From me at the-compiler.org  Mon Nov  9 00:24:25 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Mon, 9 Nov 2015 06:24:25 +0100
Subject: [pytest-dev] Moving catchlog to pytest-dev (was: Re:
 pytest-capturelog status/merging to core?)
In-Reply-To: <20151020042042.GR29343@tonks>
References: <20150925094316.GB26059@merlinux.eu>
 <CA+RQFffJ-zPsoMW900SULWJGKWqCJGWB1cavL283pPV-EVLL5g@mail.gmail.com>
 <20150925114046.GD26059@merlinux.eu> <20150925120312.GB1393@tonks>
 <CA+RQFfdP0CXUyy_1mTtLcJUHVQpHbM1LWjDhxqYOrhcXBvWf0w@mail.gmail.com>
 <43A394C1-E54E-4F0A-9B9F-75F2772FD96F@ronnypfannschmidt.de>
 <20150925175526.GE26059@merlinux.eu>
 <A7BB3652-2404-49DD-963D-8FCDAC294213@ronnypfannschmidt.de>
 <56255DE4.2050104@mailbox.org> <20151020042042.GR29343@tonks>
Message-ID: <20151109052424.GU5929@tonks>

Hey,

* Florian Bruhin <me at the-compiler.org> [2015-10-20 06:20:42 +0200]:
> * Arthur Skowronek <eisensheng at mailbox.org> [2015-10-19 23:17:24 +0200]:
> > It would be nice if the plugin could stay separated from the core package.
> > At least I am a huge fan of the modular plugin system as it can be found in
> > pytest.  I find such a modular design more preferable compared to a
> > monolithic project that comes with everything attached. Such a large project
> > makes it harder for users to track what was changed between versions and
> > keep track of changes important to them.  They don't need to care at all if
> > the API was changed but they don't use the plugin.
> > 
> > Also different contributors stated that they still would like to improve the
> > code base and add more functionality.  It would be nice to give them their
> > freedom to break the API if necessary to create something even more usable
> > with a sleeker API.  I'm looking forward to it.
> 
> I used to have a different stance on this - but after reconsidering, I
> agree it'd be best to have a "recommended" plugin (i.e. catchlog, as
> it's maintained and under pytest-dev) rather than merge things into
> core.

Backwards compatibility is restored and eisensheng released 1.2.0 - so
I think everything is ready to move pytest-catchlog to pytest-dev now.

Of course +1 from my side - any objections and/or a second +1 please?
;)

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: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20151109/d4c75ab8/attachment.sig>

From holger at merlinux.eu  Mon Nov  9 03:52:15 2015
From: holger at merlinux.eu (holger krekel)
Date: Mon, 9 Nov 2015 08:52:15 +0000
Subject: [pytest-dev] Moving catchlog to pytest-dev (was: Re:
 pytest-capturelog status/merging to core?)
In-Reply-To: <20151109052424.GU5929@tonks>
References: <CA+RQFffJ-zPsoMW900SULWJGKWqCJGWB1cavL283pPV-EVLL5g@mail.gmail.com>
 <20150925114046.GD26059@merlinux.eu> <20150925120312.GB1393@tonks>
 <CA+RQFfdP0CXUyy_1mTtLcJUHVQpHbM1LWjDhxqYOrhcXBvWf0w@mail.gmail.com>
 <43A394C1-E54E-4F0A-9B9F-75F2772FD96F@ronnypfannschmidt.de>
 <20150925175526.GE26059@merlinux.eu>
 <A7BB3652-2404-49DD-963D-8FCDAC294213@ronnypfannschmidt.de>
 <56255DE4.2050104@mailbox.org> <20151020042042.GR29343@tonks>
 <20151109052424.GU5929@tonks>
Message-ID: <20151109085215.GS16107@merlinux.eu>

On Mon, Nov 09, 2015 at 06:24 +0100, Florian Bruhin wrote:
> Hey,
> 
> * Florian Bruhin <me at the-compiler.org> [2015-10-20 06:20:42 +0200]:
> > * Arthur Skowronek <eisensheng at mailbox.org> [2015-10-19 23:17:24 +0200]:
> > > It would be nice if the plugin could stay separated from the core package.
> > > At least I am a huge fan of the modular plugin system as it can be found in
> > > pytest.  I find such a modular design more preferable compared to a
> > > monolithic project that comes with everything attached. Such a large project
> > > makes it harder for users to track what was changed between versions and
> > > keep track of changes important to them.  They don't need to care at all if
> > > the API was changed but they don't use the plugin.
> > > 
> > > Also different contributors stated that they still would like to improve the
> > > code base and add more functionality.  It would be nice to give them their
> > > freedom to break the API if necessary to create something even more usable
> > > with a sleeker API.  I'm looking forward to it.
> > 
> > I used to have a different stance on this - but after reconsidering, I
> > agree it'd be best to have a "recommended" plugin (i.e. catchlog, as
> > it's maintained and under pytest-dev) rather than merge things into
> > core.
> 
> Backwards compatibility is restored and eisensheng released 1.2.0 - so
> I think everything is ready to move pytest-catchlog to pytest-dev now.
> 
> Of course +1 from my side - any objections and/or a second +1 please?
> ;)

