| Age | Commit message (Collapse) | Author |
|---|
|
|
|
process fixed a lot of type annoyance by adding more generics.
Also changed coding style from for/while/if/switch/catch/do ( expr ) {} to for (expr) {}, hence the changes _everywhere_.
|
|
installs, is a javac apt processor, and an agent, all at once.
|
|
|
|
branch...
|
|
causes eclipse to break down; now it should just log to standard error and at least let eclipse continue running, sans lombok, if the logging to the eclipse log of the failure also fails.
|
|
which we can use to patch eclipse in specific places to ignore generated nodes.
|
|
|
|
problem, found the place in the source where the problem occurred, and even wrote a patch.
|
|
it does rather a lot of stack introspection which may slow things down far too much.
|
|
|
|
|
|
in combination with @Data). Addresses issue #37.
|
|
support explicitly listing the fields to use, via the new 'of' parameter.
We've also added any fields that start with $ to the default excludes list. Lombok itself can generate these fields ($lock of @Synchronized, for example), and in general they probably should count as effectively not part of the class.
|
|
|
|
when using @SneakyThrows (syntax colouring goes away, first character in the file is italic).
|
|
annotation @NotNull/@NonNull/@Nullable that is copied over by @Getter should no longer be causing the David Lynch bug.
|
|
generating. This really does seem to fix the David Lynch bug (#41).
|
|
the generated annotations seems to fix the David Lynch bug (issue #41).
Saving this now before further fiddling breaks things again.
|
|
ond depending on your eclipse version, a long stack trace.
The problem boiled down to the JSP compiler used by the help system also being instrumented with lombok, but that's not exactly the environment lombok was expecting. Fixed by simply disabling lombok when the environments don't match what we expect. In the process, the instrumentation has been made a little more robust; multiple separate OSGi modules can all be instrumented now, instead of the first one winning.
|
|
|
|
|
|
@NonNull/@Nullable and getter/setter/constructor generation) to 0, as eclipse mysteriously fails for annotations copied WITH source positions, but only on methods (which happens for @Getter).
|
|
Conflicts:
src/lombok/eclipse/handlers/HandleData.java
src/lombok/eclipse/handlers/HandleEqualsAndHashCode.java
src/lombok/eclipse/handlers/HandleSetter.java
src/lombok/javac/handlers/HandleData.java
src/lombok/javac/handlers/HandleEqualsAndHashCode.java
src/lombok/javac/handlers/HandleSetter.java
|
|
|
|
|
|
legal reasons for using it, so, changed it: explicitly setting 'callSuper=false' removes the warning. You only get the warning
if callSuper is false because that's the default.
Fixes issue #13
|
|
annotation
The constructor will test for null-values
The constructor and static constructor will copy the NonNull annotations from the fields
|
|
|
|
the parameter
@Getter will copy them to the getter method
Added @NonNull to lombok to support null-checks in the setter
|
|
|
|
'plus' nodes (e.g. concatenating the infix ", " and a field name literal such as "width=" into ", width=".
Also removed the [] brackets from the supercall, as, if you're chaining to another lombok-generated toString, those are superfluous - lombok's toString includes parentheses already.
|
|
new annotation, @EqualsAndHashCode.
Addresses issue #8
|
|
|
|
has now moved from HandleData to the new HandleToString.
|
|
entry now no longer causes ArrayIndexOutOfBoundsException, the setWarning method on a single item in an array initializer on eclipse
now generates the warning on just that node (like with errors), and the API of AnnotationValues has been updated to support setting errors/warning on any node.
|
|
|
|
hitting 'find callers' on a @Data annotation should find callers of the (static) constructor.
Right now it'll find callers to the *static* constructor ONLY. Letting it find callers of the public constructor if there is no static constructor just doesn't work.
|
|
|
|
If boolean fields already start with a typical getter prefix (is, has, or get), lombok's @Getter will no longer generate its own prefix as well, so a field named 'hasFoo' will result in a getter named 'hasFoo()', not 'isHasFoo()'.
Also, if any likely getter name already exists for a boolean, a getter will not be generated. Thus, if your field is called 'hasFoo', and you already have a method named 'isFoo', then @Getter will not generate anything (and warn, unless the getter is being generated due to @Data).
This last mechanism works by taking the field name *AND* any other likely base names (defined by the field name being named as prefix+baseName, with prefix being is/has/get), and then prefixing all the likely fieldnames with is/has/get, and checking if any method with that name exists.
Of course, this means weird things are going to happen if you have 2 fields named 'isFoo' and 'hasFoo', but then, you'd be a real idiot if you did that.
|
|
|
|
|
|
the class would generate errors regarding IllegalArgumentException in setSourcePosition in ASTNode.
|
|
generating it" warnings
when the getter/setter already there was in fact generated by lombok, and fixed a bug
in eclipse where a boolean array's getter method would be called isFoo() instead of getFoo().
|
|
Also fixed a bug in javac's toString() generation for the @Data constructor. It did
not include the transient fields.
|
|
upon anymore when reporting errors.
They were logging 4 or more identical warnings per problem before this change.
|
|
|
|
package, of course),
and fixed a showstopper bug in the installer that would add -javaagent:lombok.jar to
eclipse.ini, which is wrong of course; it needs to be lombok.eclipse.agent.jar.
|
|
|
|
especially the docs
on the lombok annotations in the lombok package need far more massaging.
Also added a feature to HandleSynchronized to not auto-generate the locker fields if
a specific name is provided (because, imagine you typoed those. You'd never find it!)
|
