| Age | Commit message (Collapse) | Author |
|---|
|
(exceptions from the cleanup call WILL mask exceptions from the body - this isn't intended, but it's just not possible to fix this without java 7 features or requiring a rewrite of the class file data.
Tried tactics, and why they won't work:
- Replace every 'return', 'break', and 'continue' statement (for the latter 2, only if they break/continue out of the try block) with a block that first sets a uniquely named flag before doing the operation.
Then, check that flag in the finally block to see if the cleanup call should be guarded by a try/catchThrowable. This doesn't work, because its not possible to instrument the 4th way out of a try block without throwing an exception: Just letting it run its course. Tossing a "#flag = true;" at the end may cause a compile time error if the code is not reachable, but figuring that out requires resolution and quite a bit of analysis.
- Put catch blocks in for all relevant exceptions (RuntimeException, Error, all exceptions declared as thrown by the method, and all types of exceptions of the catch blocks of encapsulating try blocks.
This doesn't work, partly because it'll fail for sneakily thrown exceptions, but mostly because you can't just catch an exception listed in the 'throws' clause of the method body; catching an exception
that no statement in the try block can throw is a compile time error, but it is perfectly allright to declare these as 'thrown'.
- Put in a blanket catch Throwable to set the flag. Two problems with this:
First, sneaky throw can't be done. Thread.stop invokes a security manager and triggers a warning, Calling a sneakyThrow method creates a runtime dependency on lombok, constructing a sneakyThrow in-class creates visible methods or at least visible class files, and creating a new class via Class.loadClass would be very slow without caching - which gets you the same issues.
Secondly, this would mean that any statements in the try body that throw an exception aren't flagged to the user as needing to be handled.
The Cleanup annotation now also calls the cleanup method for you, and will call it at the END of the current scope. The following plans have been tried and abandoned:
- Cleanup right after the final mention. This doesn't work, because the final mention may not be the final use-place. Example:
@Cleanup InputStream in = new FileInputStream(someFile);
InputStreamReader reader = new InputStreamReader(in);
reader.read(); //oops - in is already closed by now.
- Require an explicit var.cleanup() call and consider that the cue to close off the try block.
This doesn't work either, because now variables set in between the @Cleanup declaration and the var.cleanup() call become invisible to following statements. Example:
@Cleanup InputStream in = new FileInputStream(someFile);
int i = in.read();
in.close();
System.out.println(i); //fail - i is inside the generated try block but this isn't, so 'i' is not visible from here.
By running to the end of visible scope, all these problems are avoided. This does remove the flexibility of declaring where you want a close call to be executed, but there are two mitigating factors available:
1) Create an explicit scope block. You can just stick { code } in any place where you can legally write a statement, in java. This is relatively unknown, so I expect most users will go for:
2) Just call close explicitly. I've yet to see a cleanup method which isn't idempotent anyway (calling it multiple times is no different than calling it once).
During the course of investigating these options, the AST code has been extended to support live replacement of any child node, including updating the actual underlying system AST as well as our own.
Unfortunately, this code has NOT been tested. It was rather a lot of work so I'm leaving it in, and at least for eclipse it even seemed to work.
|
|
should run in java 1.5, so that an eclipse started on a 1.5 JVM will still run lombok.
|
|
|
|
is not neccessarily the proper target anyway. Rolled our own DefaultProblem subclass for problem reporting.
|
|
Eclipse will now also hold off on running @PrintAST handlers until the very end. Simple generators such as @Getter didn't need this, because PrintAST's handler will hold off until eclipse does a full parse,
but when changing the innards of methods, you would likely not see what you did. Fixed that.
