Fix #825 #834
Fix #825 #834
Conversation
|
Otherwise LGTM. |
| ctx.error("implicit classes may not be toplevel", cdef.pos) | ||
| Nil | ||
| } | ||
| else if (mods is Case) { |
There was a problem hiding this comment.
I think our coding style is } else if ... without a line break after the }.
There was a problem hiding this comment.
No I write newline } quite systematically now. Makes keywords align better.
On Wed, Oct 21, 2015 at 6:43 PM, Guillaume Martres <[email protected]
wrote:
In src/dotty/tools/dotc/ast/Desugar.scala
#834 (comment):@@ -380,18 +380,22 @@ object desugar {
// synthetic implicit C[Ts](p11: T11, ..., p1N: T1N) ... (pM1: TM1, ..., pMN: TMN): C[Ts] =
// new C[Ts](p11, ..., p1N) ... (pM1, ..., pMN) =
val implicitWrappers =
if (mods is Implicit) {if (ctx.owner is Package)ctx.error("implicit classes may not be toplevel", cdef.pos)if (mods is Case)- ctx.error("implicit classes may not case classes", cdef.pos)
if (!mods.is(Implicit))Nilelse if (ctx.owner is Package) {ctx.error("implicit classes may not be toplevel", cdef.pos)Nil}else if (mods is Case) {I think our coding style is } else if ... without a line break after the }
.—
Reply to this email directly or view it on GitHub
https://github.com/lampepfl/dotty/pull/834/files#r42649286.
Martin Odersky
EPFL
Supporess creation of implicit factory if it would be illegal.
Avoids accidental combinations of term and type flags.
Need to survive even if a value class does not have an underlying type. Also: better diagnostics if sigName goes wrong.
Enforces various restrictions of definitions.
Use more complicated control flow with returns instead. sigName is likely performance critical, so some trickiness is justified.
It's done anyway later in FirstTransform.
|
Fixed reviewers's comments and rebased to master. |
Fixes #825 as one of many well-formedness conditions. Review by @smarter.