ERROR: InexactError: trunc(Int64, NaN) when multiplying zero with range in Julia v0.7

[zsoerenm]

If I multiply zero with a range: 0.0 .* (1:8000) in Julia v0.7, I get the folliowing error:

ERROR: InexactError: trunc(Int64, NaN)
Stacktrace:
 [1] trunc at ./float.jl:693 [inlined]
 [2] round at ./float.jl:359 [inlined]
 [3] floatrange(::Type{Float64}, ::Int64, ::Int64, ::Int64, ::Int64) at ./twiceprecision.jl:361
 [4] _range(::Float64, ::Float64, ::Nothing, ::Int64) at ./twiceprecision.jl:435
 [5] #range#29 at ./range.jl:76 [inlined]
 [6] #range at ./none:0 [inlined]
 [7] broadcasted(::Base.Broadcast.DefaultArrayStyle{1}, ::typeof(*), ::Float64, ::UnitRange{Int64}) at ./broadcast.jl:977
 [8] broadcasted(::Function, ::Float64, ::UnitRange{Int64}) at ./broadcast.jl:1141
 [9] top-level scope at none:0

This did not happen in v0.6.

Julia Version 0.7.0
Commit a4cb80f3ed (2018-08-08 06:46 UTC)
Platform Info:
  OS: Linux (x86_64-pc-linux-gnu)
  CPU: Intel(R) Core(TM) i7-6820HQ CPU @ 2.70GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-6.0.0 (ORCJIT, skylake)

Workaround for now is: 0.0 .* collect(1:8000)

BTW if 0 is a Int, it produces the following error:

julia> 0 .* (1:8000)
ERROR: ArgumentError: step cannot be zero
Stacktrace:
 [1] steprange_last(::Int64, ::Int64, ::Int64) at ./range.jl:190
 [2] Type at ./range.jl:180 [inlined]
 [3] _rangestyle at ./range.jl:93 [inlined]
 [4] _range at ./range.jl:91 [inlined]
 [5] #range#29 at ./range.jl:76 [inlined]
 [6] #range at ./none:0 [inlined]
 [7] broadcasted(::Base.Broadcast.DefaultArrayStyle{1}, ::typeof(*), ::Int64, ::UnitRange{Int64}) at ./broadcast.jl:977
 [8] broadcasted(::Function, ::Int64, ::UnitRange{Int64}) at ./broadcast.jl:1141
 [9] top-level scope at none:0

[fredrikekre]

There is an open PR that fixes this: #29056

[zsoerenm]

I will close this as #29056 is merged.