aboutsummaryrefslogtreecommitdiff
path: root/src/core/lombok/javac/JavacTransformer.java
AgeCommit message (Collapse)Author
2020-07-09Remove unnecessary list copyRawi01
2020-07-09Replace JavacAst list with single object iteration to save some memoryRawi01
2019-01-22[fixes #2019] Lombok now properly deals with `@NonNull` specifically on the ↵Reinier Zwitserloot
‘type use’ of a parameter (and, in case of arrays, on the outermost dimension which is actually the first one listed. Weird corner case of the JLS).
2019-01-08[#1033] steps on the way to issue 1033: You can add cleanup tasks which are ↵Reinier Zwitserloot
deferred (during the javac run) until the end. This already fixes the exotic-to-the-point-of-nonexistent bug where setter and wither compete to steal the `@param` off of the field’s javadoc. Next are to fix builder and setter/wither competing whilst bringing javadocs to `@Builder`. Then for various other conflicts, we should defer removal of lombok imports and annotations until the end too.
2018-07-25Added key ‘dangerousconfig.lombok.disable’.Reinier Zwitserloot
2017-01-19Lombok will now also fix the typemirror info when generating bean-related ↵Reinier Zwitserloot
methods/constructors, to allow other annotation processors to see these generated methods/constructors too.
2012-08-10Added an optimization to javac: Don't force new resolution rounds if none of ↵Reinier Zwitserloot
the handlers in the next priority level need it.
2012-08-10* Added priorities to handlers, along with implementation of the priority ↵Reinier Zwitserloot
system for javac and ecj. * @Value now makes the class itself final by default.
2011-11-21Made an API for creating a (oracle javac) JavaCompiler which tracks comments ↵Reinier Zwitserloot
on a per Compilation Unit basis. The old way involved making reflective calls and detecting whether you need the 1.6 or the 1.7 variant to do this. These shenanigans are now hidden behind a nice API (lombok.javac.CommentCatcher).
2011-11-07Changed copyright noticesRoel Spilker
2011-05-30tracking if an annotation has been handled or not is now no longer doneReinier Zwitserloot
via the LombokAST object. Instead its tracked more directly in an attempt to avoid having to write all handlers as idempotent, and just in case issue #164 is a race condition (the handled-or-not is a synchronized CAS check). This does break API for other plugins, but the fix is trivial: Just make your 'handle' method return void. That 'we won't call you again' business in the decks never quite worked right anyway. Also, you might want to call Javac.(recursive)setHandledBy when you generate nodes, now.
2010-11-29lombok.val import now gets removed. Only adding lombok.val flags the ↵Roel Spilker
compilation unit as changed.
2010-11-10Delombok has been fixed to work more like a true javac run now. As a result, ↵Reinier Zwitserloot
its now compatible with resolution again (i.e. resolution based transformers are applied correctly when delomboking).
2010-11-10'val' now fully works on javac, even when referring to lombok-generated ↵Reinier Zwitserloot
code, by (ab)using the annotation processor's round system. This breaks delombok though. That'll have to be fixed next.
2010-11-10Undone something that'll never work anywayReinier Zwitserloot
2010-11-09Experiment to try and make java 'self referential' by deleting the Enter and ↵Reinier Zwitserloot
MemberEnter state. So far it doesn't work yet.
2009-11-27Thorough work on inserting comments in the proper place for delombok; should ↵Roel Spilker
now work fine with GWT native methods! NON-NLS-1 is still theoretically problematic, but that'll be a fix for another day. Also added ability to recognize that nothing has changed, which will copy the original file instead of reparsing it.
2009-11-27delombok now calls lombok. wahey!Reinier Zwitserloot