aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Fix #1286: Error on inexistent imports that are not used.Nicolas Stucki2016-10-213-0/+39
| | | | This commit also fixes #1583.
* Merge pull request #1617 from dotty-staging/change-shadow-scalaodersky2016-10-2012-32/+233
|\ | | | | Introduce scalaShadowing package
| * Move files out of the dotty packageMartin Odersky2016-10-2011-32/+20
| | | | | | | | | | | | | | | | | | | | - Remove unused classes Pair and Singleton. - Move classes from dotty.annotation.internal to scala.annotation.internal. The only classes remaining now are in dotty.runtime and DottyPredef. We should probably do something about them as well at some point.
| * Introduce scalaShadowing packageMartin Odersky2016-10-202-0/+213
|/ | | | | | | | | The `scalaShadowing` package is used to safely modify classes and objects in scala so that they can be used from dotty. They will be visible as members of the `scala` package, replacing any objects or classes with the same name. But their binary artifacts are in `scalaShadowing` so they don't clash with the same-named `scala` members at runtime.
* Merge pull request #1594 from dotty-staging/fix-#1576Olivier Blanvillain2016-10-203-1/+9
|\ | | | | Fix #1576: Parameters of Java annotations should be public
| * Fix #1576: Parameters of Java annotations should be publicMartin Odersky2016-10-133-1/+9
| | | | | | | | | | Parameters of annotation classes parsed by the Java parser should have public val parameters. Otherwise they cannot be accessed as fields.
* | Merge pull request #1595 from dotty-staging/fix-#1567odersky2016-10-203-5/+24
|\ \ | | | | | | Fix #1567: Widen private constructor in value class
| * | Fix #1567: Widen private constructor in value classMartin Odersky2016-10-143-5/+24
| |/ | | | | | | | | | | | | | | Private or protected constructors of value classes need to be widenened to public in order to enable boxing anywhere. Technically we should also do something about qualified private constructors, but since we want to get rid of them anyway it's urgent.
* | Merge pull request #1611 from ShaneDelmore/1589_Missing_error_messagesFelix Mulder2016-10-192-5/+87
|\ \ | | | | | | Improved error messages in Desugar.scala
| * | Improved error messages in Desugar.scalaShane Delmore2016-10-192-5/+87
| | |
* | | Merge pull request #1602 from ljdelight/ensimeWorkingGuillaume Martres2016-10-172-0/+2
|\ \ \ | |/ / |/| | Move 'sandbox/scalajs/hello.scala' to 'sandbox/scalajs/src/main/scala…
| * | Ignore ensime files from sbtLucas Burson2016-10-161-0/+2
| | |
| * | Move 'sandbox/scalajs/hello.scala' to ↵Lucas Burson2016-10-161-0/+0
|/ / | | | | | | | | | | | | | | | | | | | | 'sandbox/scalajs/src/main/scala/hello.scala' to fix an ensime error. After this change, `sbt ensimeConfig` works. The ensime error message: ``` [error] You have .scala files in the base of your project. Such "script style" projects [error] are not supported by ENSIME. Simply move them into src/main/scala to get support. [error] Please read https://github.com/ensime/ensime-server/issues/1432 ```
* | Merge pull request #1592 from dotty-staging/fix-#1590odersky2016-10-162-1/+33
|\ \ | | | | | | Fix #1590: Eliminate wildcards when approximating a type
| * | Refinement of interpolation directionMartin Odersky2016-10-132-1/+28
| | |
| * | Fix #1590: Eliminate wildcards when approximating a typeMartin Odersky2016-10-132-1/+6
| |/ | | | | | | | | Fixes #1590. Type variables should never be instantiated to types containing wildcards.
* | Merge pull request #1588 from dotty-staging/fix-#1502odersky2016-10-162-5/+17
|\ \ | | | | | | Fix #1544: Allow long signatures in names
| * | Fix #1544: Allow long signatures in namesMartin Odersky2016-10-132-5/+17
| | | | | | | | | | | | | | | Fixes #1544 by making the length field use 1 or 2 bytes, depending on the number of parameters in a signature.
* | | Merge pull request #1601 from sebastianharko/masterFelix Mulder2016-10-162-1/+45
|\ \ \ | | | | | | | | Missing error messages: Parsers.scala: 2064
| * | | explanation for error message related to use of early definitionsSebastian Harko2016-10-162-1/+45
|/ / /
* | | Merge pull request #1597 from dotty-staging/fix-i1540odersky2016-10-143-9/+41
|\ \ \ | | | | | | | | Fix #1540: overloaded get and isDefined in option-less patmat
| * | | add another test for i1540liu fengyun2016-10-142-1/+15
| | | |
| * | | fix #1540: overloaded get and isDefined in option-less patmatliu fengyun2016-10-142-9/+27
| | | |
* | | | Merge pull request #1598 from dotty-staging/fix-#1515Guillaume Martres2016-10-142-21/+38
|\ \ \ \ | | | | | | | | | | Fix #1515: Don't narrow gadt bounds when frozen
| * | | | Fix #1515: Don't narrow gadt bounds when frozenMartin Odersky2016-10-142-21/+38
| |/ / / | | | | | | | | | | | | Fixes #1515. Review by @smarter.
