On Thu, Jan 16, 2020 at 4:20 PM Edward K. Ream <edream...@gmail.com> wrote: [...]
> The test_one_line_pet_peeves function contains a table of snippets. Modulo > some housekeeping, it is: > > for contents in table: # table not shown. Some tests in the table fail. > contents, tokens, tree = self.make_data(contents, tag) > expected = self.blacken(contents) > results = self.beautify(contents, tag, tokens, tree) > if results != expected: > fails += 1 > assert fails == 0, fails > > *All the "one-line tests" run, even if some fail.* > > It's also convenient to "pretend" that the test passes even when it > doesn't. That way the -x (fail fast) option doesn't bother us. So, for > now, the last line is: > > assert fails == 13, fails > > As I fix the unit test failures I'll decrease the error count. > It is possible to let pytest do the bookkeeping for you by using a pytest decorator to parameterize the test. See https://docs.pytest.org/en/latest/parametrize.html. Also, the @pytest.mark.xfail decorator ( https://docs.pytest.org/en/latest/skipping.html#xfail-mark-test-functions-as-expected-to-fail) is useful for "a bug not yet fixed" which sounds like what you are describing. The parameterize decorator supports marking the input arguments with xfail independent of each other. -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/leo-editor/CAO5X8CzOBi0G%3DRxJFfUqpXjAx4DxSJs70sXSxeBxajs6tg1a1Q%40mail.gmail.com.
