RFC: address issue #20882

NEWS.md

@@ -74,8 +74,8 @@ Language changes
     `Vector{T} = Array{T,1}` or a `const` assignment.
 
   * Experimental feature: `x^n` for integer literals `n` (e.g. `x^3`
-    or `x^-3`) is now lowered to `x^Val{n}`, to enable compile-time
-    specialization for literal integer exponents ([#20530]).
+    or `x^-3`) is now lowered to `Base.literal_pow(x, Val{n}, ^)`, to enable
+    compile-time specialization for literal integer exponents ([#20530], [#20889]).
 
 Breaking changes
 ----------------

base/intfuncs.jl

@@ -195,29 +195,28 @@ end
 ^(x::Number, p::Integer)  = power_by_squaring(x,p)
 ^(x, p::Integer)          = power_by_squaring(x,p)
 
-# x^p for any literal integer p is lowered to x^Val{p},
+# x^p for any literal integer p is lowered to Base.literal_pow(x, Val{p}, ^)
 # to enable compile-time optimizations specialized to p.
-# However, we still need a fallback that calls the general ^.
-# To avoid ambiguities for methods that dispatch on the
-# first argument, we dispatch the fallback via internal_pow.
+# However, we still need a fallback that calls the function ^ which may either
+# mean Base.^ or something else, depending on context.
 # We mark these @inline since if the target is marked @inline,
 # we want to make sure that gets propagated,
 # even if it is over the inlining threshold.
-@inline ^(x, p) = internal_pow(x, p)
-@inline internal_pow{p}(x, ::Type{Val{p}}) = x^p
+@inline literal_pow{p}(x, ::Type{Val{p}}, z) = z(x,p)
 
 # Restrict inlining to hardware-supported arithmetic types, which
-# are fast enough to benefit from inlining.  This also makes it
-# easier to override ^ without having to override the Val method.
+# are fast enough to benefit from inlining.
 const HWReal = Union{Int8,Int16,Int32,Int64,UInt8,UInt16,UInt32,UInt64,Float32,Float64}
 const HWNumber = Union{HWReal, Complex{<:HWReal}, Rational{<:HWReal}}
 
 # inference.jl has complicated logic to inline x^2 and x^3 for
-# numeric types.  In terms of Val we can do it much more simply:
-@inline internal_pow(x::HWNumber, ::Type{Val{0}}) = one(x)
-@inline internal_pow(x::HWNumber, ::Type{Val{1}}) = x
-@inline internal_pow(x::HWNumber, ::Type{Val{2}}) = x*x
-@inline internal_pow(x::HWNumber, ::Type{Val{3}}) = x*x*x
+# numeric types.  In terms of Val we can do it much more simply.
+# (The third argument prevents unexpected behavior if a function ^
+# is defined that is not equal to Base.^)
+@inline literal_pow(x::HWNumber, ::Type{Val{0}}, ::typeof(^)) = one(x)
+@inline literal_pow(x::HWNumber, ::Type{Val{1}}, ::typeof(^)) = x
+@inline literal_pow(x::HWNumber, ::Type{Val{2}}, ::typeof(^)) = x*x
+@inline literal_pow(x::HWNumber, ::Type{Val{3}}, ::typeof(^)) = x*x*x
 
 # b^p mod m
 

base/promotion.jl

@@ -254,9 +254,11 @@ end
 Exponentiation operator. If `x` is a matrix, computes matrix exponentiation.
 
 If `y` is an `Int` literal (e.g. `2` in `x^2` or `-3` in `x^-3`), the Julia code
-`x^y` is transformed by the compiler to `x^Val{y}`, to enable compile-time
-specialization on the value of the exponent.  (As a default fallback,
-however, `x^Val{y}` simply calls the `^(x,y)` function.)
+`x^y` is transformed by the compiler to `Base.literal_pow(x, Val{y}, ^)`, to
+enable compile-time specialization on the value of the exponent.
+(As a default fallback, however, `Base.literal_pow(x, Val{y}, z)` simply calls
+`z(x,y)`, where usually `z == Base.^` unless `^` has been defined in the calling
+namespace.)
 
 ```jldoctest
 julia> 3^5

src/julia-syntax.scm

@@ -2057,7 +2057,7 @@
 
                  ((and (eq? f '^) (length= e 4) (integer? (cadddr e)))
                   (expand-forms
-                   `(call ^ ,(caddr e) (call (core apply_type) (top Val) ,(cadddr e)))))
+                   `(call (|.| Base (quote literal_pow)) ,(caddr e) (call (core apply_type) (top Val) ,(cadddr e)) ^)))
 
                  ((and (eq? f '*) (length= e 4))
                   (expand-transposed-op

test/numbers.jl

@@ -2903,9 +2903,12 @@ end
 end
 
 import Base.^
-immutable PR20530; end
+struct PR20530; end
+struct PR20889; x; end
 ^(::PR20530, p::Int) = 1
-^{p}(::PR20530, ::Type{Val{p}}) = 2
+^(t::PR20889, b) = t.x + b
+^(t::PR20889, b::Integer) = t.x + b
+Base.literal_pow{p}(::PR20530, ::Type{Val{p}}, ::typeof(^)) = 2
 @testset "literal powers" begin
     x = PR20530()
     p = 2
@@ -2923,6 +2926,12 @@ immutable PR20530; end
             end
         end
     end
+    @test PR20889(2)^3 == 5
+end
+module M20889 # do we get the expected behavior without importing Base.^?
+    struct PR20889; x; end
+    ^(t::PR20889, b) = t.x + b
+    Base.Test.@test PR20889(2)^3 == 5
 end
 
 @testset "iszero" begin