On Feb 17, 1:03 am, "C. Titus Brown" <[email protected]> wrote:

> directory.  What do you think of making a 'testlib' package, or a
> subdirectory that contains all of the test utilities?

sure no problem, I typically wait for it to be large enough before
refactoring into its own package, and about now is at that limit

> Test running is a little bit clumsy at the moment when you want to run
> all of the tests.  The output is verbose, and it's hard to see failing
> tests as they scroll by.

There are two ways to deal with this, may not be obvious right away.
One is to reduce the verbosity
with the  -v 0 flag, or better yet to run it in strict mode with the -
s flag. This latter stops testing at the first error and is the way to
run the tests while expecting/fixing errors. We could change the
defaults.

> unittest compatibility.  This could eventually replace runtest, which is
> nice but is a bit superfluous in the face of community-supported
> frameworks like nose.

The only pet peeve I have regarding nose is that people end up writing
tests that can only be run with nose. Those who know nose well
underestimate the overhead of learning nose and forming the proper
mental picture of its workings. So there is a tradeoff there, I think
there are people who will run fewer tests because of nose. When
something happens automatically a large number of people have no idea
what is really going behind the scenes and that (over the long term)
may end up hurting more than it helps.

> The temp directory should probably be nuked each time the tests are run.
> The second time I ran the tests I got a prompt complaining that a file
> already existed & did I want to overwrite it?

It is nuked at the beginning, but it looks like one test, the download
is run twice within each testrun, once in the pygrdata_test.py and
once when I describe its workings in the doctests. It did not show up
so far because it looks like my unzip function does not ask for
overwrite permission.

> extra spacing inside of parantheses,

ouch, I thought the PEP was the other way and asked for the spaces so
I made an effort to do it that way ... icky-ick

> There are also some 'import' statements inside of functions, which is
> generally frowned upon -- see blast_test.py.

I moved out a ton of these, looks like I still missed a few.

Istvan
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"pygr-dev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/pygr-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to