1.4.0

• Allow to customize metric parameters for example to use new lock-free reservoirs
• When logging mapping errors dump erroneous element first and only then the context
  • The context could contains lots of other elements that were already
    mapped or still to be mapped. Dumping context first makes it harder to
    digest error message. Showing erroneous element first and then the
    context makes it easier to understand how to fix the problem
• Allow to avoid duplicating information when logging mapping errors
  • Combining ElementErrorLoggerDecorator and SingleLineMesasgeFormatter leads to duplication of erroneous element and context when there is a mapping error.
    To avoid such duplication combine SingleLineMesasgeFormatter with BriefElementErrorLoggerDecorator instead.
    master
• Add a test that demonstrates typical usage example (closes #5)
• Switch back to non-indy version of Groovy to stick with default developments in latest Groovy
• Upgrade groovy to 3.0.8
• Upgrade log4j to 2.14.1
• Upgrade metrics to 4.2.0

1.3.0

Compile into invokedynamic rather than call sites in a hope to improve performance of classes without @CompileStatic. Requires Java 1.7+

1.2.0

feat: Add support for ignored fields

Allow to show that certain input or output fields are known but are explicitly ignored.
This can be done using optional field (why would mandatory field be ignored?).
If only getter is present then it means input field is ignored.
If only setter is present then it means output field is ignored.
Those getters and setters would not be actually called.
They're just for documentation purposes.

This is useful when you want to explicitly document what you do with
each input or output field and want one-to-one mapping
between documentation and code.

1.1.0

Added support for reporting time it takes to map individual fields.

This can be done by decorating FieldMapper with MeteringFieldMapperDecorator which is to be supplied with MetricNameCalculator. MetricNameCalculator allows to define how much metrics will be reported. The default implementation is PrefixedFieldIdMetricNameCalculator. It simply reports each field as a separate metric. However, by supplying own implementation of MetricNameCalculator you can have additional axes. For example, if you have separate mapping tables for different event types then you can include event type in metric name. This will allow to track how much time is needed for each field in each of mapping tables.

1.0.0

Initial open source release of translation library.