabiy/bitbucket_issue_migration which might help.
>
> Glad to hear!
>
> I started to play a bit here:
> https://github.com/The-Compiler/pytest-unofficial
>
> So far, not that much success:
>
>     [...]
>     Created: py.test does not cooperate with twisted's trial TestCase [2
> comments]
>     Created 762 issues
>     Traceback (most recent call last):
>       File "migrate.py", line 304, in <module>
>         push_issue(gh_username, gh_repository, issue, body, comments)
>       File "migrate.py", line 266, in push_issue
>         format_comment(comment),
>       File "migrate.py", line 122, in format_comment
>         comment['user'].encode('utf-8')
>     UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position
> 91: ordinal not in range(128)
>
> Time to fix that script then! ;)
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/3420d5e8/attachment.html>

From me at the-compiler.org  Wed Jun  3 17:51:18 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Wed, 3 Jun 2015 17:51:18 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
Message-ID: <20150603155118.GV26357@tonks>

* Bruno Oliveira <nicoddemus at gmail.com> [2015-06-03 15:37:24 +0000]:
> I forked the script and fixed that particular issue:
> https://github.com/nicoddemus/bitbucket_issue_migration/

Thanks! I did almost the same locally, including moving the password
input up :D

> Right now I'm hitting another error around 8th issue:
> 
> [...]
> 
> I will continue investigating, but any help here is welcome.

Same here - I printed the response and it's actually this:

    You have triggered an abuse detection mechanism and have been
    temporarily blocked from content creation. Please retry your
    request again later.

    https://developer.github.com/v3#abuse-rate-limits

Since that documentation says:

    It is not intended for this rate limit to interfere with any
    legitimate use of the API. Your normal rate limits should be the
    only limit you target. Please contact support if your use is
    affected by this rate limit.

I now contacted GitHub support.

I bet it's because of that unicode arrow (\u2192, ?) in the comment...

> Btw, what you guys think about concentrating the issue migration effort in
> a single place? I can add Florian and Anatoly and anyone else interested as
> collaborator of the fork, and we can open a PR to upstream once the script
> is ready.

For the script, or for a playground to do the migration?

+1 for fixes to the script, but for the playground I'd rather try with
my own repo, otherwise there'll be a mess if we both try to do
something ;)

Could you maybe look at migrating the repo itself and (if needed)
merge requests? Then I'll continue to look at the issues.

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/1a2f3a7e/attachment.sig>

From holger at merlinux.eu  Wed Jun  3 17:52:18 2015
From: holger at merlinux.eu (holger krekel)
Date: Wed, 3 Jun 2015 15:52:18 +0000
Subject: [pytest-dev] Test support for Python micro versions?
In-Reply-To: <20150524223607.Horde.76Icfy3dpqD0Ad0eW7_Bbg7@webmail.in-berlin.de>
References: <20150524223607.Horde.76Icfy3dpqD0Ad0eW7_Bbg7@webmail.in-berlin.de>
Message-ID: <20150603155218.GD29732@merlinux.eu>

Hi Dinu,

On Sun, May 24, 2015 at 22:36 +0200, Dinu Gherman wrote:
> Hello!
> 
> I'm looking for information about how to run py.test (or rather tox)
> on micro versions of Python releases (the micro version being part C
> in the full versioning scheme A.B.C, i.e. major.minor.micro, see
> also https://docs.python.org/devguide/devcycle.html).
> 
> I've put a well-known search engine at work and have searched on
> pytest.org, but haven't been really successful. Is this a
> non-feature or just considered a non-issue? The latter would be
> surprising to me as Python micro releases have recently shown what
> I'd call significant changes, e.g. those concerning SSL between
> 2.7.8 and 2.7.9.
> 
> So AFAICT it would be helpful to be able to run something like this:
> 
>   tox -e py278 py279 py2710
> 
> Any clues are highly appreciated, thanks!

I think you have to use the "basepython" setting, e.g.
"basepython=python2.7.8" and then maybe have a symlink
to the correct installation.  As it stands, tox has
no direct default support for distinguishing micro versions.

holger



> Cheers,
> 
> Dinu
> 
> 
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
> 

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From holger at merlinux.eu  Wed Jun  3 17:59:04 2015
From: holger at merlinux.eu (holger krekel)
Date: Wed, 3 Jun 2015 15:59:04 +0000
Subject: [pytest-dev] Fixture override not working
In-Reply-To: <CA+rtQkqBPY0u=86AuffRnNLmhPtbiQPBmkJu+kQLx0U-K-krhQ@mail.gmail.com>
References: <CA+rtQkqBPY0u=86AuffRnNLmhPtbiQPBmkJu+kQLx0U-K-krhQ@mail.gmail.com>
Message-ID: <20150603155904.GF29732@merlinux.eu>

Hi Ken,

On Wed, May 13, 2015 at 15:42 -0400, Ken Petti wrote:
> Hello all,
> 
> Last week I spoke to Ronny on the IRC channel about a fixture issue, and he
> recommended I send out an email to the larger group. Here is a gist of the
> issuehttps://gist.github.com/citizenken/861055b8778c9cce343a
> 
> I have a fixture on the module level that I want to override on a per-test
> basis. This module fixture handles app log in and is both autoused and
> function scoped. I override it with a tuple value, to use non-default
> credentials, and then use the new fixture in an alternate login fixture.

hum, haven't used myself a setup like this yet.

> What I expect to happen is that the module fixture it overridden with my
> static values, the alternate login fixture then uses those static values
> and logs in, then control goes to my tests, which executes as expected.
> What is happening is that the module fixture is never overridden, so the
> app logs in with the wrong credentials and the test fails.

Why do you want to override the module level fixture instead of
creating a new function level fixture if i may ask?
(overriding fixtures should be rarely done IMO and not sure
about modifying scope at the same time ...).

> Ronny suggested modifying my gist and using hidden fixtures like so:
> https://gist.github.com/RonnyPfannschmidt/3c17b07aa56d6231d38d

Besides, is there a reason you use "module_conftest.py" instead of
"conftest.py"?  I'd think that "module_conftest.py" is not used
at all unless you import it from "conftest.py"  or so.

holger

> This did not work. Instead, the hidden fixture was executed with the static
> values, then, without closing the browser, the original login_by_account
> fixture is called.
> 
> 
> 
> This is an excerpt from my chat with Ronny:
> [13:25] <ronny> citken: i think you missunderstood me, the idea is to
> separate th data fixture and the autouse fixture, since it seems all
> autouse instaces are made n all cases
> [13:27] <citken> ronny: can you show me an example/comment on the gist what
> i should do?
> [13:28] <ronny> citken:
> https://gist.github.com/RonnyPfannschmidt/3c17b07aa56d6231d38d
> [13:30] <citken> ok let me try that now
> [13:30] <citken> thanks, that makes more sense
> [13:31] <ronny> citken: could you please repost an issue about autouse
> fixture overriding issues
> [13:34] <citken> yeah that isn't working either
> [13:34] <citken> my fixture is overriding for all test cases
> [13:35] <citken> it should only override for the testcase that is marked
> parameterize
> [13:35] <citken> making the hidden fixture, as in your gist, does not work
> correctly
> [13:43] <ronny> then im missing something
> [13:44] <ronny> citken: what is the problem with the hidden fixture/
> [14:01] <citken> ronny: so i don't think your gist is handling the override
> correctly
> [14:02] <ronny> citken: im probably missing something, what exactly is
> happening how?
> [14:04] <citken> its weird... it looks like _ensure_login is hit first
> (expected, with autouse=True), and that has the value of my override. My
> app logs in with the override credentials. Then, without closing the
> browser or anything, the original login_by_account fixture is called, the
> one with the scope of "function"
> [14:04] <citken> is the stack trace like this?:
> [14:05] <citken> override -> autouse fixture -> function scope fixture ->
> test?
> [14:07] <ronny> citken: not sure, i dont know the exact details, holger and
> bubenkoff worked with fixture oerrides in-depth, and currently im at the
> end of what i can help with
> 
> 
> 
> 
> If anyone could give me a hand that would be great! Thanks!
> -- 
> *Ken Petti*
> Software Development Engineer in Test
> E-Mail: ken.petti at cloudlock.com <uche at cloudlock.com>
>  <http://www.cloudlock.com/>
> <http://www.twitter.com/cloudlock>  <http://www.facebook.com/cloudlock>
> <http://www.linkedin.com/company/cloudlock>
> <https://plus.google.com/106755360058827077409/posts>
> *CloudLock provides compliance and security solutions for enterprises using
> public cloud applications like Salesforce and Google Apps. The largest
> organizations in the world trust CloudLock to secure their data*.

> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev


-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From nicoddemus at gmail.com  Wed Jun  3 17:59:17 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Wed, 03 Jun 2015 15:59:17 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150603155118.GV26357@tonks>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
Message-ID: <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>

> For the script, or for a playground to do the migration?

I meant for the script... I agree that everyone should use their own
playground for testing it. Added you and Anatoly as collaborators, feel
free to commit any fixes there as you find it.

Once the script to migrate the issues is stable, we can discuss the next
steps in detail.

On Wed, Jun 3, 2015 at 12:51 PM Florian Bruhin <me at the-compiler.org> wrote:

> * Bruno Oliveira <nicoddemus at gmail.com> [2015-06-03 15:37:24 +0000]:
> > I forked the script and fixed that particular issue:
> > https://github.com/nicoddemus/bitbucket_issue_migration/
>
> Thanks! I did almost the same locally, including moving the password
> input up :D
>
> > Right now I'm hitting another error around 8th issue:
> >
> > [...]
> >
> > I will continue investigating, but any help here is welcome.
>
> Same here - I printed the response and it's actually this:
>
>     You have triggered an abuse detection mechanism and have been
>     temporarily blocked from content creation. Please retry your
>     request again later.
>
>     https://developer.github.com/v3#abuse-rate-limits
>
> Since that documentation says:
>
>     It is not intended for this rate limit to interfere with any
>     legitimate use of the API. Your normal rate limits should be the
>     only limit you target. Please contact support if your use is
>     affected by this rate limit.
>
> I now contacted GitHub support.
>
> I bet it's because of that unicode arrow (\u2192, ?) in the comment...
>
> > Btw, what you guys think about concentrating the issue migration effort
> in
> > a single place? I can add Florian and Anatoly and anyone else interested
> as
> > collaborator of the fork, and we can open a PR to upstream once the
> script
> > is ready.
>
> For the script, or for a playground to do the migration?
>
> +1 for fixes to the script, but for the playground I'd rather try with
> my own repo, otherwise there'll be a mess if we both try to do
> something ;)
>
> Could you maybe look at migrating the repo itself and (if needed)
> merge requests? Then I'll continue to look at the issues.
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/dfc6b282/attachment.html>

From me at the-compiler.org  Wed Jun  3 18:01:27 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Wed, 3 Jun 2015 18:01:27 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
Message-ID: <20150603160127.GW26357@tonks>

* Bruno Oliveira <nicoddemus at gmail.com> [2015-06-03 15:59:17 +0000]:
> > For the script, or for a playground to do the migration?
> 
> I meant for the script... I agree that everyone should use their own
> playground for testing it. Added you and Anatoly as collaborators, feel
> free to commit any fixes there as you find it.
> 
> Once the script to migrate the issues is stable, we can discuss the next
> steps in detail.

Can you please enable issues (so meta!) for the repo?

Then I suggest to continue discussions there, so people interested in
it can watch the repo, and everyone else doesn't get annoyed by the
mails on the ML :)

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/02187eb7/attachment.sig>

From nicoddemus at gmail.com  Wed Jun  3 18:12:04 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Wed, 03 Jun 2015 16:12:04 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150603160127.GW26357@tonks>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
Message-ID: <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>

> Can you please enable issues (so meta!) for the repo?

Done! :)

> Then I suggest to continue discussions there, so people interested in
> it can watch the repo, and everyone else doesn't get annoyed by the
> mails on the ML :)

Sure! Everyone interested is welcome to join.

On Wed, Jun 3, 2015 at 1:01 PM Florian Bruhin <me at the-compiler.org> wrote:

> * Bruno Oliveira <nicoddemus at gmail.com> [2015-06-03 15:59:17 +0000]:
> > > For the script, or for a playground to do the migration?
> >
> > I meant for the script... I agree that everyone should use their own
> > playground for testing it. Added you and Anatoly as collaborators, feel
> > free to commit any fixes there as you find it.
> >
> > Once the script to migrate the issues is stable, we can discuss the next
> > steps in detail.
>
> Can you please enable issues (so meta!) for the repo?
>
> Then I suggest to continue discussions there, so people interested in
> it can watch the repo, and everyone else doesn't get annoyed by the
> mails on the ML :)
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/ab486478/attachment.html>

From bubenkoff at gmail.com  Wed Jun  3 18:27:09 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Wed, 03 Jun 2015 16:27:09 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfcL+8=7HQvVj1d+v6iZH3=BzzGROOa0HZQqk2zU4BWuvA@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu>
 <CACTnz+EzYEFufspZ_KD4HxQ7AgdBzCiuw3pLQ0Lh-LOEiWbc+g@mail.gmail.com>
 <CA+RQFfcL+8=7HQvVj1d+v6iZH3=BzzGROOa0HZQqk2zU4BWuvA@mail.gmail.com>
Message-ID: <CACTnz+EvC=Q=_hCUAmh8o3NRgpLv3-AaZi07-PAzL1OynuFcJw@mail.gmail.com>

You are so fast that i'll be late i think
I'll sit tonight to check the issues on that fork

On 16:34, Wed, Jun 3, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:

> On Wed, Jun 3, 2015 at 11:22 AM Anatoly Bubenkov bubenkoff at gmail.com
> <http://mailto:bubenkoff at gmail.com> wrote:
>
> wow, i didn't expect such a topic from Holger!
>> I'll check what's available as well
>>
> Nice! :)
>
> Anatoly, let me know if I can help in any way.
>
> Cheers,
>
>
>> On Wed, Jun 3, 2015 at 4:21 PM holger krekel <holger at merlinux.eu> wrote:
>>
>>> Hi all,
>>>
>>> i we want to move pytest to github what would we do with the
>>> issues?  Can someone experiment with migrating the issues to a github
>>> repo (at some user location, not pytest-dev)?  A quick search revealed
>>> https://github.com/vbabiy/bitbucket_issue_migration which might help.
>>>
>>> best,
>>>
>>> holger
>>>
>>> --
>>> about me:    http://holgerkrekel.net/about-me/
>>> contracting: http://merlinux.eu
>>> _______________________________________________
>>> pytest-dev mailing list
>>> pytest-dev at python.org
>>> https://mail.python.org/mailman/listinfo/pytest-dev
>>>
>> _______________________________________________
>> pytest-dev mailing list
>> pytest-dev at python.org
>> https://mail.python.org/mailman/listinfo/pytest-dev
>>
> ?
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150603/ba11ad7f/attachment-0001.html>

From holger at merlinux.eu  Wed Jun  3 20:48:39 2015
From: holger at merlinux.eu (holger krekel)
Date: Wed, 3 Jun 2015 18:48:39 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
Message-ID: <20150603184839.GI29732@merlinux.eu>

On Wed, Jun 03, 2015 at 16:12 +0000, Bruno Oliveira wrote:
> > Can you please enable issues (so meta!) for the repo?
> 
> Done! :)
> 
> > Then I suggest to continue discussions there, so people interested in
> > it can watch the repo, and everyone else doesn't get annoyed by the
> > mails on the ML :)
> 
> Sure! Everyone interested is welcome to join.

For the record, i don't mind discussions here but am also happy
for discussions to go on in that repo.  Thanks all for caring!

holger

> On Wed, Jun 3, 2015 at 1:01 PM Florian Bruhin <me at the-compiler.org> wrote:
> 
> > * Bruno Oliveira <nicoddemus at gmail.com> [2015-06-03 15:59:17 +0000]:
> > > > For the script, or for a playground to do the migration?
> > >
> > > I meant for the script... I agree that everyone should use their own
> > > playground for testing it. Added you and Anatoly as collaborators, feel
> > > free to commit any fixes there as you find it.
> > >
> > > Once the script to migrate the issues is stable, we can discuss the next
> > > steps in detail.
> >
> > Can you please enable issues (so meta!) for the repo?
> >
> > Then I suggest to continue discussions there, so people interested in
> > it can watch the repo, and everyone else doesn't get annoyed by the
> > mails on the ML :)
> >
> > Florian
> >
> > --
> > http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
> >    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
> >          I love long mails! | http://email.is-not-s.ms/
> >

> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev


-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From holger at merlinux.eu  Wed Jun  3 23:53:05 2015
From: holger at merlinux.eu (holger krekel)
Date: Wed, 3 Jun 2015 21:53:05 +0000
Subject: [pytest-dev] merging PR freedom
Message-ID: <20150603215305.GM29732@merlinux.eu>

Hi Ronny and all other commiters :)

i realize that from time to time you only "approve" a PR but don't
merge it.  Please feel more free to just merge.  But please don't
forget adding a CHANGELOG entry and adding the PR person to the
"authors" file.  Of course if you are fine with a PR but want to
have others chime in, then only approving is fine.

cheers,
holger

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From nicoddemus at gmail.com  Fri Jun  5 05:06:35 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Fri, 05 Jun 2015 03:06:35 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150603184839.GI29732@merlinux.eu>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
Message-ID: <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>

After a few changes to the script, I have migrated all issues to a
temporary repository at:
https://github.com/nicoddemus/pytest-issues-migration.

Moved issues retain:
- body
- creation date
- comments
- kind and component (added as labels)

Issue and comment authors are added to each issue or comment as a note in
the body, but the issue author will always be the person that ran the
script (this is a limitation of the GitHub import API). Is this a problem?
When doing the final migration, which user should we use?

Any feedback on how we can improve the import process?




On Wed, Jun 3, 2015 at 3:48 PM holger krekel <holger at merlinux.eu> wrote:

> On Wed, Jun 03, 2015 at 16:12 +0000, Bruno Oliveira wrote:
> > > Can you please enable issues (so meta!) for the repo?
> >
> > Done! :)
> >
> > > Then I suggest to continue discussions there, so people interested in
> > > it can watch the repo, and everyone else doesn't get annoyed by the
> > > mails on the ML :)
> >
> > Sure! Everyone interested is welcome to join.
>
> For the record, i don't mind discussions here but am also happy
> for discussions to go on in that repo.  Thanks all for caring!
>
> holger
>
> > On Wed, Jun 3, 2015 at 1:01 PM Florian Bruhin <me at the-compiler.org>
> wrote:
> >
> > > * Bruno Oliveira <nicoddemus at gmail.com> [2015-06-03 15:59:17 +0000]:
> > > > > For the script, or for a playground to do the migration?
> > > >
> > > > I meant for the script... I agree that everyone should use their own
> > > > playground for testing it. Added you and Anatoly as collaborators,
> feel
> > > > free to commit any fixes there as you find it.
> > > >
> > > > Once the script to migrate the issues is stable, we can discuss the
> next
> > > > steps in detail.
> > >
> > > Can you please enable issues (so meta!) for the repo?
> > >
> > > Then I suggest to continue discussions there, so people interested in
> > > it can watch the repo, and everyone else doesn't get annoyed by the
> > > mails on the ML :)
> > >
> > > Florian
> > >
> > > --
> > > http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
> > >    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
> > >          I love long mails! | http://email.is-not-s.ms/
> > >
>
> > _______________________________________________
> > pytest-dev mailing list
> > pytest-dev at python.org
> > https://mail.python.org/mailman/listinfo/pytest-dev
>
>
> --
> about me:    http://holgerkrekel.net/about-me/
> contracting: http://merlinux.eu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150605/e2464f73/attachment.html>

From me at the-compiler.org  Fri Jun  5 06:19:12 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Fri, 5 Jun 2015 06:19:12 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
Message-ID: <20150605041912.GB26357@tonks>

* Bruno Oliveira <nicoddemus at gmail.com> [2015-06-05 03:06:35 +0000]:
> After a few changes to the script, I have migrated all issues to a
> temporary repository at:
> https://github.com/nicoddemus/pytest-issues-migration.
> 
> Moved issues retain:
> - body
> - creation date
> - comments
> - kind and component (added as labels)

Awesome, thanks for the great work!

> Issue and comment authors are added to each issue or comment as a note in
> the body, but the issue author will always be the person that ran the
> script (this is a limitation of the GitHub import API). Is this a problem?
> When doing the final migration, which user should we use?

I suggest creating a new user for the migration so it's immediately
apparent that's not the real issue author.

For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
with the pytest logo as avatar.

> Any feedback on how we can improve the import process?

I think the "Bitbucket / originally reported by" footer should be at
the top (before the issue text) instead - again so it's immediately
apparent what's going on.

Also, stuff inside <> seems to be removed?
See https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
for example.

Some other ideas which might or might not be worth the effort:

- Rewrite full links to an issue (as opposed to #nnn identifiers) to
  point to the correct/new location (or to use #nnn instead)

- Rewrite links to PRs, if PRs will be migrated

- Rewrite those "-> <<cset ...>>" comments to point at git commits
  (when the repo is migrated) instead of hg changesets).
  (See link above for a [broken] example)

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150605/b76bab0e/attachment.sig>

From nicoddemus at gmail.com  Fri Jun  5 17:33:03 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Fri, 05 Jun 2015 15:33:03 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150605041912.GB26357@tonks>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
Message-ID: <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>

On Fri, Jun 5, 2015 at 1:19 AM Florian Bruhin <me at the-compiler.org> wrote:

> > When doing the final migration, which user should we use?
>
> I suggest creating a new user for the migration so it's immediately
> apparent that's not the real issue author.


