prefix map ambiguous

Development Compiler
1

when I write sequence.prefix(n).map(f) the compiler complains that .prefix
is ambiguous. If I replace it with [..<n] it works fine. Why is this
happening? I found SR-1856 <Issues · apple/swift-issues · GitHub; which
looks like a similar issue but it’s marked fixed.

2

Hi Kelvin,

Can you give a bit more detail on the type of the sequence, and which compiler version you’re seeing this on?

I can’t reproduce this with a recent Xcode or a recent master toolchain with this code:

let sequence = 0..<10
let n = 5
let f = { $0*2 }

sequence.prefix(n).map(f)

Thanks
Ben

···

On Jan 11, 2018, at 19:11, Kelvin Ma via swift-dev <swift-dev@swift.org> wrote:

when I write sequence.prefix(n).map(f) the compiler complains that .prefix is ambiguous. If I replace it with [..<n] it works fine. Why is this happening? I found SR-1856 <Issues · apple/swift-issues · GitHub; which looks like a similar issue but it’s marked fixed.
_______________________________________________
swift-dev mailing list
swift-dev@swift.org
https://lists.swift.org/mailman/listinfo/swift-dev

3

it happens when the n is a literal

let a:[Int] = [1, 2, 3]
let f:(Int) -> Int = { $0 + 5 }

print(a.prefix(5).map(f))

ab.swift:4:7: error: ambiguous use of 'prefix'
print(a.prefix(5).map(f))
      ^
Swift.Collection:85:17: note: found this candidate
    public func prefix(_ maxLength: Int) -> Self.SubSequence
                ^
Swift.Sequence:81:17: note: found this candidate
    public func prefix(_ maxLength: Int) -> AnySequence<Self.Element>

···

On Thu, Jan 11, 2018 at 11:30 PM, Ben Cohen <ben_cohen@apple.com> wrote:

Hi Kelvin,

Can you give a bit more detail on the type of the sequence, and which
compiler version you’re seeing this on?

I can’t reproduce this with a recent Xcode or a recent master toolchain
with this code:

let sequence = 0..<10
let n = 5
let f = { $0*2 }

sequence.prefix(n).map(f)

Thanks
Ben

On Jan 11, 2018, at 19:11, Kelvin Ma via swift-dev <swift-dev@swift.org> > wrote:

when I write sequence.prefix(n).map(f) the compiler complains that .prefix
is ambiguous. If I replace it with [..<n] it works fine. Why is this
happening? I found SR-1856 <Issues · apple/swift-issues · GitHub; which
looks like a similar issue but it’s marked fixed.
_______________________________________________
swift-dev mailing list
swift-dev@swift.org
https://lists.swift.org/mailman/listinfo/swift-dev

4

This seems to be [SR-1856] prefix() chained after sorted() and before map() confuses type inference in 3.0 · Issue #44465 · apple/swift · GitHub, which is resolved in Swift 4.1.

Martin

···

Am 12.01.2018 um 07:49 schrieb Kelvin Ma via swift-dev <swift-dev@swift.org>:

it happens when the n is a literal

let a:[Int] = [1, 2, 3]
let f:(Int) -> Int = { $0 + 5 }

print(a.prefix(5).map(f))

ab.swift:4:7: error: ambiguous use of 'prefix'
print(a.prefix(5).map(f))
      ^
Swift.Collection:85:17: note: found this candidate
    public func prefix(_ maxLength: Int) -> Self.SubSequence
                ^
Swift.Sequence:81:17: note: found this candidate
    public func prefix(_ maxLength: Int) -> AnySequence<Self.Element>

On Thu, Jan 11, 2018 at 11:30 PM, Ben Cohen <ben_cohen@apple.com> wrote:
Hi Kelvin,

Can you give a bit more detail on the type of the sequence, and which compiler version you’re seeing this on?

I can’t reproduce this with a recent Xcode or a recent master toolchain with this code:

let sequence = 0..<10
let n = 5
let f = { $0*2 }

sequence.prefix(n).map(f)

Thanks
Ben

On Jan 11, 2018, at 19:11, Kelvin Ma via swift-dev <swift-dev@swift.org> wrote:

when I write sequence.prefix(n).map(f) the compiler complains that .prefix is ambiguous. If I replace it with [..<n] it works fine. Why is this happening? I found SR-1856 which looks like a similar issue but it’s marked fixed.
_______________________________________________
swift-dev mailing list
swift-dev@swift.org
https://lists.swift.org/mailman/listinfo/swift-dev

_______________________________________________
swift-dev mailing list
swift-dev@swift.org
https://lists.swift.org/mailman/listinfo/swift-dev