aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Fix hole in subtyping of modulesMartin Odersky2015-10-252-4/+18
| | | | | We did not handle correctly the case exemplified by `range.scala`: A module this type, which is compared with a getter to the same module. Seen in the wild in scala.math.BigDecimal.
* Merge pull request #863 from dotty-staging/fix/inference-methcallGuillaume Martres2015-10-253-3/+42
|\ | | | | Fix/inference methcall
| * Readd `.widen` which was accidentally removed from the previous commitGuillaume Martres2015-10-251-1/+1
| |
| * Simplify logic in adaptMartin Odersky2015-10-251-10/+2
| |
| * Fix issue that prevented instantiating some tvars before implicit searchGuillaume Martres2015-10-252-3/+29
| |
| * Less eager tvar interpolation: wait until method calls are fully appliedGuillaume Martres2015-10-252-2/+23
|/ | | | Fix #738
* Merge pull request #843 from dotty-staging/fix-#830odersky2015-10-252-3/+25
|\ | | | | Fix #830: Compiler hangs on implicit search with singleton &/|
| * Only replace intersections of constants with NothingMartin Odersky2015-10-231-4/+10
| |
| * Fix #830: Compiler hangs on implicit search with singleton &/|Martin Odersky2015-10-222-3/+19
| | | | | | | | | | | | In fact we get a deep subtype recursion when compileing i830.scala. The problem goes away once we make use of the fact that the intersection of two singleton types which are not subtypes of each other is empty.
* | Merge pull request #858 from lampepfl/DarkDimius-patch-3Dmitry Petrashko2015-10-231-1/+1
|\ \ | | | | | | Add link to gitter channel, delete travis badge.
| * | Add link to gitter channel, delete travis badge.Dmitry Petrashko2015-10-231-1/+1
|/ /
* | Merge pull request #857 from dotty-staging/change-readmeDmitry Petrashko2015-10-232-7/+86
|\ \ | | | | | | Change readme
| * | Change readme to reflect current stateMartin Odersky2015-10-231-7/+23
| | |
| * | fix typosMartin Odersky2015-10-231-8/+8
| | |
| * | fix typosMartin Odersky2015-10-231-2/+4
| | |
| * | fix typosMartin Odersky2015-10-231-3/+3
| | |
| * | fix typosMartin Odersky2015-10-231-2/+2
| | |
| * | Draft blog postMartin Odersky2015-10-231-0/+61
| | |
* | | Merge pull request #853 from dotty-staging/partest-runDmitry Petrashko2015-10-234-7/+12
|\ \ \ | |/ / |/| | Run CI-partest with bootstrapped compiler
| * | Print stack traces before creating stub symbolsDmitry Petrashko2015-10-232-3/+8
| | |
| * | Revert "Workaround #840"Dmitry Petrashko2015-10-231-1/+1
| | | | | | | | | | | | This reverts commit acd7b9d31cd0a2fbc285921bdc5a7bdc9c1d2872.
| * | Enable running pos tests in partestDmitry Petrashko2015-10-231-1/+1
| | | | | | | | | | | | | | | As junit tests are run with dotty-compiled-by-scalac, it makes sense to execute run tests with dotty-compiled-by-dotty.
| * | Run partest under bootstrapped dotty.Dmitry Petrashko2015-10-231-3/+3
| | |
| * | Workaround #840Dmitry Petrashko2015-10-231-1/+1
| | |
* | | Merge pull request #852 from dotty-staging/fix-scriptDmitry Petrashko2015-10-232-1/+2
|\ \ \ | |/ / |/| | Fix bug in dotc script
| * | Indicate that dotty is being built.Dmitry Petrashko2015-10-231-0/+1
| | |
| * | Fix bug in dotc script.Dmitry Petrashko2015-10-231-1/+1
|/ / | | | | | | Used to always use bootstrapped version
* | Merge pull request #832 from dotty-staging/fix-#815-hk-constructor-aliasDmitry Petrashko2015-10-232-1/+17
|\ \ | | | | | | Fix #815.