> For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
> with the pytest logo as avatar.
>

Yes, that seems like a good solution. :)

About your other suggestions, I agree with most of them and created an
issue with your points here:
https://github.com/nicoddemus/bitbucket_issue_migration/issues/2

There are only two which I don't think are worth the effort:

- Porting PRs seems to be tricky, since we would have to port the patches
as well;
- Update the changesets that appear in issues/comments: since those
changesets will be different when we convert from Mercurial to Git, there's
no easy way to map them;

If others have any more suggestions, feel free to comment here or at
https://github.com/nicoddemus/bitbucket_issue_migration/issues/2.

Cheers,


> I think the "Bitbucket / originally reported by" footer should be at
> the top (before the issue text) instead - again so it's immediately
> apparent what's going on.
>
> Also, stuff inside <> seems to be removed?
> See
> https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
> for example.
>
> Some other ideas which might or might not be worth the effort:
>
> - Rewrite full links to an issue (as opposed to #nnn identifiers) to
>   point to the correct/new location (or to use #nnn instead)
>
> - Rewrite links to PRs, if PRs will be migrated
>
> - Rewrite those "-> <<cset ...>>" comments to point at git commits
>   (when the repo is migrated) instead of hg changesets).
>   (See link above for a [broken] example)
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150605/50e6411a/attachment.html>

From flub at devork.be  Sat Jun  6 01:12:27 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Sat, 6 Jun 2015 00:12:27 +0100
Subject: [pytest-dev] trainer needed for pytest training at EP2015
In-Reply-To: <E656ADAE-F5FC-4050-AA6B-BDD1C2428ECE@pelme.se>
References: <20150602093049.GP29732@merlinux.eu> <20150603093148.GP26357@tonks>
 <E656ADAE-F5FC-4050-AA6B-BDD1C2428ECE@pelme.se>
Message-ID: <CAAWk_DzDw2U1SYnsvfe2+m94M=pkwXrsfvKjZxMYMdjKa_D4Qw@mail.gmail.com>

Hi Florian,

Like Andreas I'm happy to help with running the session itself which I
did last year as well.  I already have my own talk to prepare so I'm
not very keen on doing too much preparation for the training though.
And if it allows you to attend the conference even better.

Regards,
Floris


On 3 June 2015 at 10:43, Andreas Pelme <andreas at pelme.se> wrote:
> Hi Florian,
>
>> On 3 jun 2015, at 11:31, Florian Bruhin <me at the-compiler.org> wrote:
>>
>> So if there isn't someone else who'd take it and you think I could do
>> it, let's talk!
>
> I helped out last year and it was a lot of fun. I have not finalized my plans for EP and am not yet sure I will be able to make it. If I?m there, I will for sure be able help out and assist as much as I can. I?m happy if you take it on and if it works out paying your ticket that would be great. I will probably not be able to help with any preparations before EP, but during EP I can help out as much as needed. I should finalize my plans for EP in the coming week and will let you know whether I will make it or not.
>
> Holger, Floris and Brianna: I?m really excited that you?re keynoting, giving a talk on the pytest hooks and the Adopt pytest month! The EP session list looks great :)
>
> Cheers,
> Andreas
>
>
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>

From nicoddemus at gmail.com  Sat Jun  6 05:46:10 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Sat, 06 Jun 2015 03:46:10 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
Message-ID: <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>

Implemented the last suggestions by Florian, I think issue migration looks
good now.

What would be the next steps? I suggest the following list (from the top of
my head):

1. Create github.com/pytest-dev/pytest and move issues (better do this
first so the migrated issues have the same id as the original ones);
2. Convert pytest Hg repository to Git and upload to GitHub;
3. Add to bitbucket's README a notice about the move to GitHub, and that
new issues/PRs should be posted there;
4. Update all links in the documentation and PyPI;
5. Update "how to contribute" docs;
6. Upload new docs to pytest.org;
7. Ask submitters to re-create PRs at the new repository;
8. Send an email to all relevant mailing lists about the migration;

After the migration process is complete, we can start to take advantage of
GitHub's ecosystem, for example start using Travis for CI, code coverage
with coveralls.io, etc.

IMO all this must be done in a short time, because if we start the
migration process and stall without completing it, links, issues, PRs etc
might get out of sync, so it it is better to gather a few contributors and
choose a "Migration Sprint" day to start and finish all the steps required
for a full migration.

Cheers,

On Fri, Jun 5, 2015 at 12:33 PM Bruno Oliveira <nicoddemus at gmail.com> wrote:

> On Fri, Jun 5, 2015 at 1:19 AM Florian Bruhin <me at the-compiler.org> wrote:
>
>> > When doing the final migration, which user should we use?
>>
>> I suggest creating a new user for the migration so it's immediately
>> apparent that's not the real issue author.
>
>
>> For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
>> with the pytest logo as avatar.
>>
>
> Yes, that seems like a good solution. :)
>
> About your other suggestions, I agree with most of them and created an
> issue with your points here:
> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2
>
> There are only two which I don't think are worth the effort:
>
> - Porting PRs seems to be tricky, since we would have to port the patches
> as well;
> - Update the changesets that appear in issues/comments: since those
> changesets will be different when we convert from Mercurial to Git, there's
> no easy way to map them;
>
> If others have any more suggestions, feel free to comment here or at
> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2.
>
> Cheers,
>
>
>> I think the "Bitbucket / originally reported by" footer should be at
>> the top (before the issue text) instead - again so it's immediately
>> apparent what's going on.
>>
>> Also, stuff inside <> seems to be removed?
>> See
>> https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
>> for example.
>>
>> Some other ideas which might or might not be worth the effort:
>>
>> - Rewrite full links to an issue (as opposed to #nnn identifiers) to
>>   point to the correct/new location (or to use #nnn instead)
>>
>> - Rewrite links to PRs, if PRs will be migrated
>>
>> - Rewrite those "-> <<cset ...>>" comments to point at git commits
>>   (when the repo is migrated) instead of hg changesets).
>>   (See link above for a [broken] example)
>>
>> Florian
>>
>> --
>> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>>          I love long mails! | http://email.is-not-s.ms/
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150606/3e5a8aa0/attachment-0001.html>

From bubenkoff at gmail.com  Sat Jun  6 12:15:25 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Sat, 06 Jun 2015 10:15:25 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
Message-ID: <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>

Sounds like a good plan!
It's easier to implement it by one person I think and that person is you! :)

On 05:46, Sat, Jun 6, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:

> Implemented the last suggestions by Florian, I think issue migration looks
> good now.
>
> What would be the next steps? I suggest the following list (from the top
> of my head):
>
> 1. Create github.com/pytest-dev/pytest and move issues (better do this
> first so the migrated issues have the same id as the original ones);
> 2. Convert pytest Hg repository to Git and upload to GitHub;
> 3. Add to bitbucket's README a notice about the move to GitHub, and that
> new issues/PRs should be posted there;
> 4. Update all links in the documentation and PyPI;
> 5. Update "how to contribute" docs;
> 6. Upload new docs to pytest.org;
> 7. Ask submitters to re-create PRs at the new repository;
> 8. Send an email to all relevant mailing lists about the migration;
>
> After the migration process is complete, we can start to take advantage of
> GitHub's ecosystem, for example start using Travis for CI, code coverage
> with coveralls.io, etc.
>
> IMO all this must be done in a short time, because if we start the
> migration process and stall without completing it, links, issues, PRs etc
> might get out of sync, so it it is better to gather a few contributors and
> choose a "Migration Sprint" day to start and finish all the steps required
> for a full migration.
>
> Cheers,
>
> On Fri, Jun 5, 2015 at 12:33 PM Bruno Oliveira <nicoddemus at gmail.com>
> wrote:
>
>> On Fri, Jun 5, 2015 at 1:19 AM Florian Bruhin <me at the-compiler.org>
>> wrote:
>>
>>> > When doing the final migration, which user should we use?
>>>
>>> I suggest creating a new user for the migration so it's immediately
>>> apparent that's not the real issue author.
>>
>>
>>> For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
>>> with the pytest logo as avatar.
>>>
>>
>> Yes, that seems like a good solution. :)
>>
>> About your other suggestions, I agree with most of them and created an
>> issue with your points here:
>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2
>>
>> There are only two which I don't think are worth the effort:
>>
>> - Porting PRs seems to be tricky, since we would have to port the patches
>> as well;
>> - Update the changesets that appear in issues/comments: since those
>> changesets will be different when we convert from Mercurial to Git, there's
>> no easy way to map them;
>>
>> If others have any more suggestions, feel free to comment here or at
>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2.
>>
>> Cheers,
>>
>>
>>> I think the "Bitbucket / originally reported by" footer should be at
>>> the top (before the issue text) instead - again so it's immediately
>>> apparent what's going on.
>>>
>>> Also, stuff inside <> seems to be removed?
>>> See
>>> https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
>>> for example.
>>>
>>> Some other ideas which might or might not be worth the effort:
>>>
>>> - Rewrite full links to an issue (as opposed to #nnn identifiers) to
>>>   point to the correct/new location (or to use #nnn instead)
>>>
>>> - Rewrite links to PRs, if PRs will be migrated
>>>
>>> - Rewrite those "-> <<cset ...>>" comments to point at git commits
>>>   (when the repo is migrated) instead of hg changesets).
>>>   (See link above for a [broken] example)
>>>
>>> Florian
>>>
>>> --
>>> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>>>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>>>          I love long mails! | http://email.is-not-s.ms/
>>>
>> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150606/165400cc/attachment.html>

From nicoddemus at gmail.com  Sun Jun  7 18:59:37 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Sun, 07 Jun 2015 16:59:37 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
Message-ID: <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>

No problem, if everybody agrees.

I'm playing with converting the mercurial repository to Git, having used
the HgGit plugin (http://hg-git.github.io/) and pushed the converted
repository to https://github.com/nicoddemus/pytest-issues-migration. Seems
OK to me, but does anyone with more experience want to suggest some other
approach perhaps?

I think a good time to do the complete migration would be during the
weekend when there is not much going on repository and issue wise. How does
next Saturday (June 13th) sound to everyone?

On Sat, Jun 6, 2015 at 7:15 AM Anatoly Bubenkov <bubenkoff at gmail.com> wrote:

> Sounds like a good plan!
> It's easier to implement it by one person I think and that person is you!
> :)
>
> On 05:46, Sat, Jun 6, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:
>
>> Implemented the last suggestions by Florian, I think issue migration
>> looks good now.
>>
>> What would be the next steps? I suggest the following list (from the top
>> of my head):
>>
>> 1. Create github.com/pytest-dev/pytest and move issues (better do this
>> first so the migrated issues have the same id as the original ones);
>> 2. Convert pytest Hg repository to Git and upload to GitHub;
>> 3. Add to bitbucket's README a notice about the move to GitHub, and that
>> new issues/PRs should be posted there;
>> 4. Update all links in the documentation and PyPI;
>> 5. Update "how to contribute" docs;
>> 6. Upload new docs to pytest.org;
>> 7. Ask submitters to re-create PRs at the new repository;
>> 8. Send an email to all relevant mailing lists about the migration;
>>
>> After the migration process is complete, we can start to take advantage
>> of GitHub's ecosystem, for example start using Travis for CI, code coverage
>> with coveralls.io, etc.
>>
>> IMO all this must be done in a short time, because if we start the
>> migration process and stall without completing it, links, issues, PRs etc
>> might get out of sync, so it it is better to gather a few contributors and
>> choose a "Migration Sprint" day to start and finish all the steps required
>> for a full migration.
>>
>> Cheers,
>>
>> On Fri, Jun 5, 2015 at 12:33 PM Bruno Oliveira <nicoddemus at gmail.com>
>> wrote:
>>
>>> On Fri, Jun 5, 2015 at 1:19 AM Florian Bruhin <me at the-compiler.org>
>>> wrote:
>>>
>>>> > When doing the final migration, which user should we use?
>>>>
>>>> I suggest creating a new user for the migration so it's immediately
>>>> apparent that's not the real issue author.
>>>
>>>
>>>> For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
>>>> with the pytest logo as avatar.
>>>>
>>>
>>> Yes, that seems like a good solution. :)
>>>
>>> About your other suggestions, I agree with most of them and created an
>>> issue with your points here:
>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2
>>>
>>> There are only two which I don't think are worth the effort:
>>>
>>> - Porting PRs seems to be tricky, since we would have to port the
>>> patches as well;
>>> - Update the changesets that appear in issues/comments: since those
>>> changesets will be different when we convert from Mercurial to Git, there's
>>> no easy way to map them;
>>>
>>> If others have any more suggestions, feel free to comment here or at
>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2.
>>>
>>> Cheers,
>>>
>>>
>>>> I think the "Bitbucket / originally reported by" footer should be at
>>>> the top (before the issue text) instead - again so it's immediately
>>>> apparent what's going on.
>>>>
>>>> Also, stuff inside <> seems to be removed?
>>>> See
>>>> https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
>>>> for example.
>>>>
>>>> Some other ideas which might or might not be worth the effort:
>>>>
>>>> - Rewrite full links to an issue (as opposed to #nnn identifiers) to
>>>>   point to the correct/new location (or to use #nnn instead)
>>>>
>>>> - Rewrite links to PRs, if PRs will be migrated
>>>>
>>>> - Rewrite those "-> <<cset ...>>" comments to point at git commits
>>>>   (when the repo is migrated) instead of hg changesets).
>>>>   (See link above for a [broken] example)
>>>>
>>>> Florian
>>>>
>>>> --
>>>> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>>>>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>>>>          I love long mails! | http://email.is-not-s.ms/
>>>>
>>> _______________________________________________
>
>
>> pytest-dev mailing list
>> pytest-dev at python.org
>> https://mail.python.org/mailman/listinfo/pytest-dev
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150607/892883ec/attachment.html>

From bubenkoff at gmail.com  Sun Jun  7 19:40:16 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Sun, 07 Jun 2015 17:40:16 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
Message-ID: <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>

Earlier I used gitifyhg to convert
But if that tool worked, why bother with alternatives

On 18:59, Sun, Jun 7, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:

> No problem, if everybody agrees.
>
> I'm playing with converting the mercurial repository to Git, having used
> the HgGit plugin (http://hg-git.github.io/) and pushed the converted
> repository to https://github.com/nicoddemus/pytest-issues-migration.
> Seems OK to me, but does anyone with more experience want to suggest some
> other approach perhaps?
>
> I think a good time to do the complete migration would be during the
> weekend when there is not much going on repository and issue wise. How does
> next Saturday (June 13th) sound to everyone?
>
> On Sat, Jun 6, 2015 at 7:15 AM Anatoly Bubenkov <bubenkoff at gmail.com>
> wrote:
>
>> Sounds like a good plan!
>> It's easier to implement it by one person I think and that person is you!
>> :)
>>
>> On 05:46, Sat, Jun 6, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:
>>
>>> Implemented the last suggestions by Florian, I think issue migration
>>> looks good now.
>>>
>>> What would be the next steps? I suggest the following list (from the top
>>> of my head):
>>>
>>> 1. Create github.com/pytest-dev/pytest and move issues (better do this
>>> first so the migrated issues have the same id as the original ones);
>>> 2. Convert pytest Hg repository to Git and upload to GitHub;
>>> 3. Add to bitbucket's README a notice about the move to GitHub, and that
>>> new issues/PRs should be posted there;
>>> 4. Update all links in the documentation and PyPI;
>>> 5. Update "how to contribute" docs;
>>> 6. Upload new docs to pytest.org;
>>> 7. Ask submitters to re-create PRs at the new repository;
>>> 8. Send an email to all relevant mailing lists about the migration;
>>>
>>> After the migration process is complete, we can start to take advantage
>>> of GitHub's ecosystem, for example start using Travis for CI, code coverage
>>> with coveralls.io, etc.
>>>
>>> IMO all this must be done in a short time, because if we start the
>>> migration process and stall without completing it, links, issues, PRs etc
>>> might get out of sync, so it it is better to gather a few contributors and
>>> choose a "Migration Sprint" day to start and finish all the steps required
>>> for a full migration.
>>>
>>> Cheers,
>>>
>>> On Fri, Jun 5, 2015 at 12:33 PM Bruno Oliveira <nicoddemus at gmail.com>
>>> wrote:
>>>
>>>> On Fri, Jun 5, 2015 at 1:19 AM Florian Bruhin <me at the-compiler.org>
>>>> wrote:
>>>>
>>>>> > When doing the final migration, which user should we use?
>>>>>
>>>>> I suggest creating a new user for the migration so it's immediately
>>>>> apparent that's not the real issue author.
>>>>
>>>>
>>>>> For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
>>>>> with the pytest logo as avatar.
>>>>>
>>>>
>>>> Yes, that seems like a good solution. :)
>>>>
>>>> About your other suggestions, I agree with most of them and created an
>>>> issue with your points here:
>>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2
>>>>
>>>> There are only two which I don't think are worth the effort:
>>>>
>>>> - Porting PRs seems to be tricky, since we would have to port the
>>>> patches as well;
>>>> - Update the changesets that appear in issues/comments: since those
>>>> changesets will be different when we convert from Mercurial to Git, there's
>>>> no easy way to map them;
>>>>
>>>> If others have any more suggestions, feel free to comment here or at
>>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2.
>>>>
>>>> Cheers,
>>>>
>>>>
>>>>> I think the "Bitbucket / originally reported by" footer should be at
>>>>> the top (before the issue text) instead - again so it's immediately
>>>>> apparent what's going on.
>>>>>
>>>>> Also, stuff inside <> seems to be removed?
>>>>> See
>>>>> https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
>>>>> for example.
>>>>>
>>>>> Some other ideas which might or might not be worth the effort:
>>>>>
>>>>> - Rewrite full links to an issue (as opposed to #nnn identifiers) to
>>>>>   point to the correct/new location (or to use #nnn instead)
>>>>>
>>>>> - Rewrite links to PRs, if PRs will be migrated
>>>>>
>>>>> - Rewrite those "-> <<cset ...>>" comments to point at git commits
>>>>>   (when the repo is migrated) instead of hg changesets).
>>>>>   (See link above for a [broken] example)
>>>>>
>>>>> Florian
>>>>>
>>>>> --
>>>>> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>>>>>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>>>>>          I love long mails! | http://email.is-not-s.ms/
>>>>>
>>>> _______________________________________________
>>
>>
>>> pytest-dev mailing list
>>> pytest-dev at python.org
>>> https://mail.python.org/mailman/listinfo/pytest-dev
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150607/730eab20/attachment-0001.html>

From brianna.laugher at gmail.com  Tue Jun  9 00:12:09 2015
From: brianna.laugher at gmail.com (Brianna Laugher)
Date: Tue, 9 Jun 2015 00:12:09 +0200
Subject: [pytest-dev] Adopt Pytest Month the report (draft)
Message-ID: <CAPjwCjvZr2vbbyCNvZoRAiph9_ELBF23MsJ3mTBZwvTjChcRKQ@mail.gmail.com>

Hello!

Sorry to keep you all in suspense since April... But finally! here is a
draft of the APM report.

https://bitbucket.org/pytest-dev/pytest/issue-attachment/676/pytest-dev/pytest/1433801126.73/676/APMreport.pdf

Although it's publicly accessible I just want to consider it a draft for a
couple of weeks so the pytest helpers and project maintainers can tell me
if they have any concerns with what I have written. After that I can update
it (and the pytest web page I guess).

Other feedback is welcome, if anything is unclear or you think an
explanation of anything is missing. You could give feedback here, or
privately, or on bitbucket (well for the time being!)
https://bitbucket.org/pytest-dev/pytest/issue/676/adopt-pytest-month-2015

I am also giving a talk about APM at Europython, I know I'll see some of
you there. :)
https://ep2015.europython.eu/conference/talks/the-realities-of-open-source-testing-lessons-learned-from-adopt-pytest-month

Thanks!
Brianna


-- 
They've just been waiting in a mountain for the right moment:
http://modernthings.org/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150609/325e46cd/attachment.html>

From flub at devork.be  Wed Jun 10 11:44:22 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Wed, 10 Jun 2015 10:44:22 +0100
Subject: [pytest-dev] merging PR freedom
In-Reply-To: <20150603215305.GM29732@merlinux.eu>
References: <20150603215305.GM29732@merlinux.eu>
Message-ID: <CAAWk_DzVYfXJG7=0JZ0+RFv6GqcP2kZoMNourL+OqqT00K3UKw@mail.gmail.com>

On 3 June 2015 at 22:53, holger krekel <holger at merlinux.eu> wrote:
> Hi Ronny and all other commiters :)
>
> i realize that from time to time you only "approve" a PR but don't
> merge it.  Please feel more free to just merge.

I guess it's partially habit, sometimes just giving other people the
chance to comment.  The habit part is because at work we've settled
into the workflow where reviewers only ever approve and only the PR
submitter merges their own PR.  But things probably speed up when
merging right away so I'm happy with that too.


