allunique behaviour doesn't match documentation

[karajan9]

Hello everyone,
allunique doesn't seem to work on my structs. The documentation says: "allunique: Return true if all values from itr are distinct when compared with isequal."

import Base: isequal

mutable struct Cart
    x::Int
    y::Int
end

isequal(c1::Cart, c2::Cart) = return (c1.x == c2.x && c1.y == c2.y)

c1 = Cart(2, 3)
c2 = Cart(2, 3)

isequal(c1, c2)  # == true
c1 ∈ [c1, c2]  # == true

allunique([c1, c2])  # == true?!

Is this expected?

julia> versioninfo()
Julia Version 1.0.2
Commit d789231e99 (2018-11-08 20:11 UTC)
Platform Info:
  OS: Windows (x86_64-w64-mingw32)
  CPU: Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-6.0.0 (ORCJIT, ivybridge)
Environment:
  JULIA_NUM_THREADS = 2

[stev47]

I agree, it's definitely misleading.
You can work around by defining your own hash function though:

Base.hash(c::Cart, h::UInt) = hash(c.y, hash(c.x, h))

The documentation of hash() says:

hash(x[, h::UInt])

Compute an integer hash code such that isequal(x,y) implies hash(x)==hash(y). The optional second argument h is a hash code
to be mixed with the result.

So if you define your own isequal() without corresponding hash() you may violate that guarantee.

[karajan9]

Ah, that's good to know, thank you!

So, which is right then, the documentation or the behavior? Or rather, which one should be right?

[NeroBlackstone]

Same problem here, documentation should improvement