Also, PrintAST has an option to, instead of diving into the ASTNodes of bodies (methods, initializers, etc), to just render the java code, to see if the AST creation/rewriting you've been doing looks like the java code you intended.
|
|
were set to the value of the last annotation method. e.g in:
@Foo(bar=10), ALL methods in the Foo annotation were presumed to be listed, and set to 10. This was obviously causing problems. Fixed it.
|
|
|
|
There's one serious problem though: The cleanup routine modifies the eclipse internal AST, but doesn't update our bi-directional AST. Thus, or example, having a @Cleanup annotation inside the scope
of another @Cleanup fails, because the application of the second one climbs up to the wrong block level (the original block level instead of newly built try block).
|
|
|
|
createHashCode method when there are 0 fields in the type (it would generate a local variable never used warning!)
|
|
exist, and the staticConstructor is now also completed. Left: toString, hashCode, equals.
|
|
and addition. The rule is now: children traversal traverses through the tree mostly as it was when it started.
|
|
throw eclipse errors if you had 0 non-static fields.
|
|
deepToString, and added @Override in case people have warnings for missing @Override annotations on.
|
|
|
|
without any raw generics warnings - it is effectively done.
|
|
directly to other nodes (e.g. from a FieldDeclaration's type to a method argument) is NOT a good idea, as this screws up when
the TypeReference object represents a generic type (like 'T') - each instance of a generic type has a different resolution, but 1 TypeReference object can only hold 1 resolution.
Thus, a copyType() method has been written, and the Handle* classes have been updated to use it.
Also, generateEquals() is half-finished in HandleData.
|
|
** DO NOT REUSE TYPEREFERENCE OBJECTS **
because that makes the binding process go pearshaped - after hte first run, that TypeReference object's binding parameter is set, and as its set, the resolver won't bother re-resolving it.
However, each parse run starts with new scope objects, and any 2 bindings created by different scopes aren't equal to each other. urrrrrrgh!
Fortunately, a lot of code that 'fixed' methods by adding bindings and scope have all been removed, as the parser patch point is well before these bindings are created. Thus:
** NEVER CREATE YOUR OWN BINDINGS AND SCOPE OBJECTS **
because if it comes down to that, you're doing it entirely wrong. That's eclipse's job. We're patching where we are so you don't have to do this.
|
|
missing a 0!
|
|
static constructor not so much.
|
|
|
|
access level for @Getter and @Setter have now just been hardcoded in GetterHandler and SetterHandler.
Added ability to look up the Node object for any given AST object on Node itself, as you don't usually have the AST object.
Added toString() method generating to @Data, and this required some fancy footwork in finding if we've already generated methods, and editing a generated method to fill in binding and type resolutions. HandleGetter and HandleSetter have been updated to use these features.
Exceptions caused by lombok handlers show up in the eclipse error log, but now, if they are related to a CompilationUnit, also as a problem (error) on the CUD - those error log entries are easy to miss!
Our ASTs can now be appended to. When you generate a new AST node, you should add it to the AST, obviously. Getter/Setter have been updated to use this.
|
|
and setters only,
not yet a constructor, toString, hashCode, or equals.
HandleGetter and HandleSetter have been updated to handle static (theoretic; you can't put annotations on static fields normally).
You can now make AnnotationValue objects using just an annotationNode and a target type, as well as check if a given annotationNode is likely to represent a target annotation type. This is in Javac and Eclipse classes.
HandleGetter and HandleSetter can now be asked to make a getter/setter, and will grab access level off of a Getter/Setter annotation, if present.
|
|
HandleGetter a little mostly to stuff common code into PKG.
|
|
@PrintAST annotation to let you supply an optional filename. Useful particularly for IDEs, which don't usually have a viewable console.
Also renamed the printers to just 'Printer', as they are already inner classes of a specifically named type (JavacASTVisitor & co).
|
|
the annotation or not (previously, the presumption was they always handled the annotation).
This is very useful for PrintAST on eclipse, because before this change, you'd never see method contents (as the initial dietParse would come first). Now Eclipse PrintASTHandler
will skip any non-full runs, and only print non-diet. It then returns true only if it printed.
|
|
more verbose), and bumped the version number in honour of quite a bit of redesign these past few commits.
|
|
XASTPrinters in each ASTVisitor interface.
|
|
you can start your traversal at any point, not just from the top.