Regards,
Floris

From bubenkoff at gmail.com  Wed Jun 10 15:13:46 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Wed, 10 Jun 2015 13:13:46 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
Message-ID: <CACTnz+EpT-Skh1rBi-z16MBMJy2q69N=oXMC_JW5iY+j63TJjA@mail.gmail.com>

+1

On Wed, Jun 10, 2015 at 3:07 PM Bruno Oliveira <nicoddemus at gmail.com> wrote:

> Hi guys,
>
> Just wanted to know if everyone is OK with migrating pytest to GitHub this
> Saturday (June 13th).
> If all agree, I will send an email Saturday when I start the migration
> process (issues + repository), meanwhile no one should commit to BitBucket
> until the process is complete.
>
> Cheers,
>
>
>
> On Sun, Jun 7, 2015 at 2:40 PM Anatoly Bubenkov <bubenkoff at gmail.com>
> wrote:
>
>> Earlier I used gitifyhg to convert
>> But if that tool worked, why bother with alternatives
>>
>> On 18:59, Sun, Jun 7, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:
>>
>>> No problem, if everybody agrees.
>>>
>>> I'm playing with converting the mercurial repository to Git, having used
>>> the HgGit plugin (http://hg-git.github.io/) and pushed the converted
>>> repository to https://github.com/nicoddemus/pytest-issues-migration.
>>> Seems OK to me, but does anyone with more experience want to suggest some
>>> other approach perhaps?
>>>
>>> I think a good time to do the complete migration would be during the
>>> weekend when there is not much going on repository and issue wise. How does
>>> next Saturday (June 13th) sound to everyone?
>>>
>>> On Sat, Jun 6, 2015 at 7:15 AM Anatoly Bubenkov <bubenkoff at gmail.com>
>>> wrote:
>>>
>>>> Sounds like a good plan!
>>>> It's easier to implement it by one person I think and that person is
>>>> you! :)
>>>>
>>>> On 05:46, Sat, Jun 6, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:
>>>>
>>>>> Implemented the last suggestions by Florian, I think issue migration
>>>>> looks good now.
>>>>>
>>>>> What would be the next steps? I suggest the following list (from the
>>>>> top of my head):
>>>>>
>>>>> 1. Create github.com/pytest-dev/pytest and move issues (better do
>>>>> this first so the migrated issues have the same id as the original ones);
>>>>> 2. Convert pytest Hg repository to Git and upload to GitHub;
>>>>> 3. Add to bitbucket's README a notice about the move to GitHub, and
>>>>> that new issues/PRs should be posted there;
>>>>> 4. Update all links in the documentation and PyPI;
>>>>> 5. Update "how to contribute" docs;
>>>>> 6. Upload new docs to pytest.org;
>>>>> 7. Ask submitters to re-create PRs at the new repository;
>>>>> 8. Send an email to all relevant mailing lists about the migration;
>>>>>
>>>>> After the migration process is complete, we can start to take
>>>>> advantage of GitHub's ecosystem, for example start using Travis for CI,
>>>>> code coverage with coveralls.io, etc.
>>>>>
>>>>> IMO all this must be done in a short time, because if we start the
>>>>> migration process and stall without completing it, links, issues, PRs etc
>>>>> might get out of sync, so it it is better to gather a few contributors and
>>>>> choose a "Migration Sprint" day to start and finish all the steps required
>>>>> for a full migration.
>>>>>
>>>>> Cheers,
>>>>>
>>>>> On Fri, Jun 5, 2015 at 12:33 PM Bruno Oliveira <nicoddemus at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> On Fri, Jun 5, 2015 at 1:19 AM Florian Bruhin <me at the-compiler.org>
>>>>>> wrote:
>>>>>>
>>>>>>> > When doing the final migration, which user should we use?
>>>>>>>
>>>>>>> I suggest creating a new user for the migration so it's immediately
>>>>>>> apparent that's not the real issue author.
>>>>>>
>>>>>>
>>>>>>> For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
>>>>>>> with the pytest logo as avatar.
>>>>>>>
>>>>>>
>>>>>> Yes, that seems like a good solution. :)
>>>>>>
>>>>>> About your other suggestions, I agree with most of them and created
>>>>>> an issue with your points here:
>>>>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2
>>>>>>
>>>>>> There are only two which I don't think are worth the effort:
>>>>>>
>>>>>> - Porting PRs seems to be tricky, since we would have to port the
>>>>>> patches as well;
>>>>>> - Update the changesets that appear in issues/comments: since those
>>>>>> changesets will be different when we convert from Mercurial to Git, there's
>>>>>> no easy way to map them;
>>>>>>
>>>>>> If others have any more suggestions, feel free to comment here or at
>>>>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2.
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>>
>>>>>>> I think the "Bitbucket / originally reported by" footer should be at
>>>>>>> the top (before the issue text) instead - again so it's immediately
>>>>>>> apparent what's going on.
>>>>>>>
>>>>>>> Also, stuff inside <> seems to be removed?
>>>>>>> See
>>>>>>> https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
>>>>>>> for example.
>>>>>>>
>>>>>>> Some other ideas which might or might not be worth the effort:
>>>>>>>
>>>>>>> - Rewrite full links to an issue (as opposed to #nnn identifiers) to
>>>>>>>   point to the correct/new location (or to use #nnn instead)
>>>>>>>
>>>>>>> - Rewrite links to PRs, if PRs will be migrated
>>>>>>>
>>>>>>> - Rewrite those "-> <<cset ...>>" comments to point at git commits
>>>>>>>   (when the repo is migrated) instead of hg changesets).
>>>>>>>   (See link above for a [broken] example)
>>>>>>>
>>>>>>> Florian
>>>>>>>
>>>>>>> --
>>>>>>> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>>>>>>>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>>>>>>>          I love long mails! | http://email.is-not-s.ms/
>>>>>>>
>>>>>> _______________________________________________
>>>>
>>>>
>>>>> pytest-dev mailing list
>>>>> pytest-dev at python.org
>>>>> https://mail.python.org/mailman/listinfo/pytest-dev
>>>>>
>>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150610/1cc17e42/attachment.html>

From nicoddemus at gmail.com  Wed Jun 10 15:07:05 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Wed, 10 Jun 2015 13:07:05 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
Message-ID: <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>

Hi guys,

Just wanted to know if everyone is OK with migrating pytest to GitHub this
Saturday (June 13th).
If all agree, I will send an email Saturday when I start the migration
process (issues + repository), meanwhile no one should commit to BitBucket
until the process is complete.

Cheers,



On Sun, Jun 7, 2015 at 2:40 PM Anatoly Bubenkov <bubenkoff at gmail.com> wrote:

> Earlier I used gitifyhg to convert
> But if that tool worked, why bother with alternatives
>
> On 18:59, Sun, Jun 7, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:
>
>> No problem, if everybody agrees.
>>
>> I'm playing with converting the mercurial repository to Git, having used
>> the HgGit plugin (http://hg-git.github.io/) and pushed the converted
>> repository to https://github.com/nicoddemus/pytest-issues-migration.
>> Seems OK to me, but does anyone with more experience want to suggest some
>> other approach perhaps?
>>
>> I think a good time to do the complete migration would be during the
>> weekend when there is not much going on repository and issue wise. How does
>> next Saturday (June 13th) sound to everyone?
>>
>> On Sat, Jun 6, 2015 at 7:15 AM Anatoly Bubenkov <bubenkoff at gmail.com>
>> wrote:
>>
>>> Sounds like a good plan!
>>> It's easier to implement it by one person I think and that person is
>>> you! :)
>>>
>>> On 05:46, Sat, Jun 6, 2015 Bruno Oliveira <nicoddemus at gmail.com> wrote:
>>>
>>>> Implemented the last suggestions by Florian, I think issue migration
>>>> looks good now.
>>>>
>>>> What would be the next steps? I suggest the following list (from the
>>>> top of my head):
>>>>
>>>> 1. Create github.com/pytest-dev/pytest and move issues (better do this
>>>> first so the migrated issues have the same id as the original ones);
>>>> 2. Convert pytest Hg repository to Git and upload to GitHub;
>>>> 3. Add to bitbucket's README a notice about the move to GitHub, and
>>>> that new issues/PRs should be posted there;
>>>> 4. Update all links in the documentation and PyPI;
>>>> 5. Update "how to contribute" docs;
>>>> 6. Upload new docs to pytest.org;
>>>> 7. Ask submitters to re-create PRs at the new repository;
>>>> 8. Send an email to all relevant mailing lists about the migration;
>>>>
>>>> After the migration process is complete, we can start to take advantage
>>>> of GitHub's ecosystem, for example start using Travis for CI, code coverage
>>>> with coveralls.io, etc.
>>>>
>>>> IMO all this must be done in a short time, because if we start the
>>>> migration process and stall without completing it, links, issues, PRs etc
>>>> might get out of sync, so it it is better to gather a few contributors and
>>>> choose a "Migration Sprint" day to start and finish all the steps required
>>>> for a full migration.
>>>>
>>>> Cheers,
>>>>
>>>> On Fri, Jun 5, 2015 at 12:33 PM Bruno Oliveira <nicoddemus at gmail.com>
>>>> wrote:
>>>>
>>>>> On Fri, Jun 5, 2015 at 1:19 AM Florian Bruhin <me at the-compiler.org>
>>>>> wrote:
>>>>>
>>>>>> > When doing the final migration, which user should we use?
>>>>>>
>>>>>> I suggest creating a new user for the migration so it's immediately
>>>>>> apparent that's not the real issue author.
>>>>>
>>>>>
>>>>>> For example, a pytest-dev/pytest-bot/pytest-issue-migration/... user
>>>>>> with the pytest logo as avatar.
>>>>>>
>>>>>
>>>>> Yes, that seems like a good solution. :)
>>>>>
>>>>> About your other suggestions, I agree with most of them and created an
>>>>> issue with your points here:
>>>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2
>>>>>
>>>>> There are only two which I don't think are worth the effort:
>>>>>
>>>>> - Porting PRs seems to be tricky, since we would have to port the
>>>>> patches as well;
>>>>> - Update the changesets that appear in issues/comments: since those
>>>>> changesets will be different when we convert from Mercurial to Git, there's
>>>>> no easy way to map them;
>>>>>
>>>>> If others have any more suggestions, feel free to comment here or at
>>>>> https://github.com/nicoddemus/bitbucket_issue_migration/issues/2.
>>>>>
>>>>> Cheers,
>>>>>
>>>>>
>>>>>> I think the "Bitbucket / originally reported by" footer should be at
>>>>>> the top (before the issue text) instead - again so it's immediately
>>>>>> apparent what's going on.
>>>>>>
>>>>>> Also, stuff inside <> seems to be removed?
>>>>>> See
>>>>>> https://github.com/nicoddemus/pytest-issues-migration/issues/6#issuecomment-109143858
>>>>>> for example.
>>>>>>
>>>>>> Some other ideas which might or might not be worth the effort:
>>>>>>
>>>>>> - Rewrite full links to an issue (as opposed to #nnn identifiers) to
>>>>>>   point to the correct/new location (or to use #nnn instead)
>>>>>>
>>>>>> - Rewrite links to PRs, if PRs will be migrated
>>>>>>
>>>>>> - Rewrite those "-> <<cset ...>>" comments to point at git commits
>>>>>>   (when the repo is migrated) instead of hg changesets).
>>>>>>   (See link above for a [broken] example)
>>>>>>
>>>>>> Florian
>>>>>>
>>>>>> --
>>>>>> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>>>>>>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>>>>>>          I love long mails! | http://email.is-not-s.ms/
>>>>>>
>>>>> _______________________________________________
>>>
>>>
>>>> pytest-dev mailing list
>>>> pytest-dev at python.org
>>>> https://mail.python.org/mailman/listinfo/pytest-dev
>>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150610/d5ea9ed2/attachment-0001.html>

From flub at devork.be  Thu Jun 11 01:38:26 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Thu, 11 Jun 2015 00:38:26 +0100
Subject: [pytest-dev] Adopt Pytest Month the report (draft)
In-Reply-To: <CAPjwCjvZr2vbbyCNvZoRAiph9_ELBF23MsJ3mTBZwvTjChcRKQ@mail.gmail.com>
References: <CAPjwCjvZr2vbbyCNvZoRAiph9_ELBF23MsJ3mTBZwvTjChcRKQ@mail.gmail.com>
Message-ID: <CAAWk_Dwci520Om=U95W=1TpOLudSERPY=Xk8onGQXOe4qsKX0w@mail.gmail.com>

On 8 June 2015 at 23:12, Brianna Laugher <brianna.laugher at gmail.com> wrote:
> Hello!
>
> Sorry to keep you all in suspense since April... But finally! here is a
> draft of the APM report.
>
> https://bitbucket.org/pytest-dev/pytest/issue-attachment/676/pytest-dev/pytest/1433801126.73/676/APMreport.pdf

That's a really nice and very comprehensive writeup!  Thanks for
putting all the effort in!

From flub at devork.be  Thu Jun 11 01:47:29 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Thu, 11 Jun 2015 00:47:29 +0100
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
Message-ID: <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>

On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> Hi guys,
>
> Just wanted to know if everyone is OK with migrating pytest to GitHub this
> Saturday (June 13th).
> If all agree, I will send an email Saturday when I start the migration
> process (issues + repository), meanwhile no one should commit to BitBucket
> until the process is complete.

What happens to existing pull requests?  If we don't merge them before
Sat we have to convert them to git ourself and re-make the PR?

I think I'm happy to decline #271 and #292.
But for #296 I'm preparing a PR at
https://bitbucket.org/flub/pytest-py35 (because it needed to be
rebased onto pytest-2.7).  I think I'd prefer to merge that before the
conversion even if not all tests are fixed by Sat (but hopefully they
will be!).  Does that sound reasonable?


Floris

From nicoddemus at gmail.com  Thu Jun 11 02:10:11 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 11 Jun 2015 00:10:11 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
References: <20150603142052.GV29732@merlinux.eu> <20150603145414.GU26357@tonks>
 <CA+RQFff28rBCGJR6yTuUePpYZQ9F=m=fzvDM_fefPqUsDbvj+g@mail.gmail.com>
 <20150603155118.GV26357@tonks>
 <CA+RQFfcOP1c5EmsEsgY94NvjbXtk1=o8uV4=w5BbBKLtfxjv=w@mail.gmail.com>
 <20150603160127.GW26357@tonks>
 <CA+RQFfe+vDJV3Gk2BunYoJV2kH2rG=R0+mkCEDVp3_iTvLbARg@mail.gmail.com>
 <20150603184839.GI29732@merlinux.eu>
 <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
Message-ID: <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>

Hi Floris,

> What happens to existing pull requests?  If we don't merge them before
> Sat we have to convert them to git ourself and re-make the PR?

Yes, as there are few PRs and there's no automatic way to do migrate them,
I think that is a reasonable approach. For PRs that; we should ask
submitters to open new PRs in GitHub after the conversion.

> But for #296 I'm preparing a PR at
> https://bitbucket.org/flub/pytest-py35 (because it needed to be
> rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> conversion even if not all tests are fixed by Sat (but hopefully they
> will be!).  Does that sound reasonable?

IMO yes, unless others disagree.

Cheers,


On Wed, Jun 10, 2015 at 8:47 PM Floris Bruynooghe <flub at devork.be> wrote:

> On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > Hi guys,
> >
> > Just wanted to know if everyone is OK with migrating pytest to GitHub
> this
> > Saturday (June 13th).
> > If all agree, I will send an email Saturday when I start the migration
> > process (issues + repository), meanwhile no one should commit to
> BitBucket
> > until the process is complete.
>
> What happens to existing pull requests?  If we don't merge them before
> Sat we have to convert them to git ourself and re-make the PR?
>
> I think I'm happy to decline #271 and #292.
> But for #296 I'm preparing a PR at
> https://bitbucket.org/flub/pytest-py35 (because it needed to be
> rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> conversion even if not all tests are fixed by Sat (but hopefully they
> will be!).  Does that sound reasonable?
>
>
> Floris
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150611/c26794a4/attachment.html>

From holger at merlinux.eu  Fri Jun 12 13:26:14 2015
From: holger at merlinux.eu (holger krekel)
Date: Fri, 12 Jun 2015 11:26:14 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
References: <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
Message-ID: <20150612112614.GF29732@merlinux.eu>

Hi Bruno,

was on travel, then ill, initial scan of your work looks great!

I am not going to be much available during the weekend and would prefer a
switch date on monday.  It's be best to have PRs merged or rejected until
then.

As to the migrated issues: i just tried with two random user handles --
the bitbucket userids map 1:1 to a github one -- so we could add a link
to the github user (stating "maybe it's this github user: ..." because
we can't be 100% sure in all cases where the user id exists).

Also is there a way to, after migration, post to each bitbucket issue 
the link to the corresponding github issue?

thanks again for the work.

holger




On Thu, Jun 11, 2015 at 00:10 +0000, Bruno Oliveira wrote:
> Hi Floris,
> 
> > What happens to existing pull requests?  If we don't merge them before
> > Sat we have to convert them to git ourself and re-make the PR?
> 
> Yes, as there are few PRs and there's no automatic way to do migrate them,
> I think that is a reasonable approach. For PRs that; we should ask
> submitters to open new PRs in GitHub after the conversion.
> 
> > But for #296 I'm preparing a PR at
> > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> > conversion even if not all tests are fixed by Sat (but hopefully they
> > will be!).  Does that sound reasonable?
> 
> IMO yes, unless others disagree.
> 
> Cheers,
> 
> 
> On Wed, Jun 10, 2015 at 8:47 PM Floris Bruynooghe <flub at devork.be> wrote:
> 
> > On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > > Hi guys,
> > >
> > > Just wanted to know if everyone is OK with migrating pytest to GitHub
> > this
> > > Saturday (June 13th).
> > > If all agree, I will send an email Saturday when I start the migration
> > > process (issues + repository), meanwhile no one should commit to
> > BitBucket
> > > until the process is complete.
> >
> > What happens to existing pull requests?  If we don't merge them before
> > Sat we have to convert them to git ourself and re-make the PR?
> >
> > I think I'm happy to decline #271 and #292.
> > But for #296 I'm preparing a PR at
> > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> > conversion even if not all tests are fixed by Sat (but hopefully they
> > will be!).  Does that sound reasonable?
> >
> >
> > Floris
> > _______________________________________________
> > pytest-dev mailing list
> > pytest-dev at python.org
> > https://mail.python.org/mailman/listinfo/pytest-dev
> >

> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev


-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From nicoddemus at gmail.com  Fri Jun 12 14:10:48 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Fri, 12 Jun 2015 12:10:48 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150612112614.GF29732@merlinux.eu>
References: <CA+RQFfcwEmgkbFu5mZeJaWzbWA+7ywC3YUnqyKHgw8SVJ3ocQg@mail.gmail.com>
 <20150605041912.GB26357@tonks>
 <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
Message-ID: <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>

Hi Holger,

> I am not going to be much available during the weekend and would prefer a
> switch date on monday.  It's be best to have PRs merged or rejected until
> then.

I suggested Saturday because usually there's not much commits happening,
but Monday works for me too.

> As to the migrated issues: i just tried with two random user handles --
> the bitbucket userids map 1:1 to a github one -- so we could add a link
> to the github user (stating "maybe it's this github user: ..." because
> we can't be 100% sure in all cases where the user id exists).

I thought about this, but was concerned that while there might be two
identical bitbucket and GitHub user ids, they might not belong to the same
person.
In practice I don't think it will be a problem though, so I will add a
another link to the GitHub user id then.

> Also is there a way to, after migration, post to each bitbucket issue
> the link to the corresponding github issue?

Yes, good idea, I will implement this as well.

> thanks again for the work.

No problem, thanks for the feedback.

Can we set the official migration date to Monday then?

On Fri, Jun 12, 2015 at 8:26 AM holger krekel <holger at merlinux.eu> wrote:

> Hi Bruno,
>
> was on travel, then ill, initial scan of your work looks great!
>
> I am not going to be much available during the weekend and would prefer a
> switch date on monday.  It's be best to have PRs merged or rejected until
> then.
>
> As to the migrated issues: i just tried with two random user handles --
> the bitbucket userids map 1:1 to a github one -- so we could add a link
> to the github user (stating "maybe it's this github user: ..." because
> we can't be 100% sure in all cases where the user id exists).
>
> Also is there a way to, after migration, post to each bitbucket issue
> the link to the corresponding github issue?
>
> thanks again for the work.
>
> holger
>
>
>
>
> On Thu, Jun 11, 2015 at 00:10 +0000, Bruno Oliveira wrote:
> > Hi Floris,
> >
> > > What happens to existing pull requests?  If we don't merge them before
> > > Sat we have to convert them to git ourself and re-make the PR?
> >
> > Yes, as there are few PRs and there's no automatic way to do migrate
> them,
> > I think that is a reasonable approach. For PRs that; we should ask
> > submitters to open new PRs in GitHub after the conversion.
> >
> > > But for #296 I'm preparing a PR at
> > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> > > conversion even if not all tests are fixed by Sat (but hopefully they
> > > will be!).  Does that sound reasonable?
> >
> > IMO yes, unless others disagree.
> >
> > Cheers,
> >
> >
> > On Wed, Jun 10, 2015 at 8:47 PM Floris Bruynooghe <flub at devork.be>
> wrote:
> >
> > > On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > > > Hi guys,
> > > >
> > > > Just wanted to know if everyone is OK with migrating pytest to GitHub
> > > this
> > > > Saturday (June 13th).
> > > > If all agree, I will send an email Saturday when I start the
> migration
> > > > process (issues + repository), meanwhile no one should commit to
> > > BitBucket
> > > > until the process is complete.
> > >
> > > What happens to existing pull requests?  If we don't merge them before
> > > Sat we have to convert them to git ourself and re-make the PR?
> > >
> > > I think I'm happy to decline #271 and #292.
> > > But for #296 I'm preparing a PR at
> > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> > > conversion even if not all tests are fixed by Sat (but hopefully they
> > > will be!).  Does that sound reasonable?
> > >
> > >
> > > Floris
> > > _______________________________________________
> > > pytest-dev mailing list
> > > pytest-dev at python.org
> > > https://mail.python.org/mailman/listinfo/pytest-dev
> > >
>
> > _______________________________________________
> > pytest-dev mailing list
> > pytest-dev at python.org
> > https://mail.python.org/mailman/listinfo/pytest-dev
>
>
> --
> about me:    http://holgerkrekel.net/about-me/
> contracting: http://merlinux.eu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150612/abaf1dda/attachment.html>

From holger at merlinux.eu  Sat Jun 13 09:05:05 2015
From: holger at merlinux.eu (holger krekel)
Date: Sat, 13 Jun 2015 07:05:05 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
References: <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
Message-ID: <20150613070505.GK29732@merlinux.eu>

On Fri, Jun 12, 2015 at 12:10 +0000, Bruno Oliveira wrote:
> Hi Holger,
> 
> > I am not going to be much available during the weekend and would prefer a
> > switch date on monday.  It's be best to have PRs merged or rejected until
> > then.
> 
> I suggested Saturday because usually there's not much commits happening,
> but Monday works for me too.
> 
> > As to the migrated issues: i just tried with two random user handles --
> > the bitbucket userids map 1:1 to a github one -- so we could add a link
> > to the github user (stating "maybe it's this github user: ..." because
> > we can't be 100% sure in all cases where the user id exists).
> 
> I thought about this, but was concerned that while there might be two
> identical bitbucket and GitHub user ids, they might not belong to the same
> person.
> In practice I don't think it will be a problem though, so I will add a
> another link to the GitHub user id then.

yes, i agree that we don't have 100% safety -- therefore my "maybe it's
this github user" suggestion for a text in the issue.

