[issue1960] test_gdbm.py converted to unittest

[Brett Cannon]

Brett Cannon <[EMAIL PROTECTED]> added the comment:

Committed in revision 61374 (w/ changes so that the key stuff is not
order-dependent). Thanks, Giampaolo!

--
resolution:  -> accepted
status: open -> closed

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Brett Cannon]

Changes by Brett Cannon:


--
assignee:  -> brett.cannon

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Christian Heimes]

Changes by Christian Heimes:


--
keywords: +patch
priority:  -> normal

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Giampaolo Rodola']

Giampaolo Rodola' added the comment:

The diff file in attachment was generated by using TortoiseSVN and it
seems to be different than using "svn diff".
The new diff file is in attachment.

Added file: http://bugs.python.org/file9363/test_gdbm.diff

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Giampaolo Rodola']

Changes by Giampaolo Rodola':


Added file: http://bugs.python.org/file9323/test_gdbm.py

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Giampaolo Rodola']

Giampaolo Rodola' added the comment:

Updated version in attachment. 
Changes:

- Added "if self.g is not None: self.g.close()" clause in tearDown.
- Used "test.test_support.unlink(filename)" instead of the try/except
statement.
- Added tests for the flag clause in the open() statement by trying all
supported mode flags. I also called the method "test_flags" instead of
"test_modes" since "mode" is a kwarg for open().
- Added tests for trying to open non existing database with flag == 'r'
and 'w'.

Untested methods which are now tested:
- keys().
- firstkey().
- nextkey().
- reorganize()

The only method currently not tested yet is sync(). I'm not sure how to
test it, maybe I'm misunderstanding its purpose.

Added file: http://bugs.python.org/file9322/test_gdbm.diff

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Brett Cannon]

Brett Cannon added the comment:

On Jan 28, 2008 6:20 PM, Giampaolo Rodola' <[EMAIL PROTECTED]> wrote:
>
> Giampaolo Rodola' added the comment:
>
> > Well, if you have an opinion, feel free to leave a comment in this
> > issue about it. I will most likely be the one who does the checkin and
> > I will read this issue before I commit.
>
>  * One of the things I dislike is the fact that the student used "self.g
> = gdbm.open(self.filename, 'c')" in setUp method since if that fails
> there will be a NameError exception raised in tearDown method and the
> errors reported by unittest would be 2 where in fact it would be only 1.
> Probably a thing like this one would have been better:
>
> def setUp(self):
> self.g = None
>
> def tearDown(self):
> if self.g is not None:
> self.g.close()
>   ...
>
> def test_it(self):
> self.g = gdbm.open(self.filename, 'c')
> ...
>

Could also still open perform the open() call in setUp() and do what
you need in the tearDown() to not be an error. There should probably
also be a way to not cause an error if the file path is just not
writable as that is not a sign of a failing test but an unavailable
resource.

> - Another thing I don't like is that os.unlink(self.filename) executed
> in tearDown would be better if protected by a try/except statement.
>

Even better is test.test_support.unlink() which handles those details for you.

> - +1 for the self.g.close() used by the student in the tearDown method.
> That was a thing I haven't considered and it's not included in my patch.
>
> - +0.5 for the "test_modes" method added by the student. Maybe it's
> useful, maybe it's not, I don't know.
>
>
> Aside from that I don't see other relevant differences since the code is
> really minimalistic. Feel free to commit the patch you consider to be
> the best.
> If you want me to do so I can provide a merged version of my patch
> including the differences I described above.

Fine by me. Or you can wait until I have time to look at your code.
It's up to you.

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Giampaolo Rodola']

Giampaolo Rodola' added the comment:

> Well, if you have an opinion, feel free to leave a comment in this
> issue about it. I will most likely be the one who does the checkin and
> I will read this issue before I commit.

 * One of the things I dislike is the fact that the student used "self.g
= gdbm.open(self.filename, 'c')" in setUp method since if that fails
there will be a NameError exception raised in tearDown method and the
errors reported by unittest would be 2 where in fact it would be only 1.
Probably a thing like this one would have been better:

def setUp(self):
self.g = None

def tearDown(self):
if self.g is not None:
self.g.close()
  ...

def test_it(self):
self.g = gdbm.open(self.filename, 'c')
...

- Another thing I don't like is that os.unlink(self.filename) executed
in tearDown would be better if protected by a try/except statement.

- +1 for the self.g.close() used by the student in the tearDown method.
That was a thing I haven't considered and it's not included in my patch.

- +0.5 for the "test_modes" method added by the student. Maybe it's
useful, maybe it's not, I don't know. 


Aside from that I don't see other relevant differences since the code is
really minimalistic. Feel free to commit the patch you consider to be
the best. 
If you want me to do so I can provide a merged version of my patch
including the differences I described above.

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Brett Cannon]

Brett Cannon added the comment:

On Jan 28, 2008 5:03 PM, Giampaolo Rodola' <[EMAIL PROTECTED]> wrote:
>
> Giampaolo Rodola' added the comment:
>
> > Have a look at
> http://code.google.com/p/google-highly-open-participation-psf/issues/detail?id=290
> > where a GHOP student did a conversion as well.
>
> Interesting, didn't know about that.
> In future I'll check that site before start working on a new patch
> involving the test suite since it seems those students are already
> working on these kind of issues.
>

We have held off on committing their code until we get contributor
agreements from them. Hopefully that will all get resolved in February
and thus you won't have to check the GHOP site for long.

> > Any chance to come up with a possible merged version
> > that takes the best from your work, Giampaolo and the
> > student's work?
>
> Don't know... the suite is extremely simple and, IMO, I don't even think
> there's a real need to do that.
> I noticed that the student added/changed some stuff like the new
> "test_modes" method which maybe could be useful to have.

Well, if you have an opinion, feel free to leave a comment in this
issue about it. I will most likely be the one who does the checkin and
I will read this issue before I commit.

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Giampaolo Rodola']

Giampaolo Rodola' added the comment:

> Have a look at
http://code.google.com/p/google-highly-open-participation-psf/issues/detail?id=290
> where a GHOP student did a conversion as well.

Interesting, didn't know about that.
In future I'll check that site before start working on a new patch
involving the test suite since it seems those students are already
working on these kind of issues.

> Any chance to come up with a possible merged version 
> that takes the best from your work, Giampaolo and the
> student's work?

Don't know... the suite is extremely simple and, IMO, I don't even think
there's a real need to do that.
I noticed that the student added/changed some stuff like the new
"test_modes" method which maybe could be useful to have.

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Brett Cannon]

Brett Cannon added the comment:

Have a look at 
http://code.google.com/p/google-highly-open-participation-psf/issues/detail?id=290
where a GHOP student did a conversion as well. Any chance to come up
with a possible merged version that takes the best from your work,
Giampaolo and the student's work?

--
nosy: +brett.cannon

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Giampaolo Rodola']

New submission from Giampaolo Rodola':

In attachment. All existent tests are unchanged.

--
components: Tests
files: test_gdbm.diff
messages: 61806
nosy: giampaolo.rodola
severity: normal
status: open
title: test_gdbm.py converted to unittest
type: rfe
versions: Python 2.6
Added file: http://bugs.python.org/file9315/test_gdbm.diff

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue1960] test_gdbm.py converted to unittest

[Giampaolo Rodola']

Changes by Giampaolo Rodola':


Added file: http://bugs.python.org/file9316/test_gdbm.py

__
Tracker <[EMAIL PROTECTED]>

__
___
Python-bugs-list mailing list 
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com