* | | | Merge pull request #1596 from dotty-staging/fix-#1531Guillaume Martres2016-10-143-5/+11
|\ \ \ \ | | | | | | | | | | Fix #1531: Ignore private members when looking for abstract ones
| * | | | Fix #1531: Ignore private members when looking for abstract onesMartin Odersky2016-10-143-5/+11
| |/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | Private members do not override abstract ones. So when looking for abstract members we need to search with `nonPrivateMember`, not `member`. Fixes #1531. Review by @smarter.
* | | | Merge pull request #749 from smarter/fix/escaping-refs-2odersky2016-10-143-14/+15
|\ \ \ \ | |/ / / |/| | | Typer#escapingRefs: don't let the types of lower bounds escape
| * | | Typer#escapingRefs: remove dead codeGuillaume Martres2016-03-241-9/+1
| | | |
| * | | Typer#escapingRefs: don't let the types of lower bounds escapeGuillaume Martres2016-03-242-5/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In 0efa171e8ccca0d49fc6d800fd21e29f7b7336fd I changed the definition of NamedPartsAccumulator to exclude lower bounds as this is required for the implicit search, but NamedPartsAccumulator is also used by Typer#escapingRefs so in the following code: class Foo[T] val z = { class C ??? : Foo[_ >: C] } the type of z was inferred to be Foo[_ >: C] instead of Foo. To avoid this, NamedPartsAccumulator will only exclude lower bounds if the parameter excludeLowerBounds is explicitely set to true. No test because there is no way to detect that a type has escaped, this might be something that could be added to TreeChecker.
* | | | Merge pull request #1565 from Blaisorblade/warningsFelix Mulder2016-10-146-12/+10
|\ \ \ \ | | | | | | | | | | Small fixes to some warnings/comment typos
| * | | | Fix non-spurious unchecked warningPaolo G. Giarrusso2016-10-071-1/+1
| | | | |
| * | | | Suppress some spurious unchecked warningsPaolo G. Giarrusso2016-10-071-2/+2
| | | | | | | | | | | | | | | | | | | | The warning is invalid but Scalac can't tell.
| * | | | Fix compilation warningsPaolo G. Giarrusso2016-10-072-5/+3
| | | | |
| * | | | Fix typosPaolo G. Giarrusso2016-10-072-4/+4
| | | | |
* | | | | Merge pull request #1577 from dotty-staging/fix-1543-2odersky2016-10-133-3/+76
|\ \ \ \ \ | | | | | | | | | | | | Fix #1543: Ignore IgnoredProto when normalizing method application.
| * | | | | Fix #1543: Ignore IgnoredProto when normalizing method application.Nicolas Stucki2016-10-113-3/+76
| | | | | |
* | | | | | Merge pull request #1586 from dotty-staging/fix/poly-unitGuillaume Martres2016-10-132-1/+17
|\ \ \ \ \ \ | |_|_|_|_|/ |/| | | | | Fix erasure of "def foo[T]: Unit", do not box the return type
| * | | | | Fix erasure of "def foo[T]: Unit", do not box the return typeGuillaume Martres2016-10-132-1/+17
| | | | | |
* | | | | | Merge pull request #1427 from dotty-staging/fix-i1274Felix Mulder2016-10-134-16/+46
|\ \ \ \ \ \ | | | | | | | | | | | | | | fix #1274: test for dotty bootstrap based on tasty
| * | | | | | fix #1274: test for dotty bootstrap based on tastyliu fengyun2016-10-134-16/+46
|/ / / / / /
* | | | | | Merge pull request #1585 from dos65/repl_file_loadingFelix Mulder2016-10-131-16/+7
|\ \ \ \ \ \ | |/ / / / / |/| | | | | Fix #1552: REPL file loading
| * | | | | Fix #1552: loading file inside REPLdos652016-10-131-16/+7
| | | | | |
* | | | | | Merge pull request #1560 from dotty-staging/change-one-polytypeGuillaume Martres2016-10-1339-481/+446
|\ \ \ \ \ \ | | | | | | | | | | | | | | Harmonize PolyType and TypeLambda
| * | | | | | Update tests to use new PolyType syntax.Martin Odersky2016-10-127-20/+20
| | | | | | |
| * | | | | | Use => instead of -> for PolyTypesMartin Odersky2016-10-125-24/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If PolyTypes are to become value types we want to keep `=>` as the arrow for consistency. `->` should be reserved for PolyTypes that do not have side effects on instantiation.
| * | | | | | Merge GenericType, TypeLambda and PolyTypeMartin Odersky2016-10-1232-260/+185
| | | | | | |
| * | | | | | Refactor from distribute{And,Or} to {meet,join}InfoMartin Odersky2016-10-122-176/+196
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The refactored logic only applies to infos of denotations, not general types. The reactoring avoids special cases down the road then PolyTypes and MethodTypes can be used for terms as well as type(bounds).
| * | | | | | Make PolyType a subtype of TypeLambdaMartin Odersky2016-10-122-30/+45
| | | | | | |
| * | | | | | Don't let import symbols go staleMartin Odersky2016-10-122-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Otherwise we can run into problems when checking imports for e.g. scala2Mode in later runs.
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-16 21:09:32 +0000