> > Also is there a way to, after migration, post to each bitbucket issue
> > the link to the corresponding github issue?
> 
> Yes, good idea, I will implement this as well.

i guess that this should be a "closing" post so that people
recognize they need to use github.

> > thanks again for the work.
> 
> No problem, thanks for the feedback.
> 
> Can we set the official migration date to Monday then?

Yes.  Can you announce in a top-level post here on the list along
with the proposed way of what's going to happen?


holger


> 
> On Fri, Jun 12, 2015 at 8:26 AM holger krekel <holger at merlinux.eu> wrote:
> 
> > Hi Bruno,
> >
> > was on travel, then ill, initial scan of your work looks great!
> >
> > I am not going to be much available during the weekend and would prefer a
> > switch date on monday.  It's be best to have PRs merged or rejected until
> > then.
> >
> > As to the migrated issues: i just tried with two random user handles --
> > the bitbucket userids map 1:1 to a github one -- so we could add a link
> > to the github user (stating "maybe it's this github user: ..." because
> > we can't be 100% sure in all cases where the user id exists).
> >
> > Also is there a way to, after migration, post to each bitbucket issue
> > the link to the corresponding github issue?
> >
> > thanks again for the work.
> >
> > holger
> >
> >
> >
> >
> > On Thu, Jun 11, 2015 at 00:10 +0000, Bruno Oliveira wrote:
> > > Hi Floris,
> > >
> > > > What happens to existing pull requests?  If we don't merge them before
> > > > Sat we have to convert them to git ourself and re-make the PR?
> > >
> > > Yes, as there are few PRs and there's no automatic way to do migrate
> > them,
> > > I think that is a reasonable approach. For PRs that; we should ask
> > > submitters to open new PRs in GitHub after the conversion.
> > >
> > > > But for #296 I'm preparing a PR at
> > > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > > rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> > > > conversion even if not all tests are fixed by Sat (but hopefully they
> > > > will be!).  Does that sound reasonable?
> > >
> > > IMO yes, unless others disagree.
> > >
> > > Cheers,
> > >
> > >
> > > On Wed, Jun 10, 2015 at 8:47 PM Floris Bruynooghe <flub at devork.be>
> > wrote:
> > >
> > > > On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > > > > Hi guys,
> > > > >
> > > > > Just wanted to know if everyone is OK with migrating pytest to GitHub
> > > > this
> > > > > Saturday (June 13th).
> > > > > If all agree, I will send an email Saturday when I start the
> > migration
> > > > > process (issues + repository), meanwhile no one should commit to
> > > > BitBucket
> > > > > until the process is complete.
> > > >
> > > > What happens to existing pull requests?  If we don't merge them before
> > > > Sat we have to convert them to git ourself and re-make the PR?
> > > >
> > > > I think I'm happy to decline #271 and #292.
> > > > But for #296 I'm preparing a PR at
> > > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > > rebased onto pytest-2.7).  I think I'd prefer to merge that before the
> > > > conversion even if not all tests are fixed by Sat (but hopefully they
> > > > will be!).  Does that sound reasonable?
> > > >
> > > >
> > > > Floris
> > > > _______________________________________________
> > > > pytest-dev mailing list
> > > > pytest-dev at python.org
> > > > https://mail.python.org/mailman/listinfo/pytest-dev
> > > >
> >
> > > _______________________________________________
> > > pytest-dev mailing list
> > > pytest-dev at python.org
> > > https://mail.python.org/mailman/listinfo/pytest-dev
> >
> >
> > --
> > about me:    http://holgerkrekel.net/about-me/
> > contracting: http://merlinux.eu
> >

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From nicoddemus at gmail.com  Sat Jun 13 18:32:11 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Sat, 13 Jun 2015 16:32:11 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150613070505.GK29732@merlinux.eu>
References: <CA+RQFfenyotBhBfvUJU_b9ONgbk8voHcdr7xN4Knpkf5Hrnp_g@mail.gmail.com>
 <CA+RQFfc7v9b8nZmf1JRWdJ_Cwyvitj1KGEHmFH-DRd=uhz133g@mail.gmail.com>
 <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
Message-ID: <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>

> Yes.  Can you announce in a top-level post here on the list along
> with the proposed way of what's going to happen?

Sure, although I don't think all PRs (currently 6) will be merged/rejected
until Monday... should we migrate anyway on Monday, or give it a few more
days to sort the remaining PRs out and do the actual migration, say,
Wednesday?

Btw, made another round of migration (issues + repository) at
https://github.com/pytestbot/pytest-issues-migration if anybody wants to
take another look.

Cheers,


On Sat, Jun 13, 2015 at 4:05 AM holger krekel <holger at merlinux.eu> wrote:

> On Fri, Jun 12, 2015 at 12:10 +0000, Bruno Oliveira wrote:
> > Hi Holger,
> >
> > > I am not going to be much available during the weekend and would
> prefer a
> > > switch date on monday.  It's be best to have PRs merged or rejected
> until
> > > then.
> >
> > I suggested Saturday because usually there's not much commits happening,
> > but Monday works for me too.
> >
> > > As to the migrated issues: i just tried with two random user handles --
> > > the bitbucket userids map 1:1 to a github one -- so we could add a link
> > > to the github user (stating "maybe it's this github user: ..." because
> > > we can't be 100% sure in all cases where the user id exists).
> >
> > I thought about this, but was concerned that while there might be two
> > identical bitbucket and GitHub user ids, they might not belong to the
> same
> > person.
> > In practice I don't think it will be a problem though, so I will add a
> > another link to the GitHub user id then.
>
> yes, i agree that we don't have 100% safety -- therefore my "maybe it's
> this github user" suggestion for a text in the issue.
>
> > > Also is there a way to, after migration, post to each bitbucket issue
> > > the link to the corresponding github issue?
> >
> > Yes, good idea, I will implement this as well.
>
> i guess that this should be a "closing" post so that people
> recognize they need to use github.
>
> > > thanks again for the work.
> >
> > No problem, thanks for the feedback.
> >
> > Can we set the official migration date to Monday then?
>
> Yes.  Can you announce in a top-level post here on the list along
> with the proposed way of what's going to happen?
>
>
> holger
>
>
> >
> > On Fri, Jun 12, 2015 at 8:26 AM holger krekel <holger at merlinux.eu>
> wrote:
> >
> > > Hi Bruno,
> > >
> > > was on travel, then ill, initial scan of your work looks great!
> > >
> > > I am not going to be much available during the weekend and would
> prefer a
> > > switch date on monday.  It's be best to have PRs merged or rejected
> until
> > > then.
> > >
> > > As to the migrated issues: i just tried with two random user handles --
> > > the bitbucket userids map 1:1 to a github one -- so we could add a link
> > > to the github user (stating "maybe it's this github user: ..." because
> > > we can't be 100% sure in all cases where the user id exists).
> > >
> > > Also is there a way to, after migration, post to each bitbucket issue
> > > the link to the corresponding github issue?
> > >
> > > thanks again for the work.
> > >
> > > holger
> > >
> > >
> > >
> > >
> > > On Thu, Jun 11, 2015 at 00:10 +0000, Bruno Oliveira wrote:
> > > > Hi Floris,
> > > >
> > > > > What happens to existing pull requests?  If we don't merge them
> before
> > > > > Sat we have to convert them to git ourself and re-make the PR?
> > > >
> > > > Yes, as there are few PRs and there's no automatic way to do migrate
> > > them,
> > > > I think that is a reasonable approach. For PRs that; we should ask
> > > > submitters to open new PRs in GitHub after the conversion.
> > > >
> > > > > But for #296 I'm preparing a PR at
> > > > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > > > rebased onto pytest-2.7).  I think I'd prefer to merge that before
> the
> > > > > conversion even if not all tests are fixed by Sat (but hopefully
> they
> > > > > will be!).  Does that sound reasonable?
> > > >
> > > > IMO yes, unless others disagree.
> > > >
> > > > Cheers,
> > > >
> > > >
> > > > On Wed, Jun 10, 2015 at 8:47 PM Floris Bruynooghe <flub at devork.be>
> > > wrote:
> > > >
> > > > > On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com>
> wrote:
> > > > > > Hi guys,
> > > > > >
> > > > > > Just wanted to know if everyone is OK with migrating pytest to
> GitHub
> > > > > this
> > > > > > Saturday (June 13th).
> > > > > > If all agree, I will send an email Saturday when I start the
> > > migration
> > > > > > process (issues + repository), meanwhile no one should commit to
> > > > > BitBucket
> > > > > > until the process is complete.
> > > > >
> > > > > What happens to existing pull requests?  If we don't merge them
> before
> > > > > Sat we have to convert them to git ourself and re-make the PR?
> > > > >
> > > > > I think I'm happy to decline #271 and #292.
> > > > > But for #296 I'm preparing a PR at
> > > > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > > > rebased onto pytest-2.7).  I think I'd prefer to merge that before
> the
> > > > > conversion even if not all tests are fixed by Sat (but hopefully
> they
> > > > > will be!).  Does that sound reasonable?
> > > > >
> > > > >
> > > > > Floris
> > > > > _______________________________________________
> > > > > pytest-dev mailing list
> > > > > pytest-dev at python.org
> > > > > https://mail.python.org/mailman/listinfo/pytest-dev
> > > > >
> > >
> > > > _______________________________________________
> > > > pytest-dev mailing list
> > > > pytest-dev at python.org
> > > > https://mail.python.org/mailman/listinfo/pytest-dev
> > >
> > >
> > > --
> > > about me:    http://holgerkrekel.net/about-me/
> > > contracting: http://merlinux.eu
> > >
>
> --
> about me:    http://holgerkrekel.net/about-me/
> contracting: http://merlinux.eu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150613/8aaac626/attachment.html>

From holger at merlinux.eu  Mon Jun 15 11:35:13 2015
From: holger at merlinux.eu (holger krekel)
Date: Mon, 15 Jun 2015 09:35:13 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
References: <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
Message-ID: <20150615093513.GR29732@merlinux.eu>

On Sat, Jun 13, 2015 at 16:32 +0000, Bruno Oliveira wrote:
> > Yes.  Can you announce in a top-level post here on the list along
> > with the proposed way of what's going to happen?
> 
> Sure, although I don't think all PRs (currently 6) will be merged/rejected
> until Monday... should we migrate anyway on Monday, or give it a few more
> days to sort the remaining PRs out and do the actual migration, say,
> Wednesday?
> 
> Btw, made another round of migration (issues + repository) at
> https://github.com/pytestbot/pytest-issues-migration if anybody wants to
> take another look.

we can go for migration anyway, i am now on IRC.

holger

> Cheers,
> 
> 
> On Sat, Jun 13, 2015 at 4:05 AM holger krekel <holger at merlinux.eu> wrote:
> 
> > On Fri, Jun 12, 2015 at 12:10 +0000, Bruno Oliveira wrote:
> > > Hi Holger,
> > >
> > > > I am not going to be much available during the weekend and would
> > prefer a
> > > > switch date on monday.  It's be best to have PRs merged or rejected
> > until
> > > > then.
> > >
> > > I suggested Saturday because usually there's not much commits happening,
> > > but Monday works for me too.
> > >
> > > > As to the migrated issues: i just tried with two random user handles --
> > > > the bitbucket userids map 1:1 to a github one -- so we could add a link
> > > > to the github user (stating "maybe it's this github user: ..." because
> > > > we can't be 100% sure in all cases where the user id exists).
> > >
> > > I thought about this, but was concerned that while there might be two
> > > identical bitbucket and GitHub user ids, they might not belong to the
> > same
> > > person.
> > > In practice I don't think it will be a problem though, so I will add a
> > > another link to the GitHub user id then.
> >
> > yes, i agree that we don't have 100% safety -- therefore my "maybe it's
> > this github user" suggestion for a text in the issue.
> >
> > > > Also is there a way to, after migration, post to each bitbucket issue
> > > > the link to the corresponding github issue?
> > >
> > > Yes, good idea, I will implement this as well.
> >
> > i guess that this should be a "closing" post so that people
> > recognize they need to use github.
> >
> > > > thanks again for the work.
> > >
> > > No problem, thanks for the feedback.
> > >
> > > Can we set the official migration date to Monday then?
> >
> > Yes.  Can you announce in a top-level post here on the list along
> > with the proposed way of what's going to happen?
> >
> >
> > holger
> >
> >
> > >
> > > On Fri, Jun 12, 2015 at 8:26 AM holger krekel <holger at merlinux.eu>
> > wrote:
> > >
> > > > Hi Bruno,
> > > >
> > > > was on travel, then ill, initial scan of your work looks great!
> > > >
> > > > I am not going to be much available during the weekend and would
> > prefer a
> > > > switch date on monday.  It's be best to have PRs merged or rejected
> > until
> > > > then.
> > > >
> > > > As to the migrated issues: i just tried with two random user handles --
> > > > the bitbucket userids map 1:1 to a github one -- so we could add a link
> > > > to the github user (stating "maybe it's this github user: ..." because
> > > > we can't be 100% sure in all cases where the user id exists).
> > > >
> > > > Also is there a way to, after migration, post to each bitbucket issue
> > > > the link to the corresponding github issue?
> > > >
> > > > thanks again for the work.
> > > >
> > > > holger
> > > >
> > > >
> > > >
> > > >
> > > > On Thu, Jun 11, 2015 at 00:10 +0000, Bruno Oliveira wrote:
> > > > > Hi Floris,
> > > > >
> > > > > > What happens to existing pull requests?  If we don't merge them
> > before
> > > > > > Sat we have to convert them to git ourself and re-make the PR?
> > > > >
> > > > > Yes, as there are few PRs and there's no automatic way to do migrate
> > > > them,
> > > > > I think that is a reasonable approach. For PRs that; we should ask
> > > > > submitters to open new PRs in GitHub after the conversion.
> > > > >
> > > > > > But for #296 I'm preparing a PR at
> > > > > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > > > > rebased onto pytest-2.7).  I think I'd prefer to merge that before
> > the
> > > > > > conversion even if not all tests are fixed by Sat (but hopefully
> > they
> > > > > > will be!).  Does that sound reasonable?
> > > > >
> > > > > IMO yes, unless others disagree.
> > > > >
> > > > > Cheers,
> > > > >
> > > > >
> > > > > On Wed, Jun 10, 2015 at 8:47 PM Floris Bruynooghe <flub at devork.be>
> > > > wrote:
> > > > >
> > > > > > On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com>
> > wrote:
> > > > > > > Hi guys,
> > > > > > >
> > > > > > > Just wanted to know if everyone is OK with migrating pytest to
> > GitHub
> > > > > > this
> > > > > > > Saturday (June 13th).
> > > > > > > If all agree, I will send an email Saturday when I start the
> > > > migration
> > > > > > > process (issues + repository), meanwhile no one should commit to
> > > > > > BitBucket
> > > > > > > until the process is complete.
> > > > > >
> > > > > > What happens to existing pull requests?  If we don't merge them
> > before
> > > > > > Sat we have to convert them to git ourself and re-make the PR?
> > > > > >
> > > > > > I think I'm happy to decline #271 and #292.
> > > > > > But for #296 I'm preparing a PR at
> > > > > > https://bitbucket.org/flub/pytest-py35 (because it needed to be
> > > > > > rebased onto pytest-2.7).  I think I'd prefer to merge that before
> > the
> > > > > > conversion even if not all tests are fixed by Sat (but hopefully
> > they
> > > > > > will be!).  Does that sound reasonable?
> > > > > >
> > > > > >
> > > > > > Floris
> > > > > > _______________________________________________
> > > > > > pytest-dev mailing list
> > > > > > pytest-dev at python.org
> > > > > > https://mail.python.org/mailman/listinfo/pytest-dev
> > > > > >
> > > >
> > > > > _______________________________________________
> > > > > pytest-dev mailing list
> > > > > pytest-dev at python.org
> > > > > https://mail.python.org/mailman/listinfo/pytest-dev
> > > >
> > > >
> > > > --
> > > > about me:    http://holgerkrekel.net/about-me/
> > > > contracting: http://merlinux.eu
> > > >
> >
> > --
> > about me:    http://holgerkrekel.net/about-me/
> > contracting: http://merlinux.eu
> >

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From holger at merlinux.eu  Mon Jun 15 11:55:08 2015
From: holger at merlinux.eu (holger krekel)
Date: Mon, 15 Jun 2015 09:55:08 +0000
Subject: [pytest-dev] Adopt Pytest Month the report (draft)
In-Reply-To: <CAPjwCjvZr2vbbyCNvZoRAiph9_ELBF23MsJ3mTBZwvTjChcRKQ@mail.gmail.com>
References: <CAPjwCjvZr2vbbyCNvZoRAiph9_ELBF23MsJ3mTBZwvTjChcRKQ@mail.gmail.com>
Message-ID: <20150615095508.GV29732@merlinux.eu>

Hi Brianna,

On Tue, Jun 09, 2015 at 00:12 +0200, Brianna Laugher wrote:
> Hello!
> 
> Sorry to keep you all in suspense since April... But finally! here is a
> draft of the APM report.
> 
> https://bitbucket.org/pytest-dev/pytest/issue-attachment/676/pytest-dev/pytest/1433801126.73/676/APMreport.pdf
> 
> Although it's publicly accessible I just want to consider it a draft for a
> couple of weeks so the pytest helpers and project maintainers can tell me
> if they have any concerns with what I have written. After that I can update
> it (and the pytest web page I guess).

I like it.  If possible i'd change the layout of the "appended" forms
and survey, email texts to use another font or so.

I think the fact that some efforts happened already before april could
still be counted as a result of APM because it triggered/facilitated it.
In other words, i'd view all effects which can be attributed towards
the APM initiative as being part of the results :)

holger

> Other feedback is welcome, if anything is unclear or you think an
> explanation of anything is missing. You could give feedback here, or
> privately, or on bitbucket (well for the time being!)
> https://bitbucket.org/pytest-dev/pytest/issue/676/adopt-pytest-month-2015
> 
> I am also giving a talk about APM at Europython, I know I'll see some of
> you there. :)
> https://ep2015.europython.eu/conference/talks/the-realities-of-open-source-testing-lessons-learned-from-adopt-pytest-month
> 
> Thanks!
> Brianna
> 
> 
> -- 
> They've just been waiting in a mountain for the right moment:
> http://modernthings.org/

> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev


-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From nicoddemus at gmail.com  Mon Jun 15 14:59:59 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Mon, 15 Jun 2015 12:59:59 +0000
Subject: [pytest-dev] Pytest migration to GitHub
Message-ID: <CA+RQFfc57Tuwqvq88zQtL92nrka=zkx1=97JBx90kZjDn86xFA@mail.gmail.com>

Hi everyone,

Today we will migrate the pytest repository from bitbucket (Hg) to GitHub
(Git), sometime after 6pm (GMT -03:00).

Until then feel free to commit and handle issues as usual, but when the
process starts please refrain to make any changes to issues or new commits
until the process finishes. I will send an email to this list when the
process officially starts, and it is expected to take no longer than 1 hour
before normal work, now on the new repository, can resume.

For those interested, here are the steps for the migration:

1. Create github.com/pytest-dev/pytest and move current issues;
2. Convert pytest Hg repository to Git and upload to GitHub;
3. Add to bitbucket's README a notice about the move to GitHub, and that
new issues/PRs should be posted there;

(At this point, work on the new repository can resume)

4. Ask submitters to re-create PRs at the new repository;
5. Update all links in the documentation and PyPI;
6. Update "how to contribute" docs;
7. Upload new docs to pytest.org;
8. Send an email to all relevant mailing lists about the migration;

Suggestions and comments are welcome!

Cheers,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150615/8bb52516/attachment.html>

From nicoddemus at gmail.com  Mon Jun 15 22:31:07 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Mon, 15 Jun 2015 20:31:07 +0000
Subject: [pytest-dev] Pytest migration to GitHub
In-Reply-To: <CA+RQFfc57Tuwqvq88zQtL92nrka=zkx1=97JBx90kZjDn86xFA@mail.gmail.com>
References: <CA+RQFfc57Tuwqvq88zQtL92nrka=zkx1=97JBx90kZjDn86xFA@mail.gmail.com>
Message-ID: <CA+RQFfcrUd54fF17KqK0=VNRYVTKbhncKoW2aoYxM1C=OfuXkw@mail.gmail.com>

Hi everyone,

The migration process has started now. I will send an email once the first
steps are complete.