I noticed that there now also is a pytest-logging release from Pedro Algarvio:

    https://pypi.python.org/pypi/pytest-logging

which adds to the confusion around logging and pytest, unfortunately.

I haven't used pytest-catchlog myself yet but intend to do so.
It'd be great if eisenheng or you could get in contact with Pedro
and see how we can all arrive in a situation which is less confusing
to users, somewhat independently from the question of merging catchlog
into pytest-dev.

holger



> 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 me at the-compiler.org  Mon Nov  9 04:22:52 2015
From: me at the-compiler.org (Florian Bruhin)
Date: Mon, 9 Nov 2015 10:22:52 +0100
Subject: [pytest-dev] Moving catchlog to pytest-dev (was: Re:
 pytest-capturelog status/merging to core?)
In-Reply-To: <20151109085215.GS16107@merlinux.eu>
References: <20150925114046.GD26059@merlinux.eu> <20150925120312.GB1393@tonks>
 <CA+RQFfdP0CXUyy_1mTtLcJUHVQpHbM1LWjDhxqYOrhcXBvWf0w@mail.gmail.com>
 <43A394C1-E54E-4F0A-9B9F-75F2772FD96F@ronnypfannschmidt.de>
 <20150925175526.GE26059@merlinux.eu>
 <A7BB3652-2404-49DD-963D-8FCDAC294213@ronnypfannschmidt.de>
 <56255DE4.2050104@mailbox.org> <20151020042042.GR29343@tonks>
 <20151109052424.GU5929@tonks> <20151109085215.GS16107@merlinux.eu>
Message-ID: <20151109092252.GW5929@tonks>

* holger krekel <holger at merlinux.eu> [2015-11-09 08:52:15 +0000]:
> On Mon, Nov 09, 2015 at 06:24 +0100, Florian Bruhin wrote:
> > Hey,
> > 
> > * Florian Bruhin <me at the-compiler.org> [2015-10-20 06:20:42 +0200]:
> > > * Arthur Skowronek <eisensheng at mailbox.org> [2015-10-19 23:17:24 +0200]:
> > > > It would be nice if the plugin could stay separated from the core package.
> > > > At least I am a huge fan of the modular plugin system as it can be found in
> > > > pytest.  I find such a modular design more preferable compared to a
> > > > monolithic project that comes with everything attached. Such a large project
> > > > makes it harder for users to track what was changed between versions and
> > > > keep track of changes important to them.  They don't need to care at all if
> > > > the API was changed but they don't use the plugin.
> > > > 
> > > > Also different contributors stated that they still would like to improve the
> > > > code base and add more functionality.  It would be nice to give them their
> > > > freedom to break the API if necessary to create something even more usable
> > > > with a sleeker API.  I'm looking forward to it.
> > > 
> > > I used to have a different stance on this - but after reconsidering, I
> > > agree it'd be best to have a "recommended" plugin (i.e. catchlog, as
> > > it's maintained and under pytest-dev) rather than merge things into
> > > core.
> > 
> > Backwards compatibility is restored and eisensheng released 1.2.0 - so
> > I think everything is ready to move pytest-catchlog to pytest-dev now.
> > 
> > Of course +1 from my side - any objections and/or a second +1 please?
> > ;)
> 
> I noticed that there now also is a pytest-logging release from Pedro Algarvio:
> 
>     https://pypi.python.org/pypi/pytest-logging
> 
> which adds to the confusion around logging and pytest, unfortunately.
> 
> I haven't used pytest-catchlog myself yet but intend to do so.
> It'd be great if eisenheng or you could get in contact with Pedro
> and see how we can all arrive in a situation which is less confusing
> to users, somewhat independently from the question of merging catchlog
> into pytest-dev.

