New issue 50 by zerbie: overhaul tests
http://code.google.com/p/twill/issues/detail?id=50
I said to myself, self, I said, the last thing you need to do is take on
Testing Guy about the tests in his own project. So I determined to keep my
mouth shut, but apparently it's against my nature. Because: They are poorly
organized and very confusing and the nose output isn't useful because there
are like 18 tests per test function, so when one of them fails, it's really
hard to figure out what happened. And man, twilltestserver is getting
unwieldy, and it's only going to get worse. And hardly anything is
documented. And and and.
So. I think they need some work. And by "work," I'm pretty sure I mean "a
complete overhaul." Sorry! (But not really.)
... do we fight now? ;)
--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings
I have no objections, actually; twill was my first foray into testing.
A major simplification, I think, would be to be able to discover at least a
subset of
twill tests automatically without having an explicit Python test runner.
No other suggestions, though. I haven't thought much about it.
One note -- you breaka my coverage, I breaka your head!
I'll help.
One comment -- please be gentle. Framework overhauls: good. Specific issue
overhauls: good. Random rewrite for no good reason other than that you
hate my (~5
yr old) coding/testing style: bad.
:) Yeah, random rewrites for no particular reason aren't really my idea of
a great
time. My plan (such as it is) involves some reorganization and cleanup,
followed by a
nice long period of reflection. I like to think I haven't *completely* lost
my mind.
evidence, please?
;)
sounds good.
evidence of my continued sanity? it's a little thin on the ground. you'll
just have
to trust me!