I'd like to hear, what are other problems or common things people
encounter when writing tests?
On 04/14/2015 11:07 AM, Cory Johns wrote:
As Nate pointed out, I forgot the link. >_<
[1]: https://docs.python.org/2/library/unittest.html
On Tue, Apr 14, 2015 at 1:59 PM, Cory Johns <[email protected]> wrote:
José,
That's a good point that we definitely need recommended best practices
for writing charm tests. I am also behind using the built-in
unittest[1] library to organize and structure the tests. I am also
behind encouraging combining tests inside test cases where possible,
with the caveat that we frame the recommendation with explicitly
calling out the tear-down behavior and explaining why it is
recommended to combine test cases, since there will be cases where
that is not the best approach because a new environment is required
(even though it will be the right thing to do most of the time).
I think we should also include a call-out to the template test
provided by `charm add tests` (or `charm create`) as the best way to
get started, as it also incorporates these practices. However, we
should note about changing the name of the generated test case file
away from 99-autogen.
Thanks for bringing this up and offering these good suggestions.
On Thu, Apr 9, 2015 at 6:17 PM, José Antonio Rey <[email protected]> wrote:
Hello everyone,
A couple days ago I was going through the docs, and found out that there are
not best practices or recommendations for tests. Now that we are trying to
ensure quality, I believe it would be a good idea to have them.
I know that one of the benefits of Juju being so open is that you can write
your charm or tests in any language. However, having these best practices
will allow us to set a recommendation for when people are getting started,
or don't know how to proceed. I understand that this may be seen as a
limitation, but it is instead a guidance for when someone comes to the
ecosystem, or wants to get some tips on how to write tests.
I was thinking on starting with two points. These are assuming that you are
using Python and Amulet to write your tests. The first one, the use of the
unit_tests packages. As a starter with Python, I have found out the use of
it quite easy, and helps me condensate everything into one big file. Which
takes me to the second point, having everything in one file. This would
allow us to save resources and time (since bundletester tears down for each
file), and would make it look nicer. At least that's how I've been
proceeding for my last tests.
What do you think? I'm all ears to new suggestions or why these aren't good.
--
José Antonio Rey
--
Juju mailing list
[email protected]
Modify settings or unsubscribe at:
https://lists.ubuntu.com/mailman/listinfo/juju
--
José Antonio Rey
--
Juju mailing list
[email protected]
Modify settings or unsubscribe at:
https://lists.ubuntu.com/mailman/listinfo/juju