Fun! It seems that plugin simply adds --logging-format and
--logging-date-format options, and configures the verbosity based on
-v:

https://github.com/saltstack/pytest-logging/blob/master/pytest_logging/plugin.py

I opened an issue there discussing merging it into pytest-catchlog:

https://github.com/saltstack/pytest-logging/issues/1

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: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20151109/b16fd8ff/attachment.sig>

From florian.schulze at gmx.net  Mon Nov  9 07:27:44 2015
From: florian.schulze at gmx.net (Florian Schulze)
Date: Mon, 09 Nov 2015 13:27:44 +0100
Subject: [pytest-dev] Moving catchlog to pytest-dev (was: Re:
 pytest-capturelog status/merging to core?)
In-Reply-To: <20151109085215.GS16107@merlinux.eu>
References: <CA+RQFffJ-zPsoMW900SULWJGKWqCJGWB1cavL283pPV-EVLL5g@mail.gmail.com>
 <20150925114046.GD26059@merlinux.eu> <20150925120312.GB1393@tonks>
 <CA+RQFfdP0CXUyy_1mTtLcJUHVQpHbM1LWjDhxqYOrhcXBvWf0w@mail.gmail.com>
 <43A394C1-E54E-4F0A-9B9F-75F2772FD96F@ronnypfannschmidt.de>
 <20150925175526.GE26059@merlinux.eu>
 <A7BB3652-2404-49DD-963D-8FCDAC294213@ronnypfannschmidt.de>
 <56255DE4.2050104@mailbox.org> <20151020042042.GR29343@tonks>
 <20151109052424.GU5929@tonks> <20151109085215.GS16107@merlinux.eu>
Message-ID: <90096DCE-CC5A-46B3-A359-475B012C0273@gmx.net>



On 9 Nov 2015, at 9:52, holger krekel wrote:

> On Mon, Nov 09, 2015 at 06:24 +0100, Florian Bruhin wrote:
>> Hey,
>>
>> * Florian Bruhin <me at the-compiler.org> [2015-10-20 06:20:42 +0200]:
>>> * Arthur Skowronek <eisensheng at mailbox.org> [2015-10-19 23:17:24 
>>> +0200]:
>>>> It would be nice if the plugin could stay separated from the core 
>>>> package.
>>>> At least I am a huge fan of the modular plugin system as it can be 
>>>> found in
>>>> pytest.  I find such a modular design more preferable compared to a
>>>> monolithic project that comes with everything attached. Such a 
>>>> large project
>>>> makes it harder for users to track what was changed between 
>>>> versions and
>>>> keep track of changes important to them.  They don't need to care 
>>>> at all if
>>>> the API was changed but they don't use the plugin.
>>>>
>>>> Also different contributors stated that they still would like to 
>>>> improve the
>>>> code base and add more functionality.  It would be nice to give 
>>>> them their
>>>> freedom to break the API if necessary to create something even more 
>>>> usable
>>>> with a sleeker API.  I'm looking forward to it.
>>>
>>> I used to have a different stance on this - but after reconsidering, 
>>> I
>>> agree it'd be best to have a "recommended" plugin (i.e. catchlog, as
>>> it's maintained and under pytest-dev) rather than merge things into
>>> core.
>>
>> Backwards compatibility is restored and eisensheng released 1.2.0 - 
>> so
>> I think everything is ready to move pytest-catchlog to pytest-dev 
>> now.
>>
>> Of course +1 from my side - any objections and/or a second +1 please?
>> ;)
>
> I noticed that there now also is a pytest-logging release from Pedro 
> Algarvio:
>
>  https://pypi.python.org/pypi/pytest-logging
>
> which adds to the confusion around logging and pytest, unfortunately.
>
> I haven't used pytest-catchlog myself yet but intend to do so.

I just tested it with devpi-server and it's mostly a dropin replacement.
It doesn't support Python 2.6 though, so we either have to drop support 
for that as well, or restore compatibility. For now it looks like one 
place is affected in catchlog.
Unfortunately catchlog doesn't fix the pypy test failures in 
devpi-server, but at least there we can get a fix in if we find one.

Regards,
Florian Schulze

From flub at devork.be  Thu Nov 12 08:34:44 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Thu, 12 Nov 2015 13:34:44 +0000
Subject: [pytest-dev] ANN: pytest-logbook
Message-ID: <CAAWk_Dz3gjWLNqPTZ-UuEkjW6tD53SA3TwK2HwKegcMDv8+U2g@mail.gmail.com>

Hi all,

To add to the confusion of a whole bunch of py.test plugins related to
logging I've just released a very first version of pytest-logbook.
This plugin captures log records emitted via logbook and shows them in
test reports as well as making them available via some fixtures.

pypi: https://pypi.python.org/pypi/pytest-logbook/1.0.1
repo: https://bitbucket.org/flub/pytest-logbook

Hopefully you find this useful if you use py.test and logbook.  Feel
free to give any feedback you might have.

