aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
| * | Simplify methods implemented in #436Dmitry Petrashko2015-04-023-18/+15
| | |
| * | Revert "Workaround #440 in FirstTransform."Dmitry Petrashko2015-04-021-2/+2
| | | | | | | | | | | | This reverts commit b653007ed0f30298b44dcc67bf032c1d2d58bcaf.
| * | Both module and class being unpickled need to register links.Dmitry Petrashko2015-03-301-4/+10
| | |
| * | Fix #443, set moduleClass of class being lazily unpickled.Dmitry Petrashko2015-03-301-1/+5
| | |
| * | Allow to enter private symbols into Frozen scopes.Dmitry Petrashko2015-03-301-1/+1
| | | | | | | | | | | | As private symbols aren't inherited, this does not break caching.
| * | companionModule needs to return ModuleVal for Module.Dmitry Petrashko2015-03-301-6/+8
| | |
| * | Do not synthesizeCompanionMethod twice, and do not rewrite the existing symbol.Dmitry Petrashko2015-03-301-5/+11
| | |
| * | Fix #442.Dmitry Petrashko2015-03-301-2/+6
| | | | | | | | | | | | | | | | | | | | | There's a non-standard interaction between explicitOuter and Pattern matcher, as patmat can request outer symbols to be available earlier. Note that this fix makes code between pattern matcher & explicitOuter non-Ycheck-able, as patmat adds reference to future symbol.
| * | Fix #440: entering symbol into scope also enters it into future scopes.Dmitry Petrashko2015-03-302-3/+9
| | |
| * | Workaround #440 in FirstTransform.Dmitry Petrashko2015-03-301-2/+2
| | |
| * | Fix companion_class_method nameDmitry Petrashko2015-03-301-1/+1
| | |
| * | Use methods to find companion modulesDmitry Petrashko2015-03-282-16/+12
| | |
| * | Make companion-module links in ClassfileParserDmitry Petrashko2015-03-281-2/+8
| | |
| * | Make companion-module links in UnPicklerDmitry Petrashko2015-03-281-1/+6
| | |
| * | Guard against absent symbols in synthesizeCompanionMethod.Dmitry Petrashko2015-03-281-1/+1
| | |
| * | Remove code duplication between Namer, ClassfileParser and UnPicklerDmitry Petrashko2015-03-265-36/+21
| | |
| * | Fix error message in typerDmitry Petrashko2015-03-261-1/+1
| | |
| * | #435 Fix conflict between package object and case class with same nameDmitry Petrashko2015-03-263-3/+5
| | |
| * | Add late companion symbols in firstTransformDmitry Petrashko2015-03-261-2/+14
| | |
| * | Add companion link symbols early only if companion actually existsDmitry Petrashko2015-03-262-17/+48
| | | | | | | | | | | | Otherwise we'll trigger early creation of companions that could shadow something.
| * | Fix installAfter for a Denotation List of single denotationDmitry Petrashko2015-03-261-1/+5
| | |
| * | Add companion class methods for files read from scala & javaDmitry Petrashko2015-03-252-0/+20
| | |
| * | #353 use methods to find companion classDmitry Petrashko2015-03-254-14/+49
| | |
* | | Merge pull request #452 from ↵Dmitry Petrashko2015-04-021-1/+1
|\ \ \ | | | | | | | | | | | | | | | | dotty-staging/fixTTM-structural-sharing-between-classes Fix TreeTypeMap and mapSymbols to create a copy of decls for duplicated classes.
| * | | Fix TreeTypeMap and mapSymbols to create a copy of decls for duplicated classes.Dmitry Petrashko2015-04-021-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Without this fix the duplicated classes and the original ones share the same reference to a scope, instead of having each a separate one. @alexsikia this should fix your problems
* | | | Merge pull request #449 from dotty-staging/fix-context-in-forwardersDmitry Petrashko2015-04-012-7/+15
|\ \ \ \ | | | | | | | | | | Fixes that make bridges smaller.
