[Core] Add location to tag expression expcetion

[cyocum]

Summary

This integrates changes to the way that TagExpressionParser works in Cucumber-JVM to fix #1976.

Details

Updated the TagPredicate and the CoreHookDefinition to catch TagExpressionException and add information from the relevant classes to the exception message which is then wrapped in a RuntimeException and rethrown. The rethrow as a RuntimeException is due to the fact that the constructor for TagExpressionParser is package scope only and is not available in other packages.

Motivation and Context

This is to fix issue #1976.

How Has This Been Tested?

Added tests to both TagPredicateTest and CoreHookDefinition test to see if the proper exception was rethrown and if the information expected was included in the expected exception.

Screenshots (if appropriate):

Types of changes

• Bug fix (non-breaking change which fixes an issue).
• New feature (non-breaking change which adds functionality).
• Breaking change (fix or feature that would cause existing functionality to not work as expected).

Checklist:

• I've added tests for my code.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.

[mpkorstanje]

You originally pushed this down into the different hook implementations. I think because I suggested something about being as close to the source as possible. However this means that we'll have to implement the correct location reporting in all different hook implementations. Using the getLocation method we can get the location of the hook. So this is close enough.

[mpkorstanje]

We'll have to look at this. Currently the problem is that it doesn't report back where the problem came from. But this is true for all other CLI arguments as well.

[cyocum]

It may be easier if we have a shared session to discuss the options.

[mpkorstanje]

I've reverted the changes to cucumber-java and cucumber-java8. I also cleaned up some non-idiomatic java stuff. You can use Optional to avoid null checks but only when a method already returns an optional type. Using Optional.offNullable(thing).ifPresent(...) doesn't help.

This usually means that a bit more code has to change. Something like:

        @Override
        public Optional<CucumberOptionsAnnotationParser.CucumberOptions> getOptions(Class<?> clazz) {
            final CucumberOptions annotation = clazz.getAnnotation(CucumberOptions.class);
            if (annotation == null) {
                return Optional.empty();
            }
            return Optional.of(new CoreCucumberOptions(annotation));
        }

    }

            requireNonNull(optionsProvider).getOptions(classWithOptions)
                    .ifPresent(options -> {
                        addDryRun(options, args);
                        addMonochrome(options, args);
                        addTags(clazz, options, args);
                        addPlugins(options, args);
                        addStrict(options, args);
                        addName(options, args);
                        addSnippets(options, args);
                        addGlue(options, args);
                        addFeatures(options, args);
                        addObjectFactory(options, args);
                    });
        }

[cyocum]

Thanks for the changes. I have not had a chance to look over them yet. I know you had suggested to revert the change that I made to HookDefinition. I have done that locally but I am having serious difficulties with a test that I cannot see why it is failing so I am a bit stuck at the moment.

[coveralls]

Coverage increased (+0.04%) to 85.995% when pulling b03016f on tag_exception_loc into afa3a99 on master.