aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Made Erroneous a pre-complete flag.Martin Odersky2013-04-051-1/+2
|
* Made & , | NoType-aware.Martin Odersky2013-04-054-10/+9
| | | | For the pruposes of & and |, NoType is now treated as top type, above Any.
* More tests in worksheets.Martin Odersky2013-04-057-18/+99
|
* Made module class names end in '$'.Martin Odersky2013-04-055-14/+35
| | | | | | If module classes have the same names as their companion classes there is the latent trap that TypeRefs refer to one or the other. To avoid cross-talk, module classes need to have a different name. Also, some fixes to baseType, isSubClass in SymDenotations.
* Fix to array erasure.Martin Odersky2013-04-051-2/+3
| | | | Now treats arrays of value classes correctly.
* Fix to flag translation.Martin Odersky2013-04-051-2/+2
| | | | Now ensures that only termFlags/typeFlags are produced.
* Added new classSymbol method for Types.Martin Odersky2013-04-051-1/+26
|
* Ensures ScalaValueClasses are properly initialized.Martin Odersky2013-04-051-2/+12
|
* Refactored test frameworkMartin Odersky2013-04-0510-30/+82
| | | | Moves worksheets to test. Factored out class DottyTest and made it usable from worksheets. Added worksheet test for signatures.
* Added some tracing to asSeenFrom.Martin Odersky2013-04-051-4/+5
|
* Fixed hashcode computation.Martin Odersky2013-04-051-2/+2
| | | | Embarrassiing: Used == instead of = before.
* Adding debug code to disambiguation.Martin Odersky2013-04-051-2/+11
|
* Small change to disjointAsSeenFrom.Martin Odersky2013-04-051-1/+1
| | | | Not sure this matters, but seems to be prudent not ot force signature here.
* Fix to pareamSignatire for arrays.Martin Odersky2013-04-051-1/+1
|
* Fix to superId.Martin Odersky2013-04-051-5/+5
| | | | Big blooper: superId was always -1.
* Fixes to baseType; now respects linearization order.Martin Odersky2013-04-051-22/+14
|
* Consolidated tracing under -YdebugTrace.Martin Odersky2013-04-0410-14/+33
|
* Fxing cyclic references during findMemberMartin Odersky2013-04-042-37/+38
| | | | | | 1) Classes and alias type now take precedence over abstract types. 2) When computing members, parents are visited right to left, which corresponds to existing taking precedence over new in mergeDenot.
* Trying to get at the root of cyclic reference problems.Martin Odersky2013-04-045-28/+35
| | | | Part 1: Enabling tracing and avoiding subtype computations in margeDenot.
* Improvements in tracing and some Type fixes.Martin Odersky2013-04-039-75/+103
|
* Getting rid of forceProperType in UnpicklersMartin Odersky2013-03-261-14/+4
|
* Some more fixes of classfile reading.Martin Odersky2013-03-265-10/+12
|
* More fixes to classfile reading.Martin Odersky2013-03-259-110/+172
| | | | Can now read entire contents of scala distribution.
* More fixes for classfile reading.Martin Odersky2013-03-2510-27/+77
| | | | In particular: Survive malformed type errors. Make trees cloneable. and others more.
* More fixes to classfile loading.Martin Odersky2013-03-246-4/+36
|
* Fixes fingerprint problems.Martin Odersky2013-03-2410-46/+51
| | | | Also reducing debug output. Can now read and display all classes and objects in scala.collection.
* Attempt to fix fingerprint problem (disabled for now)Martin Odersky2013-03-236-24/+52
|
* Some more fixes for classfile loadingMartin Odersky2013-03-233-5/+14
|
* More fixes to classfile reading.Martin Odersky2013-03-238-41/+80
| | | | Most important one: Getting equality of NamedTypes right.
* More fixes for classfile reading.Martin Odersky2013-03-227-53/+97
| | | | Including a refactoring of symbol loaders and unpickler traits.
* More fixes to classfile reading.Martin Odersky2013-03-2110-52/+113
| | | | Can now read all classes in scala.collection.generic. Some failures remain for their companion objects.
* Pretty-printing improvements.Martin Odersky2013-03-2017-253/+315
|
* First draft of pretty printing abstractionsMartin Odersky2013-03-201-0/+123
|
* Fixing problem with reading typeParams correctly in UnpicklerMartin Odersky2013-03-193-8/+30
|
* Adding missing methods for type equalityMartin Odersky2013-03-191-9/+46
|
* Some improvements to printingMartin Odersky2013-03-192-6/+10
|
* Excluding empty trees from type mapping tree transformsMartin Odersky2013-03-191-1/+3
|
* More changes for higher-kinded types emulation.Martin Odersky2013-03-188-38/+123
|
* Cleanup of higher-kinded types treatmentMartin Odersky2013-03-183-20/+38
|
* Added emulation of higher-kinded types.Martin Odersky2013-03-1712-21/+136
| | | | Also improvements to toString.
* Various fixes to make classfile loading work.Martin Odersky2013-03-1512-95/+195
| | | | Can now read and display info of verious classfiles as listed in the showClass test. Great end of the week!
* Defining standard dotty classes used in Definitions.Martin Odersky2013-03-147-4/+34
|
* Improvements to stub handling.Martin Odersky2013-03-145-47/+51
|
* Various fixes that make loadDef largely work.Martin Odersky2013-03-1414-70/+170
| | | | Only problem is that there are stubs for AnyRef generated.
* Various fixes to get past Definitions#init.Martin Odersky2013-03-1319-106/+501
|
* Revising of Denotation#existsMartin Odersky2013-03-125-27/+19
| | | | | | exists now is only false for NoDenotation. A denotation with NoType as info counts as existing, but it is not accessible from any prefix. Also, renamed binary exists, forall or existsPart, forallParts to avoid ambiguities.
* Checking abstractness of types by their flags.Martin Odersky2013-03-124-48/+22
| | | | Previously we looked at the info, but this forces too much. As a consequence we now systematically prefer concrete over abstract when computing & denotations. This could have the strange(?) effect that the symbol of a joint denotation is a class, yet its info is a TypeBounds value.
* Split filterAsSeenFrom and generalized flag handling.Martin Odersky2013-03-1212-78/+81
| | | | | | | | 1. filterAsSeenFrom has been split again into its constituents "filterExcluded" and "asSeenFrom", and care was taken not to force the info unless we have to. The accessible check is no longer done when collecting members, because it would have forced the symbol through requesting privateWithin. 2. SymDenotation#is is tweaked to no longer force the denotation if the flags are in "FromStartFlags", i.e. set upon symbol creation. We can then eliminate special cases isModuleXXX, isPackageXXX. 3. Other tweaks mostly having to do with weakening sym.exists checks to avoid CyclicReference errros.
* Fixed too hasty commit that caused build errors.Martin Odersky2013-03-112-4/+7
|
* First steps to make compiler start work.Martin Odersky2013-03-1114-81/+170
| | | | Still fighting with CyclicReference errors.
generated by cgit v1.2.3 (git 2.39.1) at 2024-09-22 16:42:05 +0000