| * | Fix #815.Martin Odersky2015-10-202-1/+17
| | | | | | | | | | | | Need to handle hk-aliases specially in typedNew
* | | Merge pull request #842 from dotty-staging/fix-838Dmitry Petrashko2015-10-232-2/+2
|\ \ \ | | | | | | | | Fix #838. Fixes to non-local modules& static forwarders.
| * | | Update version of bundled backend.Dmitry Petrashko2015-10-221-1/+1
| | | |
| * | | Fix DottyBackendInterafec.isStaticModuleClass.Dmitry Petrashko2015-10-221-1/+1
| | | | | | | | | | | | | | | | Used to pass wrong context.
* | | | Merge pull request #849 from dotty-staging/fix-#840Dmitry Petrashko2015-10-232-2/+2
|\ \ \ \ | | | | | | | | | | Fix #840
| * | | | Don't drop impure expressions when eliding isInstanceOf tests.Martin Odersky2015-10-231-1/+1
| | | | |
| * | | | Fix #840Martin Odersky2015-10-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem was that interceptTypeTestCasts was run at wrong phase. It saw after erasura a type of the form `x.Array$$T`. Before erasure that type is simply an alias of another type, but after erasure, Arraya$$T is defined to be a type alias of a Wildcard type.
* | | | | Merge pull request #846 from dotty-staging/fix-#827odersky2015-10-235-44/+43
|\ \ \ \ \ | |/ / / / |/| | | | Fix #827
| * | | | Address reviewer commentsMartin Odersky2015-10-221-5/+1
| | | | |
| * | | | Simplify treatment of merge conflictsMartin Odersky2015-10-221-37/+16
| | | | | | | | | | | | | | | | | | | | Now, we always throw a merge error. No more exceptions.
| * | | | Better handling of merge errorsMartin Odersky2015-10-224-4/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of picking one at random, throw a MergeError which might be caught later in mergeDenot. MergeDenot has enough info to pick a simulate Scala2 linarization if the prefix comes from Scala2, or it rethrows the exception so that it becomes a type error.
| * | | | Check for cyclic inheritance.Martin Odersky2015-10-221-1/+8
|/ / / /
* | | | Merge pull request #841 from dotty-staging/fix-#831-object-selfodersky2015-10-223-19/+33
|\ \ \ \ | | | | | | | | | | Fix #831 object self
| * | | | Adapt sourceModule to new selfInfo scheme.Martin Odersky2015-10-222-3/+6
| | | | |
| * | | | Fix #831Martin Odersky2015-10-222-17/+28
| | | | | | | | | | | | | | | | | | | | | | | | | Need to create a self symbols for modules with explicit self type, but need to take care it is already typed, or sourceModule risks running into CyclicReferences.
* | | | | Merge pull request #835 from dotty-staging/add-more-testsodersky2015-10-2211-1/+131
|\ \ \ \ \ | |/ / / / |/| | | | Add more tests
| * | | | Remove empty test fileMartin Odersky2015-10-221-0/+0
| | | | |
| * | | | More testsMartin Odersky2015-10-2111-0/+130
| | | | |
| * | | | Remove disgnostic printlnMartin Odersky2015-10-211-1/+1
| |/ / /
* | | | Merge pull request #834 from dotty-staging/fix-#825odersky2015-10-2215-63/+222
|\ \ \ \ | | |_|/ | |/| | Fix #825
| * | | Avoid mutation in checkWellFormedMartin Odersky2015-10-221-4/+5
| | | | | | | | | | | | | | | | It's done anyway later in FirstTransform.
| * | | Avoid caching in sigNameMartin Odersky2015-10-221-8/+7
| | | | | | | | | | | | | | | | | | | | | | | | Use more complicated control flow with returns instead. sigName is likely performance critical, so some trickiness is justified.