Cheers,

On Mon, Jun 15, 2015 at 9:59 AM Bruno Oliveira <nicoddemus at gmail.com> wrote:

> Hi everyone,
>
> Today we will migrate the pytest repository from bitbucket (Hg) to GitHub
> (Git), sometime after 6pm (GMT -03:00).
>
> Until then feel free to commit and handle issues as usual, but when the
> process starts please refrain to make any changes to issues or new commits
> until the process finishes. I will send an email to this list when the
> process officially starts, and it is expected to take no longer than 1 hour
> before normal work, now on the new repository, can resume.
>
> For those interested, here are the steps for the migration:
>
> 1. Create github.com/pytest-dev/pytest and move current issues;
> 2. Convert pytest Hg repository to Git and upload to GitHub;
> 3. Add to bitbucket's README a notice about the move to GitHub, and that
> new issues/PRs should be posted there;
>
> (At this point, work on the new repository can resume)
>
> 4. Ask submitters to re-create PRs at the new repository;
> 5. Update all links in the documentation and PyPI;
> 6. Update "how to contribute" docs;
> 7. Upload new docs to pytest.org;
> 8. Send an email to all relevant mailing lists about the migration;
>
> Suggestions and comments are welcome!
>
> Cheers,
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150615/6aff8951/attachment.html>

From nicoddemus at gmail.com  Mon Jun 15 23:08:01 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Mon, 15 Jun 2015 21:08:01 +0000
Subject: [pytest-dev] Pytest migration to GitHub
In-Reply-To: <CA+RQFfcrUd54fF17KqK0=VNRYVTKbhncKoW2aoYxM1C=OfuXkw@mail.gmail.com>
References: <CA+RQFfc57Tuwqvq88zQtL92nrka=zkx1=97JBx90kZjDn86xFA@mail.gmail.com>
 <CA+RQFfcrUd54fF17KqK0=VNRYVTKbhncKoW2aoYxM1C=OfuXkw@mail.gmail.com>
Message-ID: <CA+RQFfezfQqdJH6AdmtUkoGVg54aPYX9XeY3_fvESghCw=GyWQ@mail.gmail.com>

Hi everyone,

We have our new repository ready:

https://github.com/pytest-dev/pytest

Issues and commits may now continue on the new repository.

The few remaining tasks to fully complete the move are listed here:
https://github.com/pytest-dev/pytest/issues/769

Thanks everyone for your patience!

Cheers,



On Mon, Jun 15, 2015 at 5:31 PM Bruno Oliveira <nicoddemus at gmail.com> wrote:

> Hi everyone,
>
> The migration process has started now. I will send an email once the first
> steps are complete.
>
> Cheers,
>
> On Mon, Jun 15, 2015 at 9:59 AM Bruno Oliveira <nicoddemus at gmail.com>
> wrote:
>
>> Hi everyone,
>>
>> Today we will migrate the pytest repository from bitbucket (Hg) to GitHub
>> (Git), sometime after 6pm (GMT -03:00).
>>
>> Until then feel free to commit and handle issues as usual, but when the
>> process starts please refrain to make any changes to issues or new commits
>> until the process finishes. I will send an email to this list when the
>> process officially starts, and it is expected to take no longer than 1 hour
>> before normal work, now on the new repository, can resume.
>>
>> For those interested, here are the steps for the migration:
>>
>> 1. Create github.com/pytest-dev/pytest and move current issues;
>> 2. Convert pytest Hg repository to Git and upload to GitHub;
>> 3. Add to bitbucket's README a notice about the move to GitHub, and that
>> new issues/PRs should be posted there;
>>
>> (At this point, work on the new repository can resume)
>>
>> 4. Ask submitters to re-create PRs at the new repository;
>> 5. Update all links in the documentation and PyPI;
>> 6. Update "how to contribute" docs;
>> 7. Upload new docs to pytest.org;
>> 8. Send an email to all relevant mailing lists about the migration;
>>
>> Suggestions and comments are welcome!
>>
>> Cheers,
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150615/0e24507c/attachment.html>

From nicoddemus at gmail.com  Tue Jun 16 01:23:07 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Mon, 15 Jun 2015 23:23:07 +0000
Subject: [pytest-dev] ANN: pytest development has moved to GitHub
Message-ID: <CA+RQFfdwAtBRbrP_+YoogX=W4Et7U=t9dzHSCswPH2yXD40StA@mail.gmail.com>

Hi,

pytest development has moved to GitHub at
https://github.com/pytest-dev/pytest. All issues have been migrated as well.

For those interested, the move was the result of requests from several
users and some pytest-core developers, and the reasons for the move can be
found in this issue: https://github.com/pytest-dev/pytest/issues/699.

Cheers,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150615/0f8a2fd8/attachment-0001.html>

From nicoddemus at gmail.com  Tue Jun 16 01:25:15 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Mon, 15 Jun 2015 23:25:15 +0000
Subject: [pytest-dev] Pytest migration to GitHub
In-Reply-To: <CA+RQFfezfQqdJH6AdmtUkoGVg54aPYX9XeY3_fvESghCw=GyWQ@mail.gmail.com>
References: <CA+RQFfc57Tuwqvq88zQtL92nrka=zkx1=97JBx90kZjDn86xFA@mail.gmail.com>
 <CA+RQFfcrUd54fF17KqK0=VNRYVTKbhncKoW2aoYxM1C=OfuXkw@mail.gmail.com>
 <CA+RQFfezfQqdJH6AdmtUkoGVg54aPYX9XeY3_fvESghCw=GyWQ@mail.gmail.com>
Message-ID: <CA+RQFffzMmLFCNdkY6YohV6m=e2EMSoeGpmMBe=__imt-MUbmQ@mail.gmail.com>

Hi everyone,

Here's the list of the tasks done for the move to GitHub:
https://github.com/pytest-dev/pytest/issues/769

If you see any obsolete link or documentation piece that needs to be
updated, please don't refrain to point it out or creating a PR.

Cheers,

On Mon, Jun 15, 2015 at 6:07 PM Bruno Oliveira <nicoddemus at gmail.com> wrote:

> Hi everyone,
>
> We have our new repository ready:
>
> https://github.com/pytest-dev/pytest
>
> Issues and commits may now continue on the new repository.
>
> The few remaining tasks to fully complete the move are listed here:
> https://github.com/pytest-dev/pytest/issues/769
>
> Thanks everyone for your patience!
>
> Cheers,
>
>
>
> On Mon, Jun 15, 2015 at 5:31 PM Bruno Oliveira <nicoddemus at gmail.com>
> wrote:
>
>> Hi everyone,
>>
>> The migration process has started now. I will send an email once the
>> first steps are complete.
>>
>> Cheers,
>>
>> On Mon, Jun 15, 2015 at 9:59 AM Bruno Oliveira <nicoddemus at gmail.com>
>> wrote:
>>
>>> Hi everyone,
>>>
>>> Today we will migrate the pytest repository from bitbucket (Hg) to
>>> GitHub (Git), sometime after 6pm (GMT -03:00).
>>>
>>> Until then feel free to commit and handle issues as usual, but when the
>>> process starts please refrain to make any changes to issues or new commits
>>> until the process finishes. I will send an email to this list when the
>>> process officially starts, and it is expected to take no longer than 1 hour
>>> before normal work, now on the new repository, can resume.
>>>
>>> For those interested, here are the steps for the migration:
>>>
>>> 1. Create github.com/pytest-dev/pytest and move current issues;
>>> 2. Convert pytest Hg repository to Git and upload to GitHub;
>>> 3. Add to bitbucket's README a notice about the move to GitHub, and that
>>> new issues/PRs should be posted there;
>>>
>>> (At this point, work on the new repository can resume)
>>>
>>> 4. Ask submitters to re-create PRs at the new repository;
>>> 5. Update all links in the documentation and PyPI;
>>> 6. Update "how to contribute" docs;
>>> 7. Upload new docs to pytest.org;
>>> 8. Send an email to all relevant mailing lists about the migration;
>>>
>>> Suggestions and comments are welcome!
>>>
>>> Cheers,
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150615/ab8ad86e/attachment.html>

From bubenkoff at gmail.com  Tue Jun 16 04:20:56 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Tue, 16 Jun 2015 02:20:56 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150615093513.GR29732@merlinux.eu>
References: <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
Message-ID: <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>

after such a success with pytest move, should we start the same process for
other repos under pytest-dev on bb?

On Mon, Jun 15, 2015 at 11:35 AM holger krekel <holger at merlinux.eu> wrote:

> On Sat, Jun 13, 2015 at 16:32 +0000, Bruno Oliveira wrote:
> > > Yes.  Can you announce in a top-level post here on the list along
> > > with the proposed way of what's going to happen?
> >
> > Sure, although I don't think all PRs (currently 6) will be
> merged/rejected
> > until Monday... should we migrate anyway on Monday, or give it a few more
> > days to sort the remaining PRs out and do the actual migration, say,
> > Wednesday?
> >
> > Btw, made another round of migration (issues + repository) at
> > https://github.com/pytestbot/pytest-issues-migration if anybody wants to
> > take another look.
>
> we can go for migration anyway, i am now on IRC.
>
> holger
>
> > Cheers,
> >
> >
> > On Sat, Jun 13, 2015 at 4:05 AM holger krekel <holger at merlinux.eu>
> wrote:
> >
> > > On Fri, Jun 12, 2015 at 12:10 +0000, Bruno Oliveira wrote:
> > > > Hi Holger,
> > > >
> > > > > I am not going to be much available during the weekend and would
> > > prefer a
> > > > > switch date on monday.  It's be best to have PRs merged or rejected
> > > until
> > > > > then.
> > > >
> > > > I suggested Saturday because usually there's not much commits
> happening,
> > > > but Monday works for me too.
> > > >
> > > > > As to the migrated issues: i just tried with two random user
> handles --
> > > > > the bitbucket userids map 1:1 to a github one -- so we could add a
> link
> > > > > to the github user (stating "maybe it's this github user: ..."
> because
> > > > > we can't be 100% sure in all cases where the user id exists).
> > > >
> > > > I thought about this, but was concerned that while there might be two
> > > > identical bitbucket and GitHub user ids, they might not belong to the
> > > same
> > > > person.
> > > > In practice I don't think it will be a problem though, so I will add
> a
> > > > another link to the GitHub user id then.
> > >
> > > yes, i agree that we don't have 100% safety -- therefore my "maybe it's
> > > this github user" suggestion for a text in the issue.
> > >
> > > > > Also is there a way to, after migration, post to each bitbucket
> issue
> > > > > the link to the corresponding github issue?
> > > >
> > > > Yes, good idea, I will implement this as well.
> > >
> > > i guess that this should be a "closing" post so that people
> > > recognize they need to use github.
> > >
> > > > > thanks again for the work.
> > > >
> > > > No problem, thanks for the feedback.
> > > >
> > > > Can we set the official migration date to Monday then?
> > >
> > > Yes.  Can you announce in a top-level post here on the list along
> > > with the proposed way of what's going to happen?
> > >
> > >
> > > holger
> > >
> > >
> > > >
> > > > On Fri, Jun 12, 2015 at 8:26 AM holger krekel <holger at merlinux.eu>
> > > wrote:
> > > >
> > > > > Hi Bruno,
> > > > >
> > > > > was on travel, then ill, initial scan of your work looks great!
> > > > >
> > > > > I am not going to be much available during the weekend and would
> > > prefer a
> > > > > switch date on monday.  It's be best to have PRs merged or rejected
> > > until
> > > > > then.
> > > > >
> > > > > As to the migrated issues: i just tried with two random user
> handles --
> > > > > the bitbucket userids map 1:1 to a github one -- so we could add a
> link
> > > > > to the github user (stating "maybe it's this github user: ..."
> because
> > > > > we can't be 100% sure in all cases where the user id exists).
> > > > >
> > > > > Also is there a way to, after migration, post to each bitbucket
> issue
> > > > > the link to the corresponding github issue?
> > > > >
> > > > > thanks again for the work.
> > > > >
> > > > > holger
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > On Thu, Jun 11, 2015 at 00:10 +0000, Bruno Oliveira wrote:
> > > > > > Hi Floris,
> > > > > >
> > > > > > > What happens to existing pull requests?  If we don't merge them
> > > before
> > > > > > > Sat we have to convert them to git ourself and re-make the PR?
> > > > > >
> > > > > > Yes, as there are few PRs and there's no automatic way to do
> migrate
> > > > > them,
> > > > > > I think that is a reasonable approach. For PRs that; we should
> ask
> > > > > > submitters to open new PRs in GitHub after the conversion.
> > > > > >
> > > > > > > But for #296 I'm preparing a PR at
> > > > > > > https://bitbucket.org/flub/pytest-py35 (because it needed to
> be
> > > > > > > rebased onto pytest-2.7).  I think I'd prefer to merge that
> before
> > > the
> > > > > > > conversion even if not all tests are fixed by Sat (but
> hopefully
> > > they
> > > > > > > will be!).  Does that sound reasonable?
> > > > > >
> > > > > > IMO yes, unless others disagree.
> > > > > >
> > > > > > Cheers,
> > > > > >
> > > > > >
> > > > > > On Wed, Jun 10, 2015 at 8:47 PM Floris Bruynooghe <
> flub at devork.be>
> > > > > wrote:
> > > > > >
> > > > > > > On 10 June 2015 at 14:07, Bruno Oliveira <nicoddemus at gmail.com
> >
> > > wrote:
> > > > > > > > Hi guys,
> > > > > > > >
> > > > > > > > Just wanted to know if everyone is OK with migrating pytest
> to
> > > GitHub
> > > > > > > this
> > > > > > > > Saturday (June 13th).
> > > > > > > > If all agree, I will send an email Saturday when I start the
> > > > > migration
> > > > > > > > process (issues + repository), meanwhile no one should
> commit to
> > > > > > > BitBucket
> > > > > > > > until the process is complete.
> > > > > > >
> > > > > > > What happens to existing pull requests?  If we don't merge them
> > > before
> > > > > > > Sat we have to convert them to git ourself and re-make the PR?
> > > > > > >
> > > > > > > I think I'm happy to decline #271 and #292.
> > > > > > > But for #296 I'm preparing a PR at
> > > > > > > https://bitbucket.org/flub/pytest-py35 (because it needed to
> be
> > > > > > > rebased onto pytest-2.7).  I think I'd prefer to merge that
> before
> > > the
> > > > > > > conversion even if not all tests are fixed by Sat (but
> hopefully
> > > they
> > > > > > > will be!).  Does that sound reasonable?
> > > > > > >
> > > > > > >
> > > > > > > Floris
> > > > > > > _______________________________________________
> > > > > > > pytest-dev mailing list
> > > > > > > pytest-dev at python.org
> > > > > > > https://mail.python.org/mailman/listinfo/pytest-dev
> > > > > > >
> > > > >
> > > > > > _______________________________________________
> > > > > > pytest-dev mailing list
> > > > > > pytest-dev at python.org
> > > > > > https://mail.python.org/mailman/listinfo/pytest-dev
> > > > >
> > > > >
> > > > > --
> > > > > about me:    http://holgerkrekel.net/about-me/
> > > > > contracting: http://merlinux.eu
> > > > >
> > >
> > > --
> > > about me:    http://holgerkrekel.net/about-me/
> > > contracting: http://merlinux.eu
> > >
>
> --
> about me:    http://holgerkrekel.net/about-me/
> contracting: http://merlinux.eu
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150616/78321482/attachment-0001.html>

From andreas at pelme.se  Tue Jun 16 16:52:56 2015
From: andreas at pelme.se (Andreas Pelme)
Date: Tue, 16 Jun 2015 16:52:56 +0200
Subject: [pytest-dev] trainer needed for pytest training at EP2015
In-Reply-To: <E656ADAE-F5FC-4050-AA6B-BDD1C2428ECE@pelme.se>
References: <20150602093049.GP29732@merlinux.eu> <20150603093148.GP26357@tonks>
 <E656ADAE-F5FC-4050-AA6B-BDD1C2428ECE@pelme.se>
Message-ID: <DFE6EE64-90EF-4134-B12A-8E97F72A439D@pelme.se>

Hi,


> On 3 jun 2015, at 11:43, Andreas Pelme <andreas at pelme.se> wrote:
> 
> I helped out last year and it was a lot of fun. I have not finalized my plans for EP and am not yet sure I will be able to make it. If I?m there, I will for sure be able help out and assist as much as I can.


I have just booked my tickets for EP, so I will be there and available to help out with the tutorial! :-)

Cheers,
Andreas

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 204 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150616/38642608/attachment.sig>

From tibor.arpas at infinit.sk  Wed Jun 17 10:54:13 2015
From: tibor.arpas at infinit.sk (Tibor Arpas)
Date: Wed, 17 Jun 2015 10:54:13 +0200
Subject: [pytest-dev] EP2015 py.test gathering
Message-ID: <CADi_WY1++Y7daygpxUfKMFQxd2ZR+SD=gjmRaJArj+cZUs9EKw@mail.gmail.com>

Great,

I also booked everything yesterday. I'll be presenting "MASHING UP PY.TEST,
COVERAGE.PY AND AST.PY TO TAKE TDD TO A NEW LEVEL" (about pytest-testmon)
on Tuesday 15:45 :)

Will the developers and interested users of py.test meet at some time at
the venue and talk? Let us know please on this list if something like that
happens, I would love to attend, but I usually forget to follow twitter :).

Tibor


On Tue, Jun 16, 2015 at 4:52 PM, Andreas Pelme <andreas at pelme.se> wrote:

> Hi,
>
>
> > On 3 jun 2015, at 11:43, Andreas Pelme <andreas at pelme.se> wrote:
> >
> > I helped out last year and it was a lot of fun. I have not finalized my
> plans for EP and am not yet sure I will be able to make it. If I?m there, I
> will for sure be able help out and assist as much as I can.
>
>
> I have just booked my tickets for EP, so I will be there and available to
> help out with the tutorial! :-)
>
> Cheers,
> Andreas
>
>
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150617/7d10635f/attachment.html>

From nicoddemus at gmail.com  Wed Jun 17 14:31:44 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Wed, 17 Jun 2015 12:31:44 +0000
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
Message-ID: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>

Hi, Do we have an idea of when we would like have a 2.8.0 release, or at
least what we would like to see in the new version? And about 2.7.2? I see
there are a couple of fixes that are in master but should be backported to
pytest-2.7 branch before making a 2.7.2 release. Cheers,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150617/4d1e2adb/attachment.html>

From nicoddemus at gmail.com  Thu Jun 18 06:03:05 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 18 Jun 2015 04:03:05 +0000
Subject: [pytest-dev] Trove Classifier for pytest plugins
Message-ID: <CA+RQFfdKbjPjf86YO4RNvZgBoByDSrW=V51VGO-S8OfutaZcPA@mail.gmail.com>

Hi,

Recently Jason R. Coombs proposed to create an specific trove classifier
for pytest plugins:

https://github.com/pytest-dev/plugincompat/issues/16

One benefit is that it would make it easier for people to find pytest
plugins, as the recommended approach of searching for "pytest-" named
packages leaves some of the existing plugins out.

I like the idea, but would like to hear from the people here first.

Cheers,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/198c67fc/attachment.html>

From flub at devork.be  Thu Jun 18 07:06:35 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Wed, 17 Jun 2015 22:06:35 -0700
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
Message-ID: <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>

On 17 June 2015 at 05:31, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> Hi, Do we have an idea of when we would like have a 2.8.0 release, or at
> least what we would like to see in the new version?

Nope, maybe the pluggy dependency?  But not sure if that was the plan.

> And about 2.7.2?

I think we should release that as soon as the python 3.5 compatibility
is sorted.  I'm on holiday for a few days though so won't get to look
at this again until mid-next week (but don't let that stop someone
else ;-)).

I think Ronny had some 2.7.2 plans as well or wanted it soon for some reason?

> I see
> there are a couple of fixes that are in master but should be backported to
> pytest-2.7 branch before making a 2.7.2 release.

Hmm, if they where fixes I guess they should have gone to 2.7 right
away.  Sometimes the line is a bit blurry though.  It obviously
doesn't matter much if 2.8 overtakes 2.7.2, so if there's nothing
important that needs to happen for 2.8 then that would be fine too.

Regards,
Floris

From flub at devork.be  Thu Jun 18 07:15:57 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Wed, 17 Jun 2015 22:15:57 -0700
Subject: [pytest-dev] EP2015 py.test gathering
In-Reply-To: <CADi_WY1++Y7daygpxUfKMFQxd2ZR+SD=gjmRaJArj+cZUs9EKw@mail.gmail.com>
References: <CADi_WY1++Y7daygpxUfKMFQxd2ZR+SD=gjmRaJArj+cZUs9EKw@mail.gmail.com>
Message-ID: <CAAWk_DzgprjGz0N_2zfvAArpVns=rgOhNK=k0bRiguEj11kGDw@mail.gmail.com>

Hi Tibor,

On 17 June 2015 at 01:54, Tibor Arpas <tibor.arpas at infinit.sk> wrote:
> I also booked everything yesterday. I'll be presenting "MASHING UP PY.TEST,
> COVERAGE.PY AND AST.PY TO TAKE TDD TO A NEW LEVEL" (about pytest-testmon) on
> Tuesday 15:45 :)

Nice, look forward to it.

> Will the developers and interested users of py.test meet at some time at the
> venue and talk? Let us know please on this list if something like that
> happens, I would love to attend, but I usually forget to follow twitter :).

I'm sure we will meet up at some point, would be great meet you as
well.  A bit long in advance right now but I'm sure we can use the
list to organise during EP.  Maybe it even makes sense to have some
wider meetup with users and adopt-py.test-month participants as a BOF
or so?

Regards,
Floris

From holger at merlinux.eu  Thu Jun 18 07:47:26 2015
From: holger at merlinux.eu (holger krekel)
Date: Thu, 18 Jun 2015 05:47:26 +0000
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
Message-ID: <20150618054726.GN29732@merlinux.eu>

On Wed, Jun 17, 2015 at 22:06 -0700, Floris Bruynooghe wrote:
> On 17 June 2015 at 05:31, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > Hi, Do we have an idea of when we would like have a 2.8.0 release, or at
> > least what we would like to see in the new version?
> 
> Nope, maybe the pluggy dependency?  But not sure if that was the plan.

