Missing DimensionMismatch for SparseMatrixCSC assignment

[fgerick]

I've encountered a strange behaviour of assigning to a SparseMatrixCSC in 0.7:

using SparseArrays
a=spzeros(10,10)
b=sprandn(10,20,0.5)
a[:,:]=b[:,:]

This does not give a DimensionMismatch as it did in Julia v0.6, and I guess it should definitely give an error.

In 0.7 it just assignes the 1:10,1:10 elements of b to a:

b[1:10,1:10] ≈ a #true

The broadcasted assignment, however, gives an error:

a[:,:].=b[:,:]
ERROR: DimensionMismatch("array could not be broadcast to match destination")
Stacktrace:
 [1] check_broadcast_shape at ./broadcast.jl:452 [inlined]
 [2] check_broadcast_shape at ./broadcast.jl:453 [inlined]
 [3] check_broadcast_axes at ./broadcast.jl:455 [inlined]
 [4] instantiate at ./broadcast.jl:269 [inlined]
 [5] materialize!(::SubArray{Float64,2,SparseMatrixCSC{Float64,Int64},Tuple{Base.Slice{Base.OneTo{Int64}},Base.Slice{Base.OneTo{Int64}}},false}, ::Base.Broadcast.Broadcasted{SparseArrays.HigherOrderFns.SparseMatStyle,Nothing,typeof(identity),Tuple{SparseMatrixCSC{Float64,Int64}}}) at ./broadcast.jl:727
 [6] top-level scope at none:0

versioninfo()
Julia Version 0.7.0
Commit a4cb80f3ed (2018-08-08 06:46 UTC)
Platform Info:
  OS: macOS (x86_64-apple-darwin14.5.0)
  CPU: Intel(R) Core(TM) i5-7360U CPU @ 2.30GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-6.0.0 (ORCJIT, skylake)
Environment:
  JULIA_NUM_THREADS = 4

[fgerick]

Also causes segmentation faults in some cases if the dimensions mismatch.