Improve string representation of matcher in allure report

[Denis-Alexeev]

Matchers in allure report now look like:
<hamcrest.core.core.isnot.IsNot object at 0x7f7262d17c40>

PR purpose:
Make it look more pretty

[offbyone]

I would prefer something closer to this:

    def __repr__(self) -> str:
        """Returns the description."""
        return "<{}({})>".format(self.__class__.__name__, truncate(self.out, 60))

Basically, I don't think losing the type info is a good plan, and having the debug repr be an unbounded string also has issues. You'd need to implement the truncate method if we don't have one somewhere else in the library.

Also, if you do this, please put a oneliner in changelog.d for it, probably as a "feature".

[Denis-Alexeev]

I've fixed the code according to you remarks. Not sure about the reason for CI failure. Could you please help me with that?

[brunns]

Re the CI failure - if you run tox -e lint before your commit, I believe black has some formatting changes.

I think this needs a test of some kind, too.

[brunns]

Tests (In BaseMatcherTest) might look something like:

    def testMatcherReprShouldDescribeMatcher(self):
        assert repr(FailingBaseMatcher()) == "FailingBaseMatcher(SOME DESCRIPTION)"

    def testMatcherReprShouldTruncateLongDescription(self):
        class LongDescriptionMatcher(BaseMatcher):
            def describe_to(self, description):
                description.append_text("1234 " * 13)
                
        assert repr(LongDescriptionMatcher()) == "LongDescriptionMatcher(1234 1234 1234 1234 1234 1234 1234 1234 1234 1234 1234 ...)"

[brunns]

The PR is just awaiting tests and a change log AFAIC.

[Denis-Alexeev]

Hello guys! Sorry for the delay. Could you please take a look and give your comments if needed?

[brunns]

Absolutely - I'll take a look on Monday, when I'm back at my computer.