[Zope-dev] Re: Running all unittests in $INSTANCE_HOME

Tres Seaver tseaver at palladion.com
Tue Nov 22 13:10:54 EST 2005


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jim Fulton wrote:
> On Fri, 2005-11-18 at 08:43 -0500, Tres Seaver wrote:
> 
>>-----BEGIN PGP SIGNED MESSAGE-----
>>Hash: SHA1
>>
>>Chris Withers wrote:
>>
>>>Paul Winkler wrote:
>>>
>>>
>>>>On Thu, Nov 17, 2005 at 02:33:10PM -0200, Sidnei da Silva wrote:
>>>>
>>>>
>>>>>I might have asked this before, but can't seem to find it.
>>>>>
>>>>>I'm looking for a way to run all tests in $INSTANCE_HOME using the
>>>>>test.py script, but can't seem to get the majik combination of
>>>>>parameters together. Would anyone hint me?
>>>>
>>>>
>>>>"./bin/zopectl test"  is supposed to do the job.
>>>>But in practice, so many third-party product tests either have
>>>>broken dependencies or pollute the test environment somehow
>>>>that something always breaks.
>>>
>>>
>>>bin/zopectl test Products/MyProduct
>>>
>>>...has worked fine for me for many months now...
>>
>>Hmm, that should be a dotted name, rather than a path:
>>
>> $ bin/zopectl test Products.MyProduct
>>
>>Because the first argument is interpreted as a regex, using search, you
>>can leave the 'Product.' bit off, as well as supplying more complicated
>>specifications, e.g.:
>>
>> $ bin/zopectl test OneProduct|AnotherProduct
>>
>>Run with --help to get docs on the command-line syntax:
>>
>> $ bin/zopectl test --help
>>
>>That pattern breaks on the Zope trunk, where you have to pass the module
>>pattern as an option rather than an argument:
>>
>> $ bin/zopectl test -m MyProduct
> 
> 
> You can still pass a module pattern and a test pattern as positional
> arguments.  This is for backward compatibility.

It's broken, then:

 $ ../bin/python2.4 test.py -v -m PythonScripts
 Running tests at level 1
 Running unit tests:
   Running:
     .............................
   Ran 29 tests with 0 failures and 0 errors in 0.426 seconds.

versus:

 $ ../bin/python2.4 test.py -v PythonScripts
 Running tests at level 1
 Running unit tests:
   Running:
     ..................................................
     ..................................................
     (snip)
     .........................
   Ran 4175 tests with 0 failures and 0 errors in 147.654 seconds.

While we're at it, what is the rationale for the "50 column indented"
output (in "dots" mode) for the new testrunner, versus the "80 column
flush" output which is standard in Python / Java CLI test runners?


Tres.
- --
===================================================================
Tres Seaver          +1 202-558-7113          tseaver at palladion.com
Palladion Software   "Excellence by Design"    http://palladion.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFDg18u+gerLs4ltQ4RAhrGAJ0QZ+oIRnUaO7TPUF8JYuLfQc4jHQCgk2lY
KaZMTRVAEPQamq5rHwrImLE=
=ygIU
-----END PGP SIGNATURE-----



More information about the Zope-Dev mailing list