Provide common test helpers.

[blackwinter]

As mentioned in #386, it would be nice to be able to reuse the stream test pattern (cf. #339) across modules. Currently metamorph and metafacture-mangling make use of it, plus metafacture-yaml once #400 is accepted. There are more candidates that could potentially benefit from it.

I'd like to propose metafacture-test as a new module, with org.metafacture.test.StreamTestHelpers (or maybe org.metafacture.test.helpers.StreamTest) as a starting point. There's always the possibility to extract more test patterns in the future.

It would be a testImplementation dependency for the aforementioned modules.

Would something like this be acceptable? (See also the brief discussion in #386.)

P.S.: While at it, it might make sense to configure test logging (showStandardStreams = true and exceptionFormat = 'full') globally for all subprojects. Unwanted output should be suppressed or otherwise dealt with.

org.metafacture.flowcontrol.ObjectExceptionCatcherTest > shouldCatchException STANDARD_ERROR
    [Test worker] ERROR org.metafacture.flowcontrol.ObjectExceptionCatcher - 'Exception Message' while processing object: data

org.metafacture.flowcontrol.ObjectThreaderTest > shouldSplitAllObjectsToAllThreadedDownStreamReceivers STANDARD_ERROR
    [Test worker] INFO org.metafacture.flowcontrol.ObjectThreader - Adding thread 1
    [Test worker] INFO org.metafacture.flowcontrol.ObjectThreader - Adding thread 2

org.metafacture.json.JsonDecoderTest > testShouldProcessRecordsInArrayRoot STANDARD_ERROR
    SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".
    SLF4J: Defaulting to no-operation (NOP) logger implementation
    SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details.

org.metafacture.monitoring.StreamLoggerTest > shouldActAsSinkIfNoReceiverIsSet STANDARD_ERROR
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - start record 1
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - start entity entity
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - literal literal=value
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - end entity
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - end record
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - resetStream
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - closeStream

org.metafacture.monitoring.StreamLoggerTest > shouldForwardAllReceivedEvents STANDARD_ERROR
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - start record 1
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - start entity entity
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - literal literal=value
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - end entity
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - end record
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - resetStream
    [Test worker] DEBUG org.metafacture.monitoring.StreamLogger - closeStream

org.metafacture.monitoring.ObjectTimerTest > testShouldHandleImmediateCloseStreamWithNoProcessing STANDARD_ERROR
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Executions: 0; Cumulative duration: 0s; Average duration: 0s
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Time to close stream:  14µs 965ns

org.metafacture.monitoring.ObjectTimerTest > testShouldMeasureExecutionTime STANDARD_ERROR
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 1: 150ms 138µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 2: 20ms 82µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 3: 30ms 130µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 4: 202ms 164µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Executions: 4; Cumulative duration: 402ms 515µs; Average duration: 100ms 629µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Time to close stream:  6µs 170ns

org.metafacture.monitoring.StreamTimerTest > testShouldHandleImmediateCloseStreamWithNoProcessing STANDARD_ERROR
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Executions: 0; Cumulative duration: 0s; Average duration: 0s
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Time to close stream:  1µs 470ns

org.metafacture.monitoring.StreamTimerTest > testShouldMeasureExecutionTime STANDARD_ERROR
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 1: 150ms 242µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 2: 20ms 149µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 3: 30ms 123µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Execution 4: 202ms 170µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Executions: 4; Cumulative duration: 402ms 685µs; Average duration: 100ms 671µs
    [Test worker] INFO org.metafacture.monitoring.TimerBase - Time to close stream:  5µs 111ns

org.metafacture.metamorph.test.validators.StreamValidatorTest > shouldFailIfEndRecordEventIsMissing STANDARD_ERROR
    [Test worker] INFO org.metafacture.metamorph.test.validators.StreamValidator - Event Stream: 1{}

[blackwinter]

metafacture-fix could also benefit from this.

[blackwinter]

Maybe also print Morph/Fix script on failure.

[dr0i]

As these scripts can comprise several thousand lines a snippet would be fine.

[blackwinter]

As these scripts can comprise several thousand lines

Not usually in tests, though, right? Those are (almost?) always small scripts with the minimum number of instructions required for the particular test.

[blackwinter]

BTW: This idea was inspired by the Metafix test helper, only limited to test failures so it wouldn't clutter the regular test output.

[dr0i]

Ah, right, sorry, of course - it's about tests ... so just ignore my comment :)

[blackwinter]

configure test logging (showStandardStreams = true and exceptionFormat = 'full') globally for all subprojects.

Already done in #417.