1. d9c913a switch Guice from manually repackaging Guava to depending on Guava. it's still jarjar'd right now, which is causing a ~400k increase in guice-snapshot.jar. next step is to switch to ProGuard to remove the unnecessary code and cut it back down (even further?!). this will let people build from Guice source and depend directly on Guava code without having to worry about hiding internal/util. by sberlin · 13 years ago
  2. cc17f14 significantly improve error reporting for binding a key already bound in a child injector or private module. include all sources in the error msg (since it can be in many sibling private modules or child injectors), including whether or not it as a JIT binding. by sberlin · 14 years ago
  3. 0d8b52f issue 608 - child can rebind a parent binding if it was just-in-time. by sberlin · 14 years ago
  4. 45ca7f6 issue 345 - better error message for private modules. show the source of a child binding instead of just saying its in a child injector. also give a hint that you may have wanted to expose the binding. by sberlin · 14 years ago
  5. b2f1760 issue 436 -- add Injector.getTypeConverters, and expand ConvertedConstantBinding to expose the TypeConverter for it. patch graciously provided by Stuart, thanks! by sberlin · 14 years ago
  6. dfee9fd issue 531 -- don't let c.g.i.internal.util reference outside its package. by sberlin · 14 years ago
  7. b02627d issue 407 -- better error messaging for failed interception. previously exploded with a ComputationException, now uses a ConfigurationException with a good msg. by sberlin · 14 years ago
  8. 74d714c Issue 552: (step 1) move core code to its own subdirectory by mcculls · 14 years ago[Renamed from src/com/google/inject/internal/Errors.java]
  9. 890b9ef Make ProvisionException stack traces from @Provides methods clearer -- don't include all the intermediary RuntimeExceptions & InvocationTargetExceptions. Just include the user's exception (even for checked exceptions). Before this, a stack trace from a ProvisionException looked like: by sberlin · 14 years ago
  10. 9a227be Move internal utility code to separate package. by limpbizkit@gmail.com · 14 years ago
  11. 7bcec88 fixes issue 263 -- introduces ability to "deduplicate" bindings. all BindingImpl subclasses now implement equals/hashcode properly and duplicates can be erased by throwing them into a set (preferably a LinkedHashSet to preserve the order of the elements). no new API methods. by sberlin@gmail.com · 14 years ago
  12. 3748e4a issue 329 - cleanup error message for under-specified assistedinject factory. by sberlin · 14 years ago
  13. b997821 introduces InjectorBuilder.disableCircularProxies, to forcibly prevent guice from creating a proxy to resolve circular dependencies. by sberlin · 14 years ago
  14. 888a264 fixes issue 342 - ability to disable JIT bindings (require explicit bindings). this revision creates a new public facing API of: by sberlin · 15 years ago
  15. b47d75a JSR 330 strict constraints checking: by limpbizkit · 15 years ago
  16. 9b8bdc7 toConstructor SPI support. We can now create the module models for toConstructor bindings. I haven't yet added the code that takes these bindings and uses them when creating an Injector. by limpbizkit · 15 years ago
  17. b68388b Killing 'bad.bad.bad' for release. by limpbizkit · 15 years ago
  18. e89c49e Some low-hanging fruit optimizations. by limpbizkit · 15 years ago
  19. 759662b Tests and fix for issue 340. by limpbizkit · 15 years ago
  20. ee79246 More listener changes: by limpbizkit · 15 years ago
  21. a843a95 Simplifying listeners: by limpbizkit · 15 years ago
  22. 7cef5b0 We now fire InjectionListeners on member injectors. by limpbizkit · 15 years ago
  23. 0ca7f91 It is now possible to inject a MembersInjector<T> anywhere, just like how we already support injecting a Provider<T> and a TypeLiteral<T> anywhere. by limpbizkit · 15 years ago
  24. 97eac0f Injector.getMembersInjector() and Binder.getMembersInjector() APIs both implemented and tested. by limpbizkit · 15 years ago
  25. 516e2ab Error handling. Now if InjectionListeners or InjectableTypeListeners fail, we include those errors in our pretty error report. by limpbizkit · 16 years ago
