Map me if you can! Painless bean mappings with MapStruct by Gunnar Morling

The mapping between different data models is a common but unpleasant task for developers, e.g. when propagating data from an internal domain model to the elements of a REST facade, UI layer or external web service.
Writing conversion code by hand is tedious and error-prone, while reflection-based solutions suffer from poor performance and the lack of type-safety. MapStruct takes a fresh look at the issue and generates mapping code at compile time, based on Java interface definitions. The resulting code is type-safe and super-fast, has no dependencies and is easy to understand (and debug, if ever needed). Providing many type conversions out of the box, MapStruct also allows you to customize and amend mappings as needed.
We'll discuss typical mapping problems and their solution with MapStruct. You'll see the generator in action (CLI and Eclipse) and quickly forget about the pain bean mappings used to give you in the past.
Gunnar Morling is a software engineer and open-source enthusiast by heart. He’s leading the Debezium project, a tool for change data capture (CDC). As part of the Hibernate team, he contributes to Hibernate Search, OGM and Validator. Gunnar is the spec lead for Bean Validation 2.0 (JSR 380). In his spare time he leads the MapStruct project. Prior to joining Red Hat, he worked on a wide range of Java EE projects in the logistics and retail industries. He’s based in Hamburg, Germany.

[@yuriyk8002]

That is just beautiful!

[@charluz]

Excellent explanation, thank you.

[@xxxxxxxxx2947]

Excellent Explanation...

[@PatrickCornelien]

This is quite a neat project, especially that it doesn't use reflection at runtime!

[@zanelesibanyoni7951]

amazing example

[@greengrer]

Thank you, it explained a lot

[@cosmicjava]

Nice 🎉

[@ShaliniSharma21]

What about collection such as list, which has fetch type lazy, how to resolve lazy initialisation error How to map collection sources?

[@nadimmansour9792]

I need to map data using java from json file to an ontology created in proetege platform and its in rdf-xml format ? can this help ?

[@peteabc1]

The code generation is nice, but I don't like the @Mapping annotation. It's like unsafe assignment done with annotation and field name as string, so what would happen, if I start to refactor? If only we could pass somethign like Class::variable, that'll be really great for this. Or maybe some kind of DSL, but there's also problem of fields visiblity..

[@GunnarMorling]

If you refactored your mapped properties, you'd get an exception at compile time about referencing a non-existing property from within your @Mapping annotation, i.e. feedback is much faster than with reflection-based solutions. We also provide an IntelliJ plug-in (it's on the market place), which supports refactorings: if you change a mapped bean property, any references in @Mapping annotations will be adjusted as well. That's as good as it gets until true field/property references in annotations are part of the Java language (if that ever happens, who knows).

[@ahmedmk534]

you ca use the IDE's plugins to solve refactoring issues

[@handsome_man69]

Handsome man

[@Thematrixhackyou]

Can i Mapping a Absctract class ¿? can i Mapping a object with contains array of objects ¿?

[@ythalorossy]

Yes, for the second question, you can.

[@ChristopherRockhill]

Ythalo Rossy Saldanha Lira example?

[@user-wq3gw1lg6f]

Why didnt u run the code???