Also a bugfix for endVisitStatement which passed the wrong node, and method arguments in Javac are no longer misfiled as local declarations.
|
|
|
|
method argument or not (by instanceof checking if it's an Argument) is faulty, as e.g. the argument to a catch block is also an Argument object. Rewritten the visitChild method to be based on a switch on the Node's getKind(), just like JavacAST. This even looks nicer.
|
|
A) many things in lombok.eclipse moved to lombok.core to enable reuse with lombok.javac.
B) lombok.javac works now similarly to eclipse's model: We first make big ASTs that are bidirectionally traversable, then we walk through that for annotations.
C) Instead of getting an annotation instance, you now get an object that is more flexible and can e.g. give you class values in an enum as a string instead of a Class object, which may fail if that class isn't on the classpath of lombok.
D) sources to the internal sun classes for javac added to /contrib.
|
|
to be implemented:
1. The visit mode of a lombok handler (does not trigger off of annotations, instead sees every field, method, type, and statement), needs to be coded,
2. triggering off of annotations via APT's annotation handling system skips method-local classes. We'll need to recode this via an AST visitor like we need for issue #1
Other than that, triggering off of annotations works swimmingly!
|
|
|
|
annotation, and not the field/type/local/method it goes with, all the way, so that you can easily generate a warning on an annotation in a handler.
|
|
call from the Field/Type/Method/Local to the Annotation, so that you can interact with its handled flag.
|
|
updated the visitor to call a separate visitAnnotationOnX method for annotated stuff. This way, 'handled' can be set per annotation.
Also fixed a bug in AST generation that caused StackOverflowErrors on most source files, and did some cosmetic renaming of parameters.
|
|
for lombok.
This way something like @AutoClose on a local var declaration can walk up one node, find
all mentions of the variable, and add a close call right after the last mention.
|
|
to contain stuff that is useful for any lombok implementation (be it e.g. javac via apt or eclipse via agent), but not annotations and other classes that are for 'end users'.
|
|
in it.
|
|
turning them into eclipse-wide errors in the worst case, but usually in an error in the problems dialog.
|
|
(generified code in HandlerLibrary for unintelligible annotation param values), and more severe general errors for eclipse's error log.
Also unrolled the foreach loop on ServiceLoader, because any given .next() call can throw a ServiceLoaderError, which we now handle somewhat more nicely.
|
|
annotation arguments for other places!
Because this stuff works so well now, I bumped the version number as well.
|
|
will be skipped.
Now they are skipped.
|
|
|
|
parsed. w00t!
|
|
working due to
circular reference from the EclipseAST back to the CUD.
Now, patched a field into CompilationUnitDeclaration and using that, which works much better
together with the garbage collector.
|
|
HandleSetter, HandleAutoClose, etc.
|
|
occurs in the two most sane places:
- After the parser is done building a first rendition of the AST. (Usually lightweight and missing method bodies etc)
- After the parser is done taking such a lightweight AST and filling in the gaps.
Lombok then builts its own bidirectional and somewhat saner AST out of this, and hands this saner AST off for treatment. Things in the AST can be marked as 'handled'.
This seems to work swimmingly and should allow us to easily identify the annotations that are for us, and work our magic, no matter where they appear or on what, including
stuff inside method bodies.
|
|
- Split off the actual agent work into a separate src package in preparation for creating separate jars. Involved a lot of renaming
- Renamed TransformCompilationUnitDeclaration to TransformEclipseAST, as this class will also be transforming e.g. MethodDeclaration objects.
- Expanded the patching to also patch in transform calls when the parser fills in the Statement array for existing constructors, methods, and initializers.
- Redesigned the ClassLoaderWorkaround class quite a bit.
- Positioning should not work correctly ('jump to method' should jump to the getter annotation).
(Apparently, Clinit objects are always fully parsed in the original run, so no need to patch anything there).
|