| * | | | Fix refPurity: was using wrong function to detect if term is stable.Dmitry Petrashko2015-04-011-1/+1
| | | | |
| * | | | Fix wrong owning context during adaptation inside forwarders.Dmitry Petrashko2015-04-011-6/+14
| |/ / /
* | | | Merge pull request #448 from smarter/fix/bridge-emissionDmitry Petrashko2015-04-011-3/+3
|\ \ \ \ | | | | | | | | | | Erasure: actually emit bridges
| * | | | Erasure: actually emit bridgesGuillaume Martres2015-03-311-3/+3
| |/ / /
* / / / Bringing tailrec back to group: cannot trigger failure.Dmitry Petrashko2015-03-301-4/+2
|/ / /
* | | Merge pull request #432 from dotty-staging/patmat-optionsDmitry Petrashko2015-03-301-1/+7
|\ \ \ | | | | | | | | Fix #431: Options make 2 steps of decoding instead of one
| * | | Fix #431: Options make 2 steps of decoding instead of oneDmitry Petrashko2015-03-251-1/+7
| | | |
* | | | Merge pull request #441 from smarter/fix/applyOverloaded-non-initodersky2015-03-281-1/+7
|\ \ \ \ | | | | | | | | | | Make tpd#applyOverloaded work with methods other than <init>
| * | | | Make tpd#applyOverloaded work with methods other than <init>Guillaume Martres2015-03-281-1/+7
| | | | |
* | | | | Implicits#viewExists: return false after typer instead of AssertionErrorGuillaume Martres2015-03-271-0/+1
|/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | This way, tpd#applyOverloaded can safely be used after typer. This issue was encoutered while working on value classes, step 3 of SIP-15 contains the following peephole optimization: new C(e) == new C(f) => e == f Which requires us to do overloading resolution.
* | | | Merge pull request #425 from smarter/fix/isNullableClassodersky2015-03-262-11/+12
|\ \ \ \ | |/ / / |/| | | Fix isNullableClass to also work after Erasure
| * | | Fix isNullableClass to also work after ErasureGuillaume Martres2015-03-232-11/+12
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | Incidentally this means that: val d = null.asInstanceOf[Double] is now correctly transformed to: val d = scala.Double.unbox(null) Previously it was translated to: val d = null: Double Which is wrong and fails in the backend.
* | | Merge pull request #428 from dotty-staging/trait-constructorsDmitry Petrashko2015-03-255-9/+55
|\ \ \ | | | | | | | | New phase: trait constructors
| * | | New phase trait constructors.Dmitry Petrashko2015-03-252-1/+38
| | | | | | | | | | | | | | | | | | | | Renames constructors in traits so that backend will call them with invokeInterface, instead of invokeSpecial Also makes sure that renamed constructor bodies conforms to type of method
| * | | Emit abstract flag for non-abstract traitsDmitry Petrashko2015-03-251-1/+2
| | | |
| * | | Style fixes in InterceptedMethodsDmitry Petrashko2015-03-251-3/+3
| | | |
| * | | Make CollectEntryPoints stable against type aliases.Dmitry Petrashko2015-03-251-4/+12
| |/ /
* | | Merge pull request #430 from dotty-staging/fix/#429-enclosingMethodDmitry Petrashko2015-03-251-1/+2
|\ \ \ | | | | | | | | Fix #429
| * | | Fix #429Martin Odersky2015-03-251-1/+2
| |/ / | | | | | | | | | enclosingMethod should not crash on NoDenotation.
* / / Make implicit conversion from Position to Coord work for NoPositionGuillaume Martres2015-03-251-1/+3
|/ /
* | Remove check for duplicates in module val names.Dmitry Petrashko2015-03-231-4/+0
| | | | | | | | They are not lifted and they are allowed.
* | InterceptedMethods: enable interception of == and !=Guillaume Martres2015-03-231-1/+1
| |
* | Have derivedValueClass run at initial stageMartin Odersky2015-03-222-6/+13
| | | | | | | | | | | | Fixes #412, by calling derivesFrom in isDerivedValueClass at initial phase. Rewrite of 789dc0c070bde6ce8634aa89e73e31ec1233a6f8 by @smarter. Does the same thing in Symbol#isDerivedValueClass.
* | Invalidate companions of classes lifted by FlatMapMartin Odersky2015-03-211-2/+21
| | | | | | | | | | | | | | RestoreScopes enters classes lifted by FlatMap into the enclosing package scope. It should also invalidate any companions because otherwise we risk linking up a new class with a previously generated companion which exists only as a Java classfile.