FWIW I consider to vendor the pluggy dependency before pytest-2.8 is released.
The main issue is that as projects start using pluggy indepdently of pytest
there might be version conflicts when pluggy gets new releases, making
it hard to test pluggy-using projects with pytest.

> > And about 2.7.2?
> 
> I think we should release that as soon as the python 3.5 compatibility
> is sorted.  I'm on holiday for a few days though so won't get to look
> at this again until mid-next week (but don't let that stop someone
> else ;-)).
>
> I think Ronny had some 2.7.2 plans as well or wanted it soon for some reason?
> 
> > I see
> > there are a couple of fixes that are in master but should be backported to
> > pytest-2.7 branch before making a 2.7.2 release.
> 
> Hmm, if they where fixes I guess they should have gone to 2.7 right
> away.  Sometimes the line is a bit blurry though.  It obviously
> doesn't matter much if 2.8 overtakes 2.7.2, so if there's nothing
> important that needs to happen for 2.8 then that would be fine too.

Maybe we can just aim for pytest-2.8 in the next two weeks and forget
about pytest-2.7.2 and then see to have more clarity regarding PR targets.
Would it make sense to consider "master" to become the new "bug fix branch"
and have a "pytest-2.9" then where we collect new features? This way
the default is to do bugfixes which might be easier for newcomers to
the project.

best,
holger

From tibor.arpas at infinit.sk  Thu Jun 18 10:22:40 2015
From: tibor.arpas at infinit.sk (Tibor Arpas)
Date: Thu, 18 Jun 2015 10:22:40 +0200
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <20150618054726.GN29732@merlinux.eu>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
 <20150618054726.GN29732@merlinux.eu>
Message-ID: <CADi_WY1sVcNbrphWho+6LCvLUEZaW-=qtP0xCSorjv7KVkfQaA@mail.gmail.com>

How about the new hook call system vs the old __multicall__ ? Is the plan
still to deprecate the old way in 2.8 and remove in 2.9?

Any idea how many plugin-s use it and will brake (or were already updated)?

Tibor

On Thu, Jun 18, 2015 at 7:47 AM, holger krekel <holger at merlinux.eu> wrote:

> On Wed, Jun 17, 2015 at 22:06 -0700, Floris Bruynooghe wrote:
> > On 17 June 2015 at 05:31, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > > Hi, Do we have an idea of when we would like have a 2.8.0 release, or
> at
> > > least what we would like to see in the new version?
> >
> > Nope, maybe the pluggy dependency?  But not sure if that was the plan.
>
> FWIW I consider to vendor the pluggy dependency before pytest-2.8 is
> released.
> The main issue is that as projects start using pluggy indepdently of pytest
> there might be version conflicts when pluggy gets new releases, making
> it hard to test pluggy-using projects with pytest.
>
> > > And about 2.7.2?
> >
> > I think we should release that as soon as the python 3.5 compatibility
> > is sorted.  I'm on holiday for a few days though so won't get to look
> > at this again until mid-next week (but don't let that stop someone
> > else ;-)).
> >
> > I think Ronny had some 2.7.2 plans as well or wanted it soon for some
> reason?
> >
> > > I see
> > > there are a couple of fixes that are in master but should be
> backported to
> > > pytest-2.7 branch before making a 2.7.2 release.
> >
> > Hmm, if they where fixes I guess they should have gone to 2.7 right
> > away.  Sometimes the line is a bit blurry though.  It obviously
> > doesn't matter much if 2.8 overtakes 2.7.2, so if there's nothing
> > important that needs to happen for 2.8 then that would be fine too.
>
> Maybe we can just aim for pytest-2.8 in the next two weeks and forget
> about pytest-2.7.2 and then see to have more clarity regarding PR targets.
> Would it make sense to consider "master" to become the new "bug fix branch"
> and have a "pytest-2.9" then where we collect new features? This way
> the default is to do bugfixes which might be easier for newcomers to
> the project.
>
> best,
> holger
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/ee740c12/attachment-0001.html>

From holger at merlinux.eu  Thu Jun 18 10:42:41 2015
From: holger at merlinux.eu (holger krekel)
Date: Thu, 18 Jun 2015 08:42:41 +0000
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <CADi_WY1sVcNbrphWho+6LCvLUEZaW-=qtP0xCSorjv7KVkfQaA@mail.gmail.com>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
 <20150618054726.GN29732@merlinux.eu>
 <CADi_WY1sVcNbrphWho+6LCvLUEZaW-=qtP0xCSorjv7KVkfQaA@mail.gmail.com>
Message-ID: <20150618084241.GP29732@merlinux.eu>

On Thu, Jun 18, 2015 at 10:22 +0200, Tibor Arpas wrote:
> How about the new hook call system vs the old __multicall__ ? Is the plan
> still to deprecate the old way in 2.8 and remove in 2.9?

yes, that's the plan.

> Any idea how many plugin-s use it and will brake (or were already updated)?

IIRC most pytest-dev plugins don't use __multicall__ last time i checked.

holger


> Tibor
> 
> On Thu, Jun 18, 2015 at 7:47 AM, holger krekel <holger at merlinux.eu> wrote:
> 
> > On Wed, Jun 17, 2015 at 22:06 -0700, Floris Bruynooghe wrote:
> > > On 17 June 2015 at 05:31, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > > > Hi, Do we have an idea of when we would like have a 2.8.0 release, or
> > at
> > > > least what we would like to see in the new version?
> > >
> > > Nope, maybe the pluggy dependency?  But not sure if that was the plan.
> >
> > FWIW I consider to vendor the pluggy dependency before pytest-2.8 is
> > released.
> > The main issue is that as projects start using pluggy indepdently of pytest
> > there might be version conflicts when pluggy gets new releases, making
> > it hard to test pluggy-using projects with pytest.
> >
> > > > And about 2.7.2?
> > >
> > > I think we should release that as soon as the python 3.5 compatibility
> > > is sorted.  I'm on holiday for a few days though so won't get to look
> > > at this again until mid-next week (but don't let that stop someone
> > > else ;-)).
> > >
> > > I think Ronny had some 2.7.2 plans as well or wanted it soon for some
> > reason?
> > >
> > > > I see
> > > > there are a couple of fixes that are in master but should be
> > backported to
> > > > pytest-2.7 branch before making a 2.7.2 release.
> > >
> > > Hmm, if they where fixes I guess they should have gone to 2.7 right
> > > away.  Sometimes the line is a bit blurry though.  It obviously
> > > doesn't matter much if 2.8 overtakes 2.7.2, so if there's nothing
> > > important that needs to happen for 2.8 then that would be fine too.
> >
> > Maybe we can just aim for pytest-2.8 in the next two weeks and forget
> > about pytest-2.7.2 and then see to have more clarity regarding PR targets.
> > Would it make sense to consider "master" to become the new "bug fix branch"
> > and have a "pytest-2.9" then where we collect new features? This way
> > the default is to do bugfixes which might be easier for newcomers to
> > the project.
> >
> > best,
> > holger
> > _______________________________________________
> > pytest-dev mailing list
> > pytest-dev at python.org
> > https://mail.python.org/mailman/listinfo/pytest-dev
> >

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From me at the-compiler.org  Thu Jun 18 11:23:00 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Thu, 18 Jun 2015 11:23:00 +0200
Subject: [pytest-dev] plugins/pytest-qt and cx_Freeze
Message-ID: <20150618092300.GE22364@tonks>

Hi,

I'm currently trying to run my tests with my application frozen via
cx_Freeze.

To do so, I wrote a separate script to freeze the tests into a
separate executable, which uses pytest.freeze_includes():

https://github.com/The-Compiler/qutebrowser/blob/frozen-tests/scripts/freeze_tests.py
https://github.com/The-Compiler/qutebrowser/blob/frozen-tests/scripts/run_frozen_tests.py

Apart of some issues with importing distutils (which I haven't looked
into yet), it seems pytest doesn't find the qapp fixture which is
provided by pytest-qt:

http://paste.the-compiler.org/view/7b1dbbf2

Is there something I can do to make cx_Freeze freeze those plugins as
well, or to make pytest recognize them when frozen?

I tried adding pytestqt to includes=[...], but that didn't help
either.

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/f11d5504/attachment.sig>

From nicoddemus at gmail.com  Thu Jun 18 15:50:59 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 18 Jun 2015 13:50:59 +0000
Subject: [pytest-dev] plugins/pytest-qt and cx_Freeze
In-Reply-To: <20150618092300.GE22364@tonks>
References: <20150618092300.GE22364@tonks>
Message-ID: <CA+RQFfcHD-=b+yo8j1KBQkHp+-D0Kotr9ORa9sgeTpngkjN9Fw@mail.gmail.com>

Hi Florian,

I have this problem as well and was trying to find a better solution before
bringing up this issue, but let me share my solution so far.

The problem is that the mechanism for plugin discovered used by pytest
(setupttools entry points) doesn't work with frozen executables so pytest
can't find any plugins.

The workaround I currently use is to add this to my `conftest.py`:

def pytest_configure(config):
    # Manually import pytest-qt because disutils cannot find it on its own
when running tests inside
    # an executable.
    if getattr(sys, 'frozen', False):
        import pytestqt.plugin
        config.pluginmanager.register(pytestqt.plugin, 'pytest-qt')
        # and any other plugins

Another solution would be to explicitly pass the plugin in the command line
using `-p pytestq.plugin`.

I still would like for this to work out of the box somehow though, so if
anyone has any suggestions/ideas they are welcome.


On Thu, Jun 18, 2015 at 6:28 AM Florian Bruhin <me at the-compiler.org> wrote:

> Hi,
>
> I'm currently trying to run my tests with my application frozen via
> cx_Freeze.
>
> To do so, I wrote a separate script to freeze the tests into a
> separate executable, which uses pytest.freeze_includes():
>
>
> https://github.com/The-Compiler/qutebrowser/blob/frozen-tests/scripts/freeze_tests.py
>
> https://github.com/The-Compiler/qutebrowser/blob/frozen-tests/scripts/run_frozen_tests.py
>
> Apart of some issues with importing distutils (which I haven't looked
> into yet), it seems pytest doesn't find the qapp fixture which is
> provided by pytest-qt:
>
> http://paste.the-compiler.org/view/7b1dbbf2
>
> Is there something I can do to make cx_Freeze freeze those plugins as
> well, or to make pytest recognize them when frozen?
>
> I tried adding pytestqt to includes=[...], but that didn't help
> either.
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/b90b1e4b/attachment.html>

From me at the-compiler.org  Thu Jun 18 16:23:47 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Thu, 18 Jun 2015 16:23:47 +0200
Subject: [pytest-dev] plugins/pytest-qt and cx_Freeze
In-Reply-To: <CA+RQFfcHD-=b+yo8j1KBQkHp+-D0Kotr9ORa9sgeTpngkjN9Fw@mail.gmail.com>
References: <20150618092300.GE22364@tonks>
 <CA+RQFfcHD-=b+yo8j1KBQkHp+-D0Kotr9ORa9sgeTpngkjN9Fw@mail.gmail.com>
Message-ID: <20150618142347.GH22364@tonks>

Hey,

* Bruno Oliveira <nicoddemus at gmail.com> [2015-06-18 13:50:59 +0000]:
> The problem is that the mechanism for plugin discovered used by pytest
> (setupttools entry points) doesn't work with frozen executables so pytest
> can't find any plugins.

Hm, I see. I'm not really aware how setuptools entry points are
implemented, and if support for that could be added to cx_Freeze.

> The workaround I currently use is to add this to my `conftest.py`:
> 
> [...]
> 
> Another solution would be to explicitly pass the plugin in the command line
> using `-p pytestq.plugin`.

Hmm, I see. I guess that'd work with the 3 plugins I need for now, but
it's definitely less than ideal.

> I still would like for this to work out of the box somehow though, so if
> anyone has any suggestions/ideas they are welcome.

I have no idea where to start - but if I can help in any way, please
let me know!

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/219b281b/attachment.sig>

From me at the-compiler.org  Thu Jun 18 17:21:10 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Thu, 18 Jun 2015 17:21:10 +0200
Subject: [pytest-dev] plugins/pytest-qt and cx_Freeze
In-Reply-To: <CA+RQFfcHD-=b+yo8j1KBQkHp+-D0Kotr9ORa9sgeTpngkjN9Fw@mail.gmail.com>
References: <20150618092300.GE22364@tonks>
 <CA+RQFfcHD-=b+yo8j1KBQkHp+-D0Kotr9ORa9sgeTpngkjN9Fw@mail.gmail.com>
Message-ID: <20150618152110.GI22364@tonks>

> Another solution would be to explicitly pass the plugin in the command line
> using `-p pytestq.plugin`.

Sorry for the double-post - I found an (IMHO) slightly better solution
:)

As said before I'm using a separate test executable, and I noticed
pytest.main takes a plugins-argument with a list of plugin entry
points.

I now changed the script to import the plugins and run pytest
accordingly:

https://github.com/The-Compiler/qutebrowser/commit/060a7ea044cfac140c53d84258dc2857d058dfc2

Contrary to the other solutions, this makes cx_Freeze realize it
should also freeze those modules, as they're imported in the script
which gets frozen - so there's less duplication.

The only thing I had to give cx_Freeze a little hint about is the
import of the PyQt5.QtTest module, as it's unused in my project
otherwise, but pytest-qt uses it - and it uses a compatibility layer
which causes cx_Freeze to not pick this up.

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/e93e515b/attachment.sig>

From nicoddemus at gmail.com  Thu Jun 18 17:24:24 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 18 Jun 2015 15:24:24 +0000
Subject: [pytest-dev] plugins/pytest-qt and cx_Freeze
In-Reply-To: <20150618152110.GI22364@tonks>
References: <20150618092300.GE22364@tonks>
 <CA+RQFfcHD-=b+yo8j1KBQkHp+-D0Kotr9ORa9sgeTpngkjN9Fw@mail.gmail.com>
 <20150618152110.GI22364@tonks>
Message-ID: <CA+RQFfe5LTzybeSbKXJsADoO0D-Y9kp+T78v7s7Tc_ZQ0rXggQ@mail.gmail.com>

Nice, thanks for sharing! :)

Cheers,

On Thu, Jun 18, 2015 at 12:21 PM Florian Bruhin <me at the-compiler.org> wrote:

> > Another solution would be to explicitly pass the plugin in the command
> line
> > using `-p pytestq.plugin`.
>
> Sorry for the double-post - I found an (IMHO) slightly better solution
> :)
>
> As said before I'm using a separate test executable, and I noticed
> pytest.main takes a plugins-argument with a list of plugin entry
> points.
>
> I now changed the script to import the plugins and run pytest
> accordingly:
>
>
> https://github.com/The-Compiler/qutebrowser/commit/060a7ea044cfac140c53d84258dc2857d058dfc2
>
> Contrary to the other solutions, this makes cx_Freeze realize it
> should also freeze those modules, as they're imported in the script
> which gets frozen - so there's less duplication.
>
> The only thing I had to give cx_Freeze a little hint about is the
> import of the PyQt5.QtTest module, as it's unused in my project
> otherwise, but pytest-qt uses it - and it uses a compatibility layer
> which causes cx_Freeze to not pick this up.
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/66b7fbc6/attachment.html>

From tibor.arpas at infinit.sk  Thu Jun 18 17:56:09 2015
From: tibor.arpas at infinit.sk (Tibor Arpas)
Date: Thu, 18 Jun 2015 17:56:09 +0200
Subject: [pytest-dev] pytest plugins popularity
Message-ID: <CADi_WY3MKJZuSEZ8R+if-PjSAN6U6BMSTUTyr58FqgTmFUvfDQ@mail.gmail.com>

Hi,

I was curious how much usage does pytest and it's plugins get. Of course
that's quite impossible to find out. The only very loosely related thing is
a statistics of downloads from PyPi .

I tried to pick download counts of releases older than 15 days but the
script might have had a bug anyway. The results seem quite plausible so
here you go.

The top results:
[image: Inline image 1]

Later version of the script, printing CSV:

import datetime
from xmlrpclib import ServerProxy

now = datetime.datetime.now()

package_stats = []

client = ServerProxy('https://pypi.python.org/pypi', use_datetime=True)


for package in client.search({'name': 'pytest'}):
    name = package['name']
    if name.startswith('pytest'):
        days_ago = 0
        downloads = 0
        release = None
        for release in client.package_releases(name):
            for url in client.release_urls(name, release):
                days_ago = round((now -
url['upload_time']).total_seconds()/3600/24, 1)
                downloads += url['downloads']
            break
        package_stats.append([name, release, days_ago, downloads])


for stat in package_stats:
    print(",".join([str(s) for s in stat]))

---
Tibor
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/23a2b2e6/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 100703 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/23a2b2e6/attachment-0001.png>

From nicoddemus at gmail.com  Thu Jun 18 18:08:02 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 18 Jun 2015 16:08:02 +0000
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <20150618054726.GN29732@merlinux.eu>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
 <20150618054726.GN29732@merlinux.eu>
Message-ID: <CA+RQFffNQuO=ATtUGLCK7Qg0iDgCjXpt=T_f60uaWQTsmHwtDA@mail.gmail.com>

On Thu, Jun 18, 2015 at 2:47 AM holger krekel <holger at merlinux.eu> wrote:

> On Wed, Jun 17, 2015 at 22:06 -0700, Floris Bruynooghe wrote:
> > I think Ronny had some 2.7.2 plans as well or wanted it soon for some
> reason?
>

pytest-cache integration into the core perhaps?


> Maybe we can just aim for pytest-2.8 in the next two weeks and forget
> about pytest-2.7.2 and then see to have more clarity regarding PR targets.
>

That would be great, as it would avoid having to review the backlog and
port bug issues that went directly into master.

Would it make sense to consider "master" to become the new "bug fix branch"
> and have a "pytest-2.9" then where we collect new features? This way
> the default is to do bugfixes which might be easier for newcomers to
> the project.
>

I was thinking along the same lines this week. As most of the contributions
are bug-fixes, it makes sense to make life easier for PR contributors.

Cheers,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/cd59caf3/attachment.html>

From nicoddemus at gmail.com  Thu Jun 18 19:41:03 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 18 Jun 2015 17:41:03 +0000
Subject: [pytest-dev] pytest plugins popularity
In-Reply-To: <CADi_WY3MKJZuSEZ8R+if-PjSAN6U6BMSTUTyr58FqgTmFUvfDQ@mail.gmail.com>
References: <CADi_WY3MKJZuSEZ8R+if-PjSAN6U6BMSTUTyr58FqgTmFUvfDQ@mail.gmail.com>
Message-ID: <CA+RQFfdT2wsx-k7-fshRSJiC_8fRYtDnxKvT2Ov2CSiHJHpTdw@mail.gmail.com>

On Thu, Jun 18, 2015 at 1:07 PM Tibor Arpas <tibor.arpas at infinit.sk> wrote:

> Hi,
>
> I was curious how much usage does pytest and it's plugins get. Of course
> that's quite impossible to find out. The only very loosely related thing is
> a statistics of downloads from PyPi .
>
> I tried to pick download counts of releases older than 15 days but the
> script might have had a bug anyway. The results seem quite plausible so
> here you go.
>

Interesting, thanks for sharing :)

Cheers,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/8e6a9095/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 100703 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/8e6a9095/attachment-0001.png>

From lac at openend.se  Thu Jun 18 23:49:28 2015
From: lac at openend.se (Laura Creighton)
Date: Thu, 18 Jun 2015 23:49:28 +0200
Subject: [pytest-dev] Trove Classifier for pytest plugins
In-Reply-To: Message from Bruno Oliveira <nicoddemus@gmail.com> of "Thu,
 18 Jun 2015 04:03:05 -0000."
 <CA+RQFfdKbjPjf86YO4RNvZgBoByDSrW=V51VGO-S8OfutaZcPA@mail.gmail.com>
References: <CA+RQFfdKbjPjf86YO4RNvZgBoByDSrW=V51VGO-S8OfutaZcPA@mail.gmail.com>
Message-ID: <201506182149.t5ILnSne009339@fido.openend.se>

In a message of Thu, 18 Jun 2015 04:03:05 -0000, Bruno Oliveira writes:
>Hi,
>
>Recently Jason R. Coombs proposed to create an specific trove classifier
>for pytest plugins:
>
>https://github.com/pytest-dev/plugincompat/issues/16
>
>One benefit is that it would make it easier for people to find pytest
>plugins, as the recommended approach of searching for "pytest-" named
>packages leaves some of the existing plugins out.
>
>I like the idea, but would like to hear from the people here first.
>

I just discovered, grrr, half way through writing a plugin, that
it already existed.  I didn't even know about searching for
pytest -

and it wouldn?'t have found the one that I didn't need to write a second
in any event.

So I think this is a great idea.  I see no downside at all.

What would it be?

