[TIP] Testing optparse with Nosetests
Dan Wandschneider
daniel.wandschneider at schrodinger.com
Thu Nov 1 10:48:36 PDT 2012
You might try this:
def opts(args=None):
"""
Using optparser to parse any commandline options.
@return:
"""
parser = optparse.OptionParser()
parser.add_option(....
options, args = parser.parse_args(args)
return args
Then in tests do:
opts(['-testarg'])
Also, you shouldn't use an empty list as the default value for the module
variable ARGS. Instead, use None or possibly an empty tuple. See this
StackOverflow article for an explanation:
http://stackoverflow.com/q/1132941/1309332.
On Wed, Oct 31, 2012 at 11:57 PM, Tim Aerdts <fragger123 at gmail.com> wrote:
> Hello,
>
> I've decided to go with Dan's initial solution. But a little different.
> In utils.py
>
> ARGS = []
>
> def opts():
> """
> Using optparser to parse any commandline options.
> @return
> """
> parser = optparse.OptionParser()
>
> parser.add_option(....
>
> if len(ARGS):
> options, args = parser.parse_args(ARGS)
> else:
> options, args = parser.parse_args()
> return options
>
> Then in my tests I do:
> from utils import ARGS
> ARGS.append('-testarg')
>
> For example. This might not be very pythonic I don't know and I might
> refactor it at a later stage but for now it works.
>
> Thank you for all your tips and help.
> Oh and I couldn't use argparse because I am stuck with 2.6 where argparse
> was introduced in 2.7
>
> Cheers,
>
> -Tim
>
> On Wed, Oct 31, 2012 at 6:59 PM, Dan Wandschneider <
> daniel.wandschneider at schrodinger.com> wrote:
>
>> Tim-
>> Yeah, that seems... not ideal. I'd strongly recommend against adding
>> these options to your parser. It sounds like get_opts() is probably the
>> name of the theoretical function I was referring to
>> as my_parsing_function(). I truly would recommend that you give get_opts()
>> an args parameter defaulting to None. Devs may whine about this, but it is
>> good practice; they should have done it in the first place. This would
>> also allow you to test arguments that are not arguments to Nose.
>>
>> If you definitely cannot change get_opts() (the recommended solution),
>> you could modify sys.argv. I'd recommend a context manager:
>>
>> class ArgvHider(object):
>> def __init__(self, *newargs):
>> self._newargs = list(newargs)
>> def __enter__(self):
>> self._argv = sys.argv
>> sys.argv = self._newargs
>> def __exit__(self, exc_type, exc_value, traceback):
>> sys.argv = self._argv
>>
>> Use like:
>> print sys.argv
>> with ArgvHider('-arg1', 'arg1val'):
>> #argv is overridden with my custom values!
>> print sys.argv
>> #now the original argv is back!
>> print sys.argv
>>
>> -Dan
>>
>>
>>
>> On Tue, Oct 30, 2012 at 11:56 PM, Tim Aerdts <fragger123 at gmail.com>wrote:
>>
>>> Hello Dan,
>>>
>>> Thanks for your reply! However I can not use a custom argslist. I am
>>> testing some methods and functions that call a function get_opts() which
>>> parses the options. So I do not do any parse_args() calls in my tests or
>>> anywhere near it.
>>>
>>> I've more or less solved it by adding
>>>
>>> parser.add_option('--with-coverage')
>>> parser.add_option('--with-xunit')
>>> parser.add_option('--cover-xml')
>>> parser.add_option('--verbose')
>>> parser.add_option('--cover-package')
>>> parser.add_option('--nocapture')
>>>
>>> To the parser, but I am sure this is not the way to go.
>>>
>>> Also my setup of optparse might not be that ideal. Everytime I need to
>>> get something from the arguments I make a call to get_opts() which returns
>>> the options from parse_args() so if I need something called myval I always
>>> do get_opts().myval
>>>
>>> Cheers,
>>>
>>> -Tim
>>>
>>> On Tue, Oct 30, 2012 at 8:45 PM, Dan Wandschneider <
>>> daniel.wandschneider at schrodinger.com> wrote:
>>>
>>>> Tim-
>>>> optparse.parse_args() parses all arguments given on the command line,
>>>> so it is probably balking on the commands that you gave to Nose. In tests,
>>>> you should create the specific argument list that you want to test, and
>>>> then run parser.parse_args(arglist). For example:
>>>> argslist = ['-h']
>>>> with assertRaises(SystemExit):
>>>> myparser.parse_args(argslist)
>>>>
>>>> If your parser is wrapped in a function or method, use this:
>>>> def my_parsing_function(args=None):
>>>> myparser.parse_args(args)
>>>>
>>>> If args == None, OptParse will use sys.argv, so your production code is
>>>> safe. (Also, if you are writing new code, I believe that ArgParse is
>>>> preferred to OptParse)
>>>>
>>>> -Dan W.
