Wrong getter/setter names for generated for certain corner cases

[janzyka]

Lombok generates non-valid getter names for certain corner cases of Java Bean specification.

Namely fields which have second letter uppercase like xAxis or iMessage should become setxAxis(...) and setiMessage(...) - not very intuitive but it seems to be that way. Lombok on the other hand generates setXAxis(...) and setIMessage(...).

This issue usually doesn't do much harm until you do something like deserialize object annotated via @Data through Jackson. In this case those fields remain unset because no setter with valid name exists for such fields.

There is a blog post explaining this issue nicely.

[nick-yegw]

Yes, I met the same issue when using fasterxml.jackson to generate xml from javabean, which has used the "@DaTa"

And you can find the related doc from oracle about this:
http://download.oracle.com/otndocs/jcp/7224-javabeans-1.01-fr-spec-oth-JSpec/

Please refer to below section:
8.3.1 Simple properties
8.3.2 Boolean properties
8.8 Capitalization of inferred names.

Thus when we extract a property or event name from the middle of an existing Java name, we
normally convert the first character to lower case. However to support the occasional use of all
upper-case names, we check if the first two characters of the name are both upper case and if
so leave it alone. So for example,
“FooBah” becomes “fooBah”
“Z” becomes “z”
“URL” becomes “URL”
We provide a method Introspector.decapitalize which implements this conversion rule.

I think it would be better to fix this bug asap, otherwise, some magic bug would be hard to found.

[askoning]

See #504 , #757 for older discussions on this.

[anddero]

In #757 this was dropped due to Java community being split on the topic. I still believe it is worth the hassle and could be fixed in the next major release.

A better option might be to include a new strict parameter for the @Getter, @Setter and @Data annotations to allow this functionality without any breaking changes (as suggested here), for as long as it is not the default behavior.

The current workaround is to write your own getters and setters for fields related to this issue, to conform to the JavaBeans API Specification.

[randakar]

There is a simpler workaround for Jackson: The new '@Jacksonize' annotation that is in the edge release, possibly with '@JsonProperty' / '@JsonGetter' on any fields still having issues.

As for the whole bean naming issue: The whole point of it is that there is no single spec everyone adheres to. There is no such thing as 'strict' when doing that breaks half of the world.

[Rawi01]

@anddero You can also add @Setter(AccessLevel.NONE) @Getter(AccessLevel.NONE) to the field you want to exclude, that might be easier if there are only one or two fields.

One of the maintainers already mentioned that they will accept a PR that changes this (unfortunately only without a config key to get back the current behaviour...).

[anddero]

It seems, when you have a custom accessor like getxAxis(), Lombok detects this and does not generate an additional getXAxis(). This makes the workarounds much simpler.

@Rawi01 So it is not even necessary to add those annotations to specific fields.

[rzwitserloot]

Duplicate of #2693