Laura


From nicoddemus at gmail.com  Thu Jun 18 23:59:31 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 18 Jun 2015 21:59:31 +0000
Subject: [pytest-dev] Trove Classifier for pytest plugins
In-Reply-To: <201506182149.t5ILnSne009339@fido.openend.se>
References: <nicoddemus@gmail.com>
 <CA+RQFfdKbjPjf86YO4RNvZgBoByDSrW=V51VGO-S8OfutaZcPA@mail.gmail.com>
 <201506182149.t5ILnSne009339@fido.openend.se>
Message-ID: <CA+RQFfd1gGURPCbeQ_T-h5H0OzQYojox6EedSTrXE4bL=4_vEA@mail.gmail.com>

On Thu, Jun 18, 2015 at 6:49 PM Laura Creighton <lac at openend.se> wrote:

> So I think this is a great idea.  I see no downside at all.
>
> What would it be?
>

People commented on the issue I posted supporting this idea, so I moved
this forward by requesting to add a new Trove classifier here:

https://bitbucket.org/pypa/pypi/issue/301

In case people want to vote/watch it. :)

Cheers,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/76a550f8/attachment.html>

From lac at openend.se  Fri Jun 19 00:24:06 2015
From: lac at openend.se (Laura Creighton)
Date: Fri, 19 Jun 2015 00:24:06 +0200
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: Message from Tibor Arpas <tibor.arpas@infinit.sk> of "Thu,
 18 Jun 2015 10:22:40 +0200."
 <CADi_WY1sVcNbrphWho+6LCvLUEZaW-=qtP0xCSorjv7KVkfQaA@mail.gmail.com>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
 <20150618054726.GN29732@merlinux.eu><CADi_WY1sVcNbrphWho+6LCvLUEZaW-=qtP0xCSorjv7KVkfQaA@mail.gmail.com>
Message-ID: <201506182224.t5IMO6IH011066@fido.openend.se>

In a message of Thu, 18 Jun 2015 10:22:40 +0200, Tibor Arpas writes:
>How about the new hook call system vs the old __multicall__ ? Is the plan
>still to deprecate the old way in 2.8 and remove in 2.9?
>
>Any idea how many plugin-s use it and will brake (or were already updated)?
>
>Tibor
>

Eeeee.  Can somebody point me at the new hook system.  I have code
I had no idea was in immenent threat of breakage ....

Laura

From nicoddemus at gmail.com  Fri Jun 19 00:28:03 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Thu, 18 Jun 2015 22:28:03 +0000
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <201506182224.t5IMO6IH011066@fido.openend.se>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
 <20150618054726.GN29732@merlinux.eu> <tibor.arpas@infinit.sk>
 <CADi_WY1sVcNbrphWho+6LCvLUEZaW-=qtP0xCSorjv7KVkfQaA@mail.gmail.com>
 <201506182224.t5IMO6IH011066@fido.openend.se>
Message-ID: <CA+RQFffs=8W9XXvVdO_ODO0Uyyp+Vgm=dr+973E+KfFrQXXR7A@mail.gmail.com>

On Thu, Jun 18, 2015 at 7:24 PM Laura Creighton <lac at openend.se> wrote:

> Eeeee.  Can somebody point me at the new hook system.  I have code
> I had no idea was in immenent threat of breakage ....
>

Here are a couple of examples:

https://github.com/pytest-dev/pytest-qt/blob/master/pytestqt/plugin.py#L582
https://github.com/pytest-dev/pytest-qt/blob/master/pytestqt/plugin.py#L645

Cheers,
Bruno.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150618/325939f1/attachment.html>

From lac at openend.se  Fri Jun 19 00:31:51 2015
From: lac at openend.se (Laura Creighton)
Date: Fri, 19 Jun 2015 00:31:51 +0200
Subject: [pytest-dev] plugins/pytest-qt and cx_Freeze
In-Reply-To: Message from Florian Bruhin <me@the-compiler.org>
 of "Thu, 18 Jun 2015 11:23:00 +0200." <20150618092300.GE22364@tonks>
References: <20150618092300.GE22364@tonks>
Message-ID: <201506182231.t5IMVpXv011524@fido.openend.se>

cx_Freeze is buggy.  I think you hit one.

Laura


From holger at merlinux.eu  Fri Jun 19 10:19:55 2015
From: holger at merlinux.eu (holger krekel)
Date: Fri, 19 Jun 2015 08:19:55 +0000
Subject: [pytest-dev] github push notification / pytest channel
Message-ID: <20150619081955.GR29732@merlinux.eu>

Hey Bruno, all,

i notice that we don't have push notifications to the #pylib channel.
Could someone care for that?  What's good practise for github there?

With bitbucket we have a "trumpet" bot running on hq.merlinux.eu
which could probably be configured to accept github notifications.
FWIW the trumpet daemon is currently living under ronny's account on
pytest.org.

Besides, i am wondering if we should use a new #pytest channel instead of
the #pylib one.  "pylib" is not really a primary project anymore
(py.test used to be part of pylib several years ago).  I've registered
#pytest on freenode and i can probably add others.

best,
holger



-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From me at the-compiler.org  Fri Jun 19 10:25:17 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Fri, 19 Jun 2015 10:25:17 +0200
Subject: [pytest-dev] github push notification / pytest channel
In-Reply-To: <20150619081955.GR29732@merlinux.eu>
References: <20150619081955.GR29732@merlinux.eu>
Message-ID: <20150619082517.GJ22364@tonks>

* holger krekel <holger at merlinux.eu> [2015-06-19 08:19:55 +0000]:
> i notice that we don't have push notifications to the #pylib channel.
> Could someone care for that?  What's good practise for github there?

GitHub has its own IRC service (well hidden). Go to the repo settings
-> services and add IRC there.

There's also http://n.tkte.ch/ as an alternative.

We could also set up Travis IRC notifications so we notice when things
would break for some reason:

http://docs.travis-ci.com/user/notifications/#IRC-notification

> Besides, i am wondering if we should use a new #pytest channel instead of
> the #pylib one.  "pylib" is not really a primary project anymore
> (py.test used to be part of pylib several years ago).  I've registered
> #pytest on freenode and i can probably add others.

+1 - would also make it easier to find. From time to time, people in
#python ask what the pytest channel was again.

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150619/c287ea20/attachment-0001.sig>

From me at the-compiler.org  Fri Jun 19 10:28:40 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Fri, 19 Jun 2015 10:28:40 +0200
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <20150618054726.GN29732@merlinux.eu>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
 <20150618054726.GN29732@merlinux.eu>
Message-ID: <20150619082840.GK22364@tonks>

* holger krekel <holger at merlinux.eu> [2015-06-18 05:47:26 +0000]:
> Would it make sense to consider "master" to become the new "bug fix branch"
> and have a "pytest-2.9" then where we collect new features? This way
> the default is to do bugfixes which might be easier for newcomers to
> the project.

Another possibility (GitLab flow[1]) would be to make all commits to
master, and then cherry-pick bugfixes to a, say, pytest-2.7 branch.

That's what I use for my project, and I'm quite happy with it. It
means a bit more work for maintainers (as they have to cherry-pick
things), but it takes the "burden" of deciding/knowing which branch to
use from the contributors.

[1] https://about.gitlab.com/2014/09/29/gitlab-flow/

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150619/edcf54da/attachment.sig>

From nicoddemus at gmail.com  Fri Jun 19 12:35:00 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Fri, 19 Jun 2015 10:35:00 +0000
Subject: [pytest-dev] github push notification / pytest channel
In-Reply-To: <20150619082517.GJ22364@tonks>
References: <20150619081955.GR29732@merlinux.eu> <20150619082517.GJ22364@tonks>
Message-ID: <CA+RQFfdhA5qczRveTEjRF-043-NYSxSLboBOU4rmfxsUDKUiZw@mail.gmail.com>

On Fri, Jun 19, 2015 at 5:25 AM Florian Bruhin <me at the-compiler.org> wrote:

> We could also set up Travis IRC notifications so we notice when things
> would break for some reason:
>
> http://docs.travis-ci.com/user/notifications/#IRC-notification


Just merged Florian's PR which does that:
https://github.com/pytest-dev/pytest/pull/790

Let's see how it behaves online.

>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150619/89906c89/attachment.html>

From nicoddemus at gmail.com  Fri Jun 19 12:54:13 2015
From: nicoddemus at gmail.com (Bruno Oliveira)
Date: Fri, 19 Jun 2015 10:54:13 +0000
Subject: [pytest-dev] github push notification / pytest channel
In-Reply-To: <CA+RQFfdhA5qczRveTEjRF-043-NYSxSLboBOU4rmfxsUDKUiZw@mail.gmail.com>
References: <20150619081955.GR29732@merlinux.eu> <20150619082517.GJ22364@tonks>
 <CA+RQFfdhA5qczRveTEjRF-043-NYSxSLboBOU4rmfxsUDKUiZw@mail.gmail.com>
Message-ID: <CA+RQFfcFnApsMRKLZXqjuzSNgTyDg5pOBWacs1LXSp3T70fJiQ@mail.gmail.com>

Just wanted to mention that there's Gitter (https://gitter.im) which is an
in-browser chat room targeted to open source projects. From what I see, it
supports Markdown and integrates with a bunch of services.

I don't use IRC or Gitter, just thought I would mention it in case others
wanted to try it out.

Cheers,

On Fri, Jun 19, 2015 at 7:35 AM Bruno Oliveira <nicoddemus at gmail.com> wrote:

> On Fri, Jun 19, 2015 at 5:25 AM Florian Bruhin <me at the-compiler.org>
> wrote:
>
>> We could also set up Travis IRC notifications so we notice when things
>> would break for some reason:
>>
>> http://docs.travis-ci.com/user/notifications/#IRC-notification
>
>
> Just merged Florian's PR which does that:
> https://github.com/pytest-dev/pytest/pull/790
>
> Let's see how it behaves online.
>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150619/4385c7ba/attachment.html>

From tom at viner.tv  Sat Jun 20 15:42:26 2015
From: tom at viner.tv (Tom Viner)
Date: Sat, 20 Jun 2015 14:42:26 +0100
Subject: [pytest-dev] github push notification / pytest channel
Message-ID: <CANBSuyGZTEoy5MhC1yB7UCZiP6h34pX3PL994tMDYxzbz398Eg@mail.gmail.com>

+1 for Gitter. Really reduces the barrier to entry for people to engage.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150620/041eaeec/attachment.html>

From bubenkoff at gmail.com  Sat Jun 20 17:37:57 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Sat, 20 Jun 2015 15:37:57 +0000
Subject: [pytest-dev] github push notification / pytest channel
In-Reply-To: <CANBSuyGZTEoy5MhC1yB7UCZiP6h34pX3PL994tMDYxzbz398Eg@mail.gmail.com>
References: <CANBSuyGZTEoy5MhC1yB7UCZiP6h34pX3PL994tMDYxzbz398Eg@mail.gmail.com>
Message-ID: <CACTnz+GEJ+EwhF0FYie6Ysy+RcNQo44YVO0XbYGgWR5tsA7Wxg@mail.gmail.com>

How about slack for notifications and chat?

On 15:42, Sat, Jun 20, 2015 Tom Viner <tom at viner.tv> wrote:

> +1 for Gitter. Really reduces the barrier to entry for people to engage.
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150620/e0238dde/attachment.html>

From flub at devork.be  Mon Jun 22 15:46:38 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Mon, 22 Jun 2015 06:46:38 -0700
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <20150619082840.GK22364@tonks>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
 <CAAWk_DxkkEikL8Mu6ODZyhCLct+gHkCC3k8+=uZAoL9Sme_XcQ@mail.gmail.com>
 <20150618054726.GN29732@merlinux.eu> <20150619082840.GK22364@tonks>
Message-ID: <CAAWk_Dz7wrynFdJ9ERX5dbLBMTTAPXNtcSzDr5xXOpsC8N-9OQ@mail.gmail.com>

We could also follow how zeromq does things and simply not have branches in
the main repos, using forks instead for maintenance releases.  It does not
help with the fact that we'd be back in charge of cherry picking fixes for
bugfix releases though.

But I honestly don't mind too much, happy to leave the branch workflow to
people more experienced with git then me.

Floris
On 19 Jun 2015 09:28, "Florian Bruhin" <me at the-compiler.org> wrote:

> * holger krekel <holger at merlinux.eu> [2015-06-18 05:47:26 +0000]:
> > Would it make sense to consider "master" to become the new "bug fix
> branch"
> > and have a "pytest-2.9" then where we collect new features? This way
> > the default is to do bugfixes which might be easier for newcomers to
> > the project.
>
> Another possibility (GitLab flow[1]) would be to make all commits to
> master, and then cherry-pick bugfixes to a, say, pytest-2.7 branch.
>
> That's what I use for my project, and I'm quite happy with it. It
> means a bit more work for maintainers (as they have to cherry-pick
> things), but it takes the "burden" of deciding/knowing which branch to
> use from the contributors.
>
> [1] https://about.gitlab.com/2014/09/29/gitlab-flow/
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
>
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150622/f9fe171a/attachment.html>

From holger at merlinux.eu  Mon Jun 22 23:52:17 2015
From: holger at merlinux.eu (holger krekel)
Date: Mon, 22 Jun 2015 21:52:17 +0000
Subject: [pytest-dev] 2.8.0 release plan? 2.7.2?
In-Reply-To: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
References: <CA+RQFfeqwBVO8719kwoLD_GgybZFG50DU3MvV1HdqQRXayP02Q@mail.gmail.com>
Message-ID: <20150622215217.GK29732@merlinux.eu>

Hi Bruno, all,

On Wed, Jun 17, 2015 at 12:31 +0000, Bruno Oliveira wrote:
> Hi, Do we have an idea of when we would like have a 2.8.0 release, or at
> least what we would like to see in the new version? And about 2.7.2? I see
> there are a couple of fixes that are in master but should be backported to
> pytest-2.7 branch before making a 2.7.2 release. Cheers,

I am now thinking we should just do a quick pytest-2.7.2.

I just fixed an error in the traceback generation which many people
have encountered and prepared a py-1.4.29 release.  I also prepared
a pytest-2.7.2 with the fixes that are currently in.  You can install
both with:

    pip install -U -i https://devpi.net/hpk/dev pytest

Tests are passing:

    https://devpi.net/hpk/dev/py/1.4.29

    https://devpi.net/hpk/dev/pytest/2.7.2
    
Anything speaking against just pushing those releases to pypi now?

FWIW I think python3.5 compat (currently 3.5.alpha2 is out) should
be tackled for pytest-2.8.

holger

From flub at devork.be  Tue Jun 23 14:51:42 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Tue, 23 Jun 2015 13:51:42 +0100
Subject: [pytest-dev] Pytest migration to GitHub
In-Reply-To: <CA+RQFffzMmLFCNdkY6YohV6m=e2EMSoeGpmMBe=__imt-MUbmQ@mail.gmail.com>
References: <CA+RQFfc57Tuwqvq88zQtL92nrka=zkx1=97JBx90kZjDn86xFA@mail.gmail.com>
 <CA+RQFfcrUd54fF17KqK0=VNRYVTKbhncKoW2aoYxM1C=OfuXkw@mail.gmail.com>
 <CA+RQFfezfQqdJH6AdmtUkoGVg54aPYX9XeY3_fvESghCw=GyWQ@mail.gmail.com>
 <CA+RQFffzMmLFCNdkY6YohV6m=e2EMSoeGpmMBe=__imt-MUbmQ@mail.gmail.com>
Message-ID: <CAAWk_Dyy3gNQgHc01Fk1Buy1Hg=1UU1bXoAv-Qz_-nU1-+eZLQ@mail.gmail.com>

Hi,

I noticed Bruno created an issue to say that they should now be
reported on github instead.  I was wondering if we should maybe simply
disable issue tracking completely?  I've just experimented with this
in a test repo and it seems you can no longer access issues when you
disable it, however if you then re-enable the issues they are all
still there.

Also related, do we leave the code there?  Why not fully delete the
repo and redirect?

On 16 June 2015 at 00:25, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> Hi everyone,
>
> Here's the list of the tasks done for the move to GitHub:
> https://github.com/pytest-dev/pytest/issues/769
>
> If you see any obsolete link or documentation piece that needs to be
> updated, please don't refrain to point it out or creating a PR.
>
> Cheers,
>
> On Mon, Jun 15, 2015 at 6:07 PM Bruno Oliveira <nicoddemus at gmail.com> wrote:
>>
>> Hi everyone,
>>
>> We have our new repository ready:
>>
>> https://github.com/pytest-dev/pytest
>>
>> Issues and commits may now continue on the new repository.
>>
>> The few remaining tasks to fully complete the move are listed here:
>> https://github.com/pytest-dev/pytest/issues/769
>>
>> Thanks everyone for your patience!
>>
>> Cheers,
>>
>>
>>
>> On Mon, Jun 15, 2015 at 5:31 PM Bruno Oliveira <nicoddemus at gmail.com>
>> wrote:
>>>
>>> Hi everyone,
>>>
>>> The migration process has started now. I will send an email once the
>>> first steps are complete.
>>>
>>> Cheers,
>>>
>>> On Mon, Jun 15, 2015 at 9:59 AM Bruno Oliveira <nicoddemus at gmail.com>
>>> wrote:
>>>>
>>>> Hi everyone,
>>>>
>>>> Today we will migrate the pytest repository from bitbucket (Hg) to
>>>> GitHub (Git), sometime after 6pm (GMT -03:00).
>>>>
>>>> Until then feel free to commit and handle issues as usual, but when the
>>>> process starts please refrain to make any changes to issues or new commits
>>>> until the process finishes. I will send an email to this list when the
>>>> process officially starts, and it is expected to take no longer than 1 hour
>>>> before normal work, now on the new repository, can resume.
>>>>
>>>> For those interested, here are the steps for the migration:
>>>>
>>>> 1. Create github.com/pytest-dev/pytest and move current issues;
>>>> 2. Convert pytest Hg repository to Git and upload to GitHub;
>>>> 3. Add to bitbucket's README a notice about the move to GitHub, and that
>>>> new issues/PRs should be posted there;
>>>>
>>>> (At this point, work on the new repository can resume)
>>>>
>>>> 4. Ask submitters to re-create PRs at the new repository;
>>>> 5. Update all links in the documentation and PyPI;
>>>> 6. Update "how to contribute" docs;
>>>> 7. Upload new docs to pytest.org;
>>>> 8. Send an email to all relevant mailing lists about the migration;
>>>>
>>>> Suggestions and comments are welcome!
>>>>
>>>> Cheers,
>
>
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>

From flub at devork.be  Tue Jun 23 14:54:09 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Tue, 23 Jun 2015 13:54:09 +0100
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
References: <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
Message-ID: <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>

On 16 June 2015 at 03:20, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> after such a success with pytest move, should we start the same process for
> other repos under pytest-dev on bb?

Not sure if anyone saw my tweet, but I noticed (sadly too late!) that
the git repo seems to be ~50Mb to clone instead of ~20Mb for the hg
one.  Is there anything that can still be done about this?  And is
there a way to avoid this for any other repo we might migrate?

From bubenkoff at gmail.com  Tue Jun 23 14:57:21 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Tue, 23 Jun 2015 12:57:21 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
References: <CACTnz+G8tD4yZuuuTTzYiB3WPq514uT+UsCaLu-NsR22ZBC5NQ@mail.gmail.com>
 <CA+RQFffDgXWmrZfuMp3ySDoOgmHHDvhH4=GhMNL7hS0zTi8S9A@mail.gmail.com>
 <CACTnz+G5TVLxEpMQPZmx-oBztQPh1-fEXUjNp_uCfT66qjPyQA@mail.gmail.com>
 <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
 <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
Message-ID: <CACTnz+Fn4JGhmjLmT4s0tv6Ubp2Ri_pLKosoNUdGoMsmkK101A@mail.gmail.com>

but it's not neccessary (or most likely not) because of the problematic
migration
the way git and hg store changes is still different, git always store full
file contents on any change to it from revision to revision, while hg i
believe uses some periodic snapshotting mechanism, so it does full copy
after several incremental diffs

On Tue, Jun 23, 2015 at 2:54 PM Floris Bruynooghe <flub at devork.be> wrote:

> On 16 June 2015 at 03:20, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> > after such a success with pytest move, should we start the same process
> for
> > other repos under pytest-dev on bb?
>
> Not sure if anyone saw my tweet, but I noticed (sadly too late!) that
> the git repo seems to be ~50Mb to clone instead of ~20Mb for the hg
> one.  Is there anything that can still be done about this?  And is
> there a way to avoid this for any other repo we might migrate?
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150623/85f4591c/attachment.html>

From bubenkoff at gmail.com  Tue Jun 23 14:58:55 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Tue, 23 Jun 2015 12:58:55 +0000
Subject: [pytest-dev] Pytest migration to GitHub
In-Reply-To: <CAAWk_Dyy3gNQgHc01Fk1Buy1Hg=1UU1bXoAv-Qz_-nU1-+eZLQ@mail.gmail.com>
References: <CA+RQFfc57Tuwqvq88zQtL92nrka=zkx1=97JBx90kZjDn86xFA@mail.gmail.com>
 <CA+RQFfcrUd54fF17KqK0=VNRYVTKbhncKoW2aoYxM1C=OfuXkw@mail.gmail.com>
 <CA+RQFfezfQqdJH6AdmtUkoGVg54aPYX9XeY3_fvESghCw=GyWQ@mail.gmail.com>
 <CA+RQFffzMmLFCNdkY6YohV6m=e2EMSoeGpmMBe=__imt-MUbmQ@mail.gmail.com>
 <CAAWk_Dyy3gNQgHc01Fk1Buy1Hg=1UU1bXoAv-Qz_-nU1-+eZLQ@mail.gmail.com>
Message-ID: <CACTnz+Gd=_6e6PWi_57eTCHVouFZoeTtxhfe+T0HXN29xC=y=g@mail.gmail.com>

cleaning up the repo and leaving the readme with redirect seems like a good
plan - just like we need for pytest on old location under Holger's account

On Tue, Jun 23, 2015 at 2:51 PM Floris Bruynooghe <flub at devork.be> wrote:

