> Most of the unit tests are trivial to run, you just need to fill in the > config.py.tmpl with some values and run testing/run_all_tests.sh. I've > been working on them to make them standalone, i.e. SetUp and TearDown > routines as needed. Only one, roottest.py requires root privileges. > The rest only require root privileges for setup and teardown.
I guess I would like to have the tests that do not require *any* setup at all to be clearly separated and runnable out of the box, and then gradually try to decrease the dependence on external factors for the remainder of them. Even filling in config.py.tmpl is not necessarily trivial because it's not clear what you can leave empty and what you can't, etc. > I suppose we should make run_all_tests.sh into an uber-test in Python, > but have not had the time to do so. Would love to see a summary at the > end, like "152 tests run, 152 passed, 0 skipped, 0 failed". I think the > testing platform on Python will let us do that, but I need to grok it. I'm not sure if there is an idiomatic/trivial way; but since we don't have hundreds and hundreds of tests I think it's okay to be slightly redundant and just have a top-level testing/test.py that lists the tests. I'll see about having a look at the unit tests today, I should make sure I have an environment for them anyway. -- / Peter Schuller PGP userID: 0xE9758B7D or 'Peter Schuller <[email protected]>' Key retrieval: Send an E-Mail to [email protected] E-Mail: [email protected] Web: http://www.scode.org
pgpUCU293S3YQ.pgp
Description: PGP signature
_______________________________________________ Mailing list: https://launchpad.net/~duplicity-team Post to : [email protected] Unsubscribe : https://launchpad.net/~duplicity-team More help : https://help.launchpad.net/ListHelp