  26. 53664a7 Regrettably replacing jarjar'd Google Collections with minimal copies of the parts that we use. by limpbizkit · 16 years ago
  27. 03c8d26 AssistedInject Deluxe. by limpbizkit · 16 years ago
  28. b1f42f5 Removing the obsolete PrivateModules extension; PrivateModules are now folded-in to core Guice. by limpbizkit · 16 years ago
  29. fcbdf99 Pretty massive rewrite of PrivateModules so that they're now implemented in core Guice. by limpbizkit · 16 years ago
  30. b0f3c9b Guice now reifies types! If you inject a TypeLiteral<T>, Guice will deduce what T is and inject the proper TypeLiteral for you. by limpbizkit · 16 years ago
  31. 72d11dd Taking advantage of the error message changes in some extensions. by limpbizkit · 16 years ago
  32. 0689806 Hopefully the last of the big exceptions refactorings. I went through all of the places we're adding context to our Errors object and made sure we're never doubling-up -- specifing the same injection point or key multiple times. by limpbizkit · 16 years ago
  33. 490833f Cleaning up exceptions and Javadoc: by limpbizkit · 16 years ago
  34. b8db6a9 Restoring a missing 'at' statement in the Errors message by limpbizkit · 16 years ago
  35. 86cb3bb Refinements to PrivateModules: by limpbizkit · 16 years ago
  36. 5fb9d92 Rewrite of parent injectors. by limpbizkit · 16 years ago
  37. b3a8f0b Last of the planned major changes to the SPI. Aside from Javadoc I consider the SPI to be complete for v2. (with a possible addition of a ProviderMethodBinding feature also) by limpbizkit · 16 years ago
  38. a6e0e78 Moving error sources so they're now reported in the same place for configuration errors and runtime errors. This means that instead of seeing this: by limpbizkit · 16 years ago
  39. 18cb191 Splitting up finding member injectors from finding injection points. This is intended to make it so the new SPI exposes injection points on Module elements without actually having to resolve 'em. by limpbizkit · 16 years ago
  40. a98bc7a Cleaning up some internals for the SPI. by limpbizkit · 16 years ago
  41. 76a2b9a Changing InjectionPoint.member to be swapped for a serializable instance lazily rather than eagerly. This is what Bob had originally requested, and only now do I learn my lesson (from Sam's memory leak). Bob=smart. by limpbizkit · 16 years ago
  42. 4ce9cfa Changing the loading of StackTraceElements to be as lazy as possible. Sam Berlin was running into memory issues when these were loaded eagerly. by limpbizkit · 16 years ago
  43. d6becac Full test coverage for errors... by limpbizkit · 16 years ago
  44. 5789ef4 A few more errors tests. There's only about five left till we have full test coverage for our errors code. by limpbizkit · 16 years ago
  45. 918dff8 Moved Errors (an implementation detail class) out of InjectionPoint (a public API). by limpbizkit · 16 years ago
  46. e731d19 Tweaking error handling. by limpbizkit · 16 years ago
  47. d37c58b Disallowing private, no arg constructors that don't have @Inject on 'em. This is issue #120. The workaround is to increase the visibility of the constructor (package private still works) or to add @Inject on the constructed type. by limpbizkit · 16 years ago
  48. 2c2c610 More cleanup of Errors -- including sources as a factory method rather than push/pop methods (that require try/finally clauses) by limpbizkit · 16 years ago
  49. 6663d02 Axing SourceProviders and SourceProvider in favour of Binder.withSource and Binder.skipSources by limpbizkit · 16 years ago
  50. 9532e62 Eliminated static from SourceProviders. The new mechanism to specify the source of a binding is like this: by limpbizkit · 16 years ago
  51. 185d2a2 Added some new error messages: by limpbizkit · 16 years ago
  52. 76f1457 Fixing some places where we were using SourceProviders when we should have been using other sources by limpbizkit · 16 years ago
  53. 163c48a Cleanup after the big errors refactoring. by limpbizkit · 16 years ago
  54. c5c488e TypeConversion errors cleanup by limpbizkit · 16 years ago
  55. 564053f Massive refactoring to exception handling. I'm trying to simplify things, but they are currently a little bit more complicated. I'll do another round shortly. by limpbizkit · 16 years ago