>>>> http://www.schrodinger.com/
>>>>
>>>> On Tue, Oct 30, 2012 at 12:00 PM, <
>>>> testing-in-python-request at lists.idyll.org> wrote:
>>>>
>>>>> Send testing-in-python mailing list submissions to
>>>>> testing-in-python at lists.idyll.org
>>>>>
>>>>> To subscribe or unsubscribe via the World Wide Web, visit
>>>>> http://lists.idyll.org/listinfo/testing-in-python
>>>>> or, via email, send a message with subject or body 'help' to
>>>>> testing-in-python-request at lists.idyll.org
>>>>>
>>>>> You can reach the person managing the list at
>>>>> testing-in-python-owner at lists.idyll.org
>>>>>
>>>>> When replying, please edit your Subject line so it is more specific
>>>>> than "Re: Contents of testing-in-python digest..."
>>>>>
>>>>> Today's Topics:
>>>>>
>>>>> 1. Testing optparse with Nosetests (Tim Aerdts)
>>>>> 2. Re: Testing optparse with Nosetests (Tim Aerdts)
>>>>>
>>>>>
>>>>> ---------- Forwarded message ----------
>>>>> From: Tim Aerdts <fragger123 at gmail.com>
>>>>> To: testing-in-python at lists.idyll.org
>>>>> Cc:
>>>>> Date: Tue, 30 Oct 2012 09:42:49 +0100
>>>>> Subject: [TIP] Testing optparse with Nosetests
>>>>> Hello,
>>>>>
>>>>> Posting this here because it seems more active then the Nosetests
>>>>> users list.
>>>>> Anyway I am in the process of writing an application which makes use
>>>>> of Optparse. I'm running the tests with Nosetests but I fear these two
>>>>> might be interfering?
>>>>>
>>>>> When I run parser.parse_args() anywhere in my code Nosetests bugs out.
>>>>>
>>>>> Usage: nosetests [options]
>>>>> nosetests: error: no such option: --with-coverage
>>>>>
>>>>> Removing the call to parse_args() and everything runs fine.
>>>>>
>>>>> Cheers,
>>>>>
>>>>> --
>>>>> Kind regards,
>>>>> Tim Aerdts
>>>>> http://www.tuimz.nl
>>>>>
>>>>>
>>>>> ---------- Forwarded message ----------
>>>>> From: Tim Aerdts <fragger123 at gmail.com>
>>>>> To: testing-in-python at lists.idyll.org
>>>>> Cc:
>>>>> Date: Tue, 30 Oct 2012 11:00:53 +0100
>>>>> Subject: Re: [TIP] Testing optparse with Nosetests
>>>>> Just a follow-up.
>>>>>
>>>>> If I add the options that I use with nosetests (--with-coverage
>>>>> --verbose --cover-package=mypackage) to my own optparser it works as
>>>>> expected.
>>>>>
>>>>> I don't understand this, and preferably I don't need those
>>>>> dependencies in my main application..
>>>>>
>>>>> On Tue, Oct 30, 2012 at 9:42 AM, Tim Aerdts <fragger123 at gmail.com>wrote:
>>>>>
>>>>>> Hello,
>>>>>>
>>>>>> Posting this here because it seems more active then the Nosetests
>>>>>> users list.
>>>>>> Anyway I am in the process of writing an application which makes use
>>>>>> of Optparse. I'm running the tests with Nosetests but I fear these two
>>>>>> might be interfering?
>>>>>>
>>>>>> When I run parser.parse_args() anywhere in my code Nosetests bugs out.
>>>>>>
>>>>>> Usage: nosetests [options]
>>>>>> nosetests: error: no such option: --with-coverage
>>>>>>
>>>>>> Removing the call to parse_args() and everything runs fine.
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>> --
>>>>>> Kind regards,
>>>>>> Tim Aerdts
>>>>>> http://www.tuimz.nl
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Kind regards,
>>>>> Tim Aerdts
>>>>> http://www.tuimz.nl
>>>>>
>>>>> _______________________________________________
>>>>> testing-in-python mailing list
>>>>> testing-in-python at lists.idyll.org
>>>>> http://lists.idyll.org/listinfo/testing-in-python
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> testing-in-python mailing list
>>>> testing-in-python at lists.idyll.org
>>>> http://lists.idyll.org/listinfo/testing-in-python
>>>>
>>>>
>>>
>>>
>>> --
>>> Kind regards,
>>> Tim Aerdts
>>> http://www.tuimz.nl
>>>
>>
>>
>
>
> --
> Kind regards,
> Tim Aerdts
> http://www.tuimz.nl
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.idyll.org/pipermail/testing-in-python/attachments/20121101/95d7314e/attachment-0001.htm>
More information about the testing-in-python
mailing list