Regards,
Floris

From andreas at pelme.se  Thu Nov 12 08:46:36 2015
From: andreas at pelme.se (Andreas Pelme)
Date: Thu, 12 Nov 2015 14:46:36 +0100
Subject: [pytest-dev] ANN: pytest-logbook
In-Reply-To: <CAAWk_Dz3gjWLNqPTZ-UuEkjW6tD53SA3TwK2HwKegcMDv8+U2g@mail.gmail.com>
References: <CAAWk_Dz3gjWLNqPTZ-UuEkjW6tD53SA3TwK2HwKegcMDv8+U2g@mail.gmail.com>
Message-ID: <B8AC9BAC-F04B-4DDD-859C-640D21EE87BD@pelme.se>

> 
> On 12 nov 2015, at 14:34, Floris Bruynooghe <flub at devork.be> wrote:
> 
> 
> Hopefully you find this useful if you use py.test and logbook.  Feel
> free to give any feedback you might have.

Thanks Floris.

I?m using logbook and have a similar log_handler fixture that initiates/pushes TestHandler. The reporting looks nice and is something I lack with my hack. :)

Cheers,
Andreas

From flub at devork.be  Thu Nov 12 09:04:35 2015
From: flub at devork.be (Floris Bruynooghe)
Date: Thu, 12 Nov 2015 14:04:35 +0000
Subject: [pytest-dev] ANN: pytest-logbook
In-Reply-To: <B8AC9BAC-F04B-4DDD-859C-640D21EE87BD@pelme.se>
References: <CAAWk_Dz3gjWLNqPTZ-UuEkjW6tD53SA3TwK2HwKegcMDv8+U2g@mail.gmail.com>
 <B8AC9BAC-F04B-4DDD-859C-640D21EE87BD@pelme.se>
Message-ID: <CAAWk_DzQXoQ2xmHbVQrTxMGGqTe9jBe+8vtS4gcWmQeZ5MHi_Q@mail.gmail.com>

On 12 November 2015 at 13:46, Andreas Pelme <andreas at pelme.se> wrote:
>>
>> On 12 nov 2015, at 14:34, Floris Bruynooghe <flub at devork.be> wrote:
>>
>>
>> Hopefully you find this useful if you use py.test and logbook.  Feel
>> free to give any feedback you might have.
>
> Thanks Floris.
>
> I?m using logbook and have a similar log_handler fixture that initiates/pushes TestHandler. The reporting looks nice and is something I lack with my hack. :)

Yeah, we had some ad-hoc fixtures for a while but started needing them
in multiple projects and got bored of copy-paste.  So that's what led
to cleaning it up and creating pytest-logbook.  If you migrate let me
know if you have anything tricky or if it's lacking some features.
One feature I've not included that we do locally is a check that
ensures no error or warning level messages have been logged unless the
test had a mark applied.  But I thought starting minimal would be a
best.

From opensource at ronnypfannschmidt.de  Thu Nov 19 15:59:20 2015
From: opensource at ronnypfannschmidt.de (Ronny)
Date: Thu, 19 Nov 2015 21:59:20 +0100
Subject: [pytest-dev] pytest-2.8.3 released - bug fixes
Message-ID: <564E3828.1090204@ronnypfannschmidt.de>

Hello Everyone,

i'm pleased to announce the release of


pytest-2.8.3: 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 2.8.2.

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
        Florian Bruhin
        Gabe Hollombe
        Gabriel Reis
        Hartmut Goebel
        John Vandenberg
        Lee Kamentsky
        Michael Birtwell
        Raphael Pierzina
        Ronny Pfannschmidt
        William Martin Stewart

Happy testing,
The py.test Development Team


2.8.3 (compared to 2.8.2)
-----------------------------

- fix #1169: add __name__ attribute to testcases in TestCaseFunction to
  support the @unittest.skip decorator on functions and methods.
  Thanks Lee Kamentsky for the PR.

- fix #1035: collecting tests if test module level obj has __getattr__().
  Thanks Suor for the report and Bruno Oliveira / Tom Viner for the PR.

- fix #331: don't collect tests if their failure cannot be reported
correctly
  e.g. they are a callable instance of a class.

- fix #1133: fixed internal error when filtering tracebacks where one entry
  belongs to a file which is no longer available.
  Thanks Bruno Oliveira for the PR.

- enhancement made to highlight in red the name of the failing tests so
  they stand out in the output.
  Thanks Gabriel Reis for the PR.

- add more talks to the documentation
- extend documentation on the --ignore cli option
- use pytest-runner for setuptools integration
- minor fixes for interaction with OS X El Capitan system integrity
protection (thanks Florian)