[reportlab-users] Moving to Bitbucket, and development roadmap..

Dinu Gherman gherman at darwin.in-berlin.de
Tue Mar 26 18:24:49 EDT 2013


Andy Robinson:


>> That's *very* short notice... ;-) BTW, could we deprecate reporting bugs

>> on this list, except maybe for discussing bugs reported in BitBucket?

>

> Tracker is best but not sure everyone reading this has a Bitbucket

> account or wants to create one. If anyone does report a bug to the

> list we will certainly add it to the tracker.


An account on BitBucket is not needed for raising an issue.


> People also tell me 'nose' is very nice. But it looks to me like the

> enhanced unittest in the standard library since 2.7 has copied most of

> the really important features: discovery, skipping, module and class

> level setup/teardown.


I think py.test has a couple of niceties, which I'd like to see in Unittest
as well, e.g. when showing a traceback with a failed "assert equal/unequal"
you don't only get a message saying some value is different from the one
you expect, but you get a more qualified statement, saying *how* the values
differ, which is really useful for compound datastructures. See example
below.

Regards,

Dinu


$ more test.py
def test():
assert [1, 2, [3, 4, 5]] == [1, 2, 3, [4, 5]]
$
$ py.test test.py
==================================== test session starts ====================================
platform darwin -- Python 2.7.3 -- pytest-2.3.4
collected 1 items

test.py F

========================================= FAILURES ==========================================
___________________________________________ test ____________________________________________

def test():

> assert [1, 2, [3, 4, 5]] == [1, 2, 3, [4, 5]]

E assert [1, 2, [3, 4, 5]] == [1, 2, 3, [4, 5]]
E At index 2 diff: [3, 4, 5] != 3
E Right contains more items, first extra item: [4, 5]

test.py:2: AssertionError
================================= 1 failed in 0.03 seconds ==================================



More information about the reportlab-users mailing list