> Hi,
>
> I noticed Bruno created an issue to say that they should now be
> reported on github instead.  I was wondering if we should maybe simply
> disable issue tracking completely?  I've just experimented with this
> in a test repo and it seems you can no longer access issues when you
> disable it, however if you then re-enable the issues they are all
> still there.
>
> Also related, do we leave the code there?  Why not fully delete the
> repo and redirect?
>
> On 16 June 2015 at 00:25, Bruno Oliveira <nicoddemus at gmail.com> wrote:
> > Hi everyone,
> >
> > Here's the list of the tasks done for the move to GitHub:
> > https://github.com/pytest-dev/pytest/issues/769
> >
> > If you see any obsolete link or documentation piece that needs to be
> > updated, please don't refrain to point it out or creating a PR.
> >
> > Cheers,
> >
> > On Mon, Jun 15, 2015 at 6:07 PM Bruno Oliveira <nicoddemus at gmail.com>
> wrote:
> >>
> >> Hi everyone,
> >>
> >> We have our new repository ready:
> >>
> >> https://github.com/pytest-dev/pytest
> >>
> >> Issues and commits may now continue on the new repository.
> >>
> >> The few remaining tasks to fully complete the move are listed here:
> >> https://github.com/pytest-dev/pytest/issues/769
> >>
> >> Thanks everyone for your patience!
> >>
> >> Cheers,
> >>
> >>
> >>
> >> On Mon, Jun 15, 2015 at 5:31 PM Bruno Oliveira <nicoddemus at gmail.com>
> >> wrote:
> >>>
> >>> Hi everyone,
> >>>
> >>> The migration process has started now. I will send an email once the
> >>> first steps are complete.
> >>>
> >>> Cheers,
> >>>
> >>> On Mon, Jun 15, 2015 at 9:59 AM Bruno Oliveira <nicoddemus at gmail.com>
> >>> wrote:
> >>>>
> >>>> Hi everyone,
> >>>>
> >>>> Today we will migrate the pytest repository from bitbucket (Hg) to
> >>>> GitHub (Git), sometime after 6pm (GMT -03:00).
> >>>>
> >>>> Until then feel free to commit and handle issues as usual, but when
> the
> >>>> process starts please refrain to make any changes to issues or new
> commits
> >>>> until the process finishes. I will send an email to this list when the
> >>>> process officially starts, and it is expected to take no longer than
> 1 hour
> >>>> before normal work, now on the new repository, can resume.
> >>>>
> >>>> For those interested, here are the steps for the migration:
> >>>>
> >>>> 1. Create github.com/pytest-dev/pytest and move current issues;
> >>>> 2. Convert pytest Hg repository to Git and upload to GitHub;
> >>>> 3. Add to bitbucket's README a notice about the move to GitHub, and
> that
> >>>> new issues/PRs should be posted there;
> >>>>
> >>>> (At this point, work on the new repository can resume)
> >>>>
> >>>> 4. Ask submitters to re-create PRs at the new repository;
> >>>> 5. Update all links in the documentation and PyPI;
> >>>> 6. Update "how to contribute" docs;
> >>>> 7. Upload new docs to pytest.org;
> >>>> 8. Send an email to all relevant mailing lists about the migration;
> >>>>
> >>>> Suggestions and comments are welcome!
> >>>>
> >>>> Cheers,
> >
> >
> > _______________________________________________
> > pytest-dev mailing list
> > pytest-dev at python.org
> > https://mail.python.org/mailman/listinfo/pytest-dev
> >
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150623/a7475326/attachment-0001.html>

From me at the-compiler.org  Tue Jun 23 15:05:51 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Tue, 23 Jun 2015 15:05:51 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
References: <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
 <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
Message-ID: <20150623130551.GW22364@tonks>

* Floris Bruynooghe <flub at devork.be> [2015-06-23 13:54:09 +0100]:
> On 16 June 2015 at 03:20, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> > after such a success with pytest move, should we start the same process for
> > other repos under pytest-dev on bb?
> 
> Not sure if anyone saw my tweet, but I noticed (sadly too late!) that
> the git repo seems to be ~50Mb to clone instead of ~20Mb for the hg
> one.  Is there anything that can still be done about this?  And is
> there a way to avoid this for any other repo we might migrate?

It seems git calculated (because of the import?) very inefficient
deltas beetween the commits.

After doing a  git gc --aggressive  the repo shrinks from 55 MB to
9.2 MB for me.

So for future migrations it might make sense to do that in the
converted git repo before pushing.

I contacted GitHub support to ask if they can do that on the repo.

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150623/c76e497d/attachment.sig>

From flub at devork.be  Tue Jun 23 15:25:26 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Tue, 23 Jun 2015 14:25:26 +0100
Subject: [pytest-dev] github push notification / pytest channel
In-Reply-To: <CACTnz+GEJ+EwhF0FYie6Ysy+RcNQo44YVO0XbYGgWR5tsA7Wxg@mail.gmail.com>
References: <CANBSuyGZTEoy5MhC1yB7UCZiP6h34pX3PL994tMDYxzbz398Eg@mail.gmail.com>
 <CACTnz+GEJ+EwhF0FYie6Ysy+RcNQo44YVO0XbYGgWR5tsA7Wxg@mail.gmail.com>
Message-ID: <CAAWk_DxHm-0G4uagZmcAC_n-OXrLm4wHb3oOgPohoxgs5Dx4QQ@mail.gmail.com>

On 20 June 2015 at 16:37, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> How about slack for notifications and chat?
>
>
> On 15:42, Sat, Jun 20, 2015 Tom Viner <tom at viner.tv> wrote:
>>
>> +1 for Gitter. Really reduces the barrier to entry for people to engage.

Could gitter bridge bi-directionally with IRC on freenode?   There is
also webchat.freenode.net btw, maybe we should add that to the docs as
well?

As for slack, not sure if that's very suitable as it's not really
designed as a public service so you'd have to sign up people.  I guess
the feature people like is that it preserves history.  I'm also not a
huge fan of keeping adding closed products into the workflow/community
of a free software projects to be honest.  But I guess IRC must be
pretty alien to many people.


Floris

From holger at merlinux.eu  Tue Jun 23 15:43:47 2015
From: holger at merlinux.eu (holger krekel)
Date: Tue, 23 Jun 2015 13:43:47 +0000
Subject: [pytest-dev] github push notification / pytest channel
In-Reply-To: <CAAWk_DxHm-0G4uagZmcAC_n-OXrLm4wHb3oOgPohoxgs5Dx4QQ@mail.gmail.com>
References: <CANBSuyGZTEoy5MhC1yB7UCZiP6h34pX3PL994tMDYxzbz398Eg@mail.gmail.com>
 <CACTnz+GEJ+EwhF0FYie6Ysy+RcNQo44YVO0XbYGgWR5tsA7Wxg@mail.gmail.com>
 <CAAWk_DxHm-0G4uagZmcAC_n-OXrLm4wHb3oOgPohoxgs5Dx4QQ@mail.gmail.com>
Message-ID: <20150623134347.GT29732@merlinux.eu>

On Tue, Jun 23, 2015 at 14:25 +0100, Floris Bruynooghe wrote:
> On 20 June 2015 at 16:37, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> > How about slack for notifications and chat?
> >
> >
> > On 15:42, Sat, Jun 20, 2015 Tom Viner <tom at viner.tv> wrote:
> >>
> >> +1 for Gitter. Really reduces the barrier to entry for people to engage.
> 
> Could gitter bridge bi-directionally with IRC on freenode?   There is
> also webchat.freenode.net btw, maybe we should add that to the docs as
> well?
> 
> As for slack, not sure if that's very suitable as it's not really
> designed as a public service so you'd have to sign up people.  I guess
> the feature people like is that it preserves history.  I'm also not a
> huge fan of keeping adding closed products into the workflow/community
> of a free software projects to be honest.  But I guess IRC must be
> pretty alien to many people.

FWIW i am unlikely to participate in another communication channel
and generally agree on minimizing the usage of proprietary services.
What's bad about IRC and freenode which is kind of a standard for
open source developments?

holger

From bubenkoff at gmail.com  Tue Jun 23 15:46:19 2015
From: bubenkoff at gmail.com (Anatoly Bubenkov)
Date: Tue, 23 Jun 2015 13:46:19 +0000
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150623130551.GW22364@tonks>
References: <CA+RQFfeBau6oa=FT38hKW0muPxo3WiozLuVQETP2OThMXQwi9w@mail.gmail.com>
 <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
 <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
 <20150623130551.GW22364@tonks>
Message-ID: <CACTnz+G+CnMBvbFX7DobiaJZND5u6jgp=wN=d0LzeVb+HrexNw@mail.gmail.com>

nice catch!

On Tue, Jun 23, 2015 at 3:06 PM Florian Bruhin <me at the-compiler.org> wrote:

> * Floris Bruynooghe <flub at devork.be> [2015-06-23 13:54:09 +0100]:
> > On 16 June 2015 at 03:20, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> > > after such a success with pytest move, should we start the same
> process for
> > > other repos under pytest-dev on bb?
> >
> > Not sure if anyone saw my tweet, but I noticed (sadly too late!) that
> > the git repo seems to be ~50Mb to clone instead of ~20Mb for the hg
> > one.  Is there anything that can still be done about this?  And is
> > there a way to avoid this for any other repo we might migrate?
>
> It seems git calculated (because of the import?) very inefficient
> deltas beetween the commits.
>
> After doing a  git gc --aggressive  the repo shrinks from 55 MB to
> 9.2 MB for me.
>
> So for future migrations it might make sense to do that in the
> converted git repo before pushing.
>
> I contacted GitHub support to ask if they can do that on the repo.
>
> Florian
>
> --
> http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
>    GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
>          I love long mails! | http://email.is-not-s.ms/
> _______________________________________________
> pytest-dev mailing list
> pytest-dev at python.org
> https://mail.python.org/mailman/listinfo/pytest-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150623/c759419a/attachment.html>

From me at the-compiler.org  Tue Jun 23 15:50:26 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Tue, 23 Jun 2015 15:50:26 +0200
Subject: [pytest-dev] github push notification / pytest channel
In-Reply-To: <20150623134347.GT29732@merlinux.eu>
References: <CANBSuyGZTEoy5MhC1yB7UCZiP6h34pX3PL994tMDYxzbz398Eg@mail.gmail.com>
 <CACTnz+GEJ+EwhF0FYie6Ysy+RcNQo44YVO0XbYGgWR5tsA7Wxg@mail.gmail.com>
 <CAAWk_DxHm-0G4uagZmcAC_n-OXrLm4wHb3oOgPohoxgs5Dx4QQ@mail.gmail.com>
 <20150623134347.GT29732@merlinux.eu>
Message-ID: <20150623135026.GX22364@tonks>

* holger krekel <holger at merlinux.eu> [2015-06-23 13:43:47 +0000]:
> On Tue, Jun 23, 2015 at 14:25 +0100, Floris Bruynooghe wrote:
> > On 20 June 2015 at 16:37, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> > > How about slack for notifications and chat?
> > >
> > >
> > > On 15:42, Sat, Jun 20, 2015 Tom Viner <tom at viner.tv> wrote:
> > >>
> > >> +1 for Gitter. Really reduces the barrier to entry for people to engage.
> > 
> > Could gitter bridge bi-directionally with IRC on freenode?   There is
> > also webchat.freenode.net btw, maybe we should add that to the docs as
> > well?
> > 
> > As for slack, not sure if that's very suitable as it's not really
> > designed as a public service so you'd have to sign up people.  I guess
> > the feature people like is that it preserves history.  I'm also not a
> > huge fan of keeping adding closed products into the workflow/community
> > of a free software projects to be honest.  But I guess IRC must be
> > pretty alien to many people.
> 
> FWIW i am unlikely to participate in another communication channel
> and generally agree on minimizing the usage of proprietary services.
> What's bad about IRC and freenode which is kind of a standard for
> open source developments?

I agree - I won't use Gitter (unless their IRC bridge which is "very
much beta" actually works). Looking at Gitter chats for some random
Python projects (at https://gitter.im/explore/tags/python ) there are
a lot of them where it's very quiet.

I really don't see how it would be more difficult to use the Freenode
webchat compared to Gitter.

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150623/81dc0689/attachment-0001.sig>

From holger at merlinux.eu  Tue Jun 23 16:40:05 2015
From: holger at merlinux.eu (holger krekel)
Date: Tue, 23 Jun 2015 14:40:05 +0000
Subject: [pytest-dev] pytest-2.7.2: bug fixes
Message-ID: <20150623144005.GU29732@merlinux.eu>

pytest-2.7.2: bug fixes
=======================

pytest is a mature Python testing tool with more than a 1100 tests
against itself, passing on many different interpreters and platforms.
This release is supposed to be drop-in compatible to 2.7.1.

See below for the changes and see docs at:

    http://pytest.org

As usual, you can upgrade from pypi via::

    pip install -U pytest

Thanks to all who contributed to this release, among them:

    Bruno Oliveira
    Floris Bruynooghe
    Punyashloka Biswal
    Aron Curzon
    Benjamin Peterson
    Thomas De Schampheleire
    Edison Gustavo Muenz
    Holger Krekel

Happy testing,
The py.test Development Team


2.7.2 (compared to 2.7.1)
-----------------------------

- fix issue767: pytest.raises value attribute does not contain the exception
  instance on Python 2.6. Thanks Eric Siegerman for providing the test
  case and Bruno Oliveira for PR.

- Automatically create directory for junitxml and results log.
  Thanks Aron Curzon.

- fix issue713: JUnit XML reports for doctest failures.
  Thanks Punyashloka Biswal.

- fix issue735: assertion failures on debug versions of Python 3.4+
  Thanks Benjamin Peterson.

- fix issue114: skipif marker reports to internal skipping plugin;
  Thanks Floris Bruynooghe for reporting and Bruno Oliveira for the PR.

- fix issue748: unittest.SkipTest reports to internal pytest unittest plugin.
  Thanks Thomas De Schampheleire for reporting and Bruno Oliveira for the PR.

- fix issue718: failed to create representation of sets containing unsortable
  elements in python 2. Thanks Edison Gustavo Muenz

- fix issue756, fix issue752 (and similar issues): depend on py-1.4.29
  which has a refined algorithm for traceback generation.

-- 
about me:    http://holgerkrekel.net/about-me/
contracting: http://merlinux.eu

From me at the-compiler.org  Wed Jun 24 10:01:45 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Wed, 24 Jun 2015 10:01:45 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150623130551.GW22364@tonks>
References: <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
 <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
 <20150623130551.GW22364@tonks>
Message-ID: <20150624080145.GD22364@tonks>

* Florian Bruhin <me at the-compiler.org> [2015-06-23 15:05:51 +0200]:
> * Floris Bruynooghe <flub at devork.be> [2015-06-23 13:54:09 +0100]:
> > On 16 June 2015 at 03:20, Anatoly Bubenkov <bubenkoff at gmail.com> wrote:
> > > after such a success with pytest move, should we start the same process for
> > > other repos under pytest-dev on bb?
> > 
> > Not sure if anyone saw my tweet, but I noticed (sadly too late!) that
> > the git repo seems to be ~50Mb to clone instead of ~20Mb for the hg
> > one.  Is there anything that can still be done about this?  And is
> > there a way to avoid this for any other repo we might migrate?
> 
> It seems git calculated (because of the import?) very inefficient
> deltas beetween the commits.
> 
> After doing a  git gc --aggressive  the repo shrinks from 55 MB to
> 9.2 MB for me.
> 
> So for future migrations it might make sense to do that in the
> converted git repo before pushing.
> 
> I contacted GitHub support to ask if they can do that on the repo.

They now did run it, and the download is around 5.7 MB now :)

Definitely much better - thanks for bringing it up!

Florian

-- 
http://www.the-compiler.org | me at the-compiler.org (Mail/XMPP)
   GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc
         I love long mails! | http://email.is-not-s.ms/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20150624/05c9df1a/attachment.sig>

From flub at devork.be  Thu Jun 25 10:53:57 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Thu, 25 Jun 2015 09:53:57 +0100
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <20150624080145.GD22364@tonks>
References: <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
 <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
 <20150623130551.GW22364@tonks> <20150624080145.GD22364@tonks>
Message-ID: <CAAWk_DwRvgP_BYRJrH0cPkcrPObj7NdrsFveWn2XUn-eYNP9oQ@mail.gmail.com>

On 24 June 2015 at 09:01, Florian Bruhin <me at the-compiler.org> wrote:
> * Florian Bruhin <me at the-compiler.org> [2015-06-23 15:05:51 +0200]:
>> After doing a  git gc --aggressive  the repo shrinks from 55 MB to
>> 9.2 MB for me.
>>
>> So for future migrations it might make sense to do that in the
>> converted git repo before pushing.
>>
>> I contacted GitHub support to ask if they can do that on the repo.
>
> They now did run it, and the download is around 5.7 MB now :)

Nice!  Thanks for looking into this.

Floris

From florian.schulze at gmx.net  Thu Jun 25 11:13:42 2015
From: florian.schulze at gmx.net (Florian Schulze)
Date: Thu, 25 Jun 2015 11:13:42 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <CAAWk_DwRvgP_BYRJrH0cPkcrPObj7NdrsFveWn2XUn-eYNP9oQ@mail.gmail.com>
References: <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
 <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
 <20150623130551.GW22364@tonks> <20150624080145.GD22364@tonks>
 <CAAWk_DwRvgP_BYRJrH0cPkcrPObj7NdrsFveWn2XUn-eYNP9oQ@mail.gmail.com>
Message-ID: <C6A08EED-34CB-4337-BAC3-CEAD6E5353A4@gmx.net>

On 25 Jun 2015, at 10:53, Floris Bruynooghe wrote:
> On 24 June 2015 at 09:01, Florian Bruhin <me at the-compiler.org> wrote:
>> * Florian Bruhin <me at the-compiler.org> [2015-06-23 15:05:51 +0200]:
>>> After doing a  git gc --aggressive  the repo shrinks from 55 MB to
>>> 9.2 MB for me.
>>>
>>> So for future migrations it might make sense to do that in the
>>> converted git repo before pushing.
>>>
>>> I contacted GitHub support to ask if they can do that on the repo.
>>
>> They now did run it, and the download is around 5.7 MB now :)
>
> Nice!  Thanks for looking into this.

I did some experiments on some of my repositories. I had a 800MB 
repository that shrank to 220MB. You have to be careful though. By 
default the repacking has no memory limit and uses as many threads as 
you have cpu cores. My poor server with 16GB Ram and 8 cores used 20GB 
Swap extensively :) On my laptop the process died at first. You can set 
"git config pack.threads 1" and "git config pack.windowMemory 1g" in 
your repository before you run git gc to avoid that issue. The whole 
process can take a pretty long time depending on your repository.

In general it seems like this is only needed as a final step after a 
migration from another version control system or if you have a lot of 
constant changes in the same files in each commit (I backup database 
dumps with git to keep some history).

Regards,
Florian Schulze

From encukou at gmail.com  Thu Jun 25 11:23:27 2015
From: encukou at gmail.com (Petr Viktorin)
Date: Thu, 25 Jun 2015 11:23:27 +0200
Subject: [pytest-dev] moving to git / issues?
In-Reply-To: <C6A08EED-34CB-4337-BAC3-CEAD6E5353A4@gmx.net>
References: <CAAWk_DwTeaCJh_LHk4B6Zf6JuriM-CY5LNGAEDeVxatWe7EBtQ@mail.gmail.com>
 <CA+RQFfd--ABeX_HHHPArnoQBWsu4Mz_9OGO92UkxNHgt4JFGkQ@mail.gmail.com>
 <20150612112614.GF29732@merlinux.eu>
 <CA+RQFfe-YzpkG6AxBaKxR8=kw7S2SmXydB81CO+BGgHJYwctZA@mail.gmail.com>
 <20150613070505.GK29732@merlinux.eu>
 <CA+RQFfdN_SgCU57QSALW2Gfarog4EG7fCVzL7WPuUs7SA3WQbg@mail.gmail.com>
 <20150615093513.GR29732@merlinux.eu>
 <CACTnz+F5Vw3XqTFO3aAD=Ns4QprfFFti3zKmBjvXu_VayM0czQ@mail.gmail.com>
 <CAAWk_DwZnyd+XR66rTQZ3=yJm=gcro7-nh3V-r-0Xxg1pisHyg@mail.gmail.com>
 <20150623130551.GW22364@tonks> <20150624080145.GD22364@tonks>
 <CAAWk_DwRvgP_BYRJrH0cPkcrPObj7NdrsFveWn2XUn-eYNP9oQ@mail.gmail.com>
 <C6A08EED-34CB-4337-BAC3-CEAD6E5353A4@gmx.net>
Message-ID: <CA+=+wqAOGh-+_zZpqT9AC0FWoWboSWWM3Ow0otN4JYEk+=CwxQ@mail.gmail.com>

On Thu, Jun 25, 2015 at 11:13 AM, Florian Schulze
<florian.schulze at gmx.net> wrote:
> On 25 Jun 2015, at 10:53, Floris Bruynooghe wrote:
>>
>> On 24 June 2015 at 09:01, Florian Bruhin <me at the-compiler.org> wrote:
>>>
>>> * Florian Bruhin <me at the-compiler.org> [2015-06-23 15:05:51 +0200]:
>>>>
>>>> After doing a  git gc --aggressive  the repo shrinks from 55 MB to
>>>> 9.2 MB for me.
>>>>
>>>> So for future migrations it might make sense to do that in the
>>>> converted git repo before pushing.
>>>>
>>>> I contacted GitHub support to ask if they can do that on the repo.
>>>
>>>
>>> They now did run it, and the download is around 5.7 MB now :)
>>
>>
>> Nice!  Thanks for looking into this.
>
>
> I did some experiments on some of my repositories. I had a 800MB repository
> that shrank to 220MB. You have to be careful though. By default the
> repacking has no memory limit and uses as many threads as you have cpu
> cores. My poor server with 16GB Ram and 8 cores used 20GB Swap extensively
> :) On my laptop the process died at first. You can set "git config
> pack.threads 1" and "git config pack.windowMemory 1g" in your repository
> before you run git gc to avoid that issue. The whole process can take a
> pretty long time depending on your repository.
>
> In general it seems like this is only needed as a final step after a
> migration from another version control system

Exactly.

> or if you have a lot of
> constant changes in the same files in each commit (I backup database dumps
> with git to keep some history).

Here's a mail with lots of details, including a better command to use
than git gc --aggressive:
https://gcc.gnu.org/ml/gcc/2007-12/msg00165.html