more consistently ignore whitespace in readdlm

[stevengj]

readcsv(IOBuffer(" 1, 2,3")) gives [1.0 2.0 3.0], but readcsv(IOBuffer(" 1 , 2,3")) gives [" 1 ", 2.0, 3.0].

As another example, readcsv(IOBuffer("\"hello, again\",\"goodbye\"")) gives ["hello, again" "goodbye"], but readcsv(IOBuffer(" \"hello, again\",\"goodbye\"")) gives " \"hello" " again\"" "goodbye". And readcsv(IOBuffer("\"hello, again\" ,\"goodbye\"")) gives an error (unexpected character ' ' after quoted field at row 1 column 1).

It seems like it would be better to treat all of these cases the same, and ignore whitespace before and after the delimiter.

[KristofferC]

The first case seems fixed here:

julia> readcsv(IOBuffer(" 1, 2,3"))
1×3 Array{Float64,2}:
 1.0  2.0  3.0

julia> readcsv(IOBuffer(" 1 , 2,3"))
1×3 Array{Float64,2}:
 1.0  2.0  3.0

The other ones are the same as when the issue was made.

[ViralBShah]

Ok to close?

[stevengj]

The numeric case works now:

julia> readdlm(IOBuffer(" 1 , 2,3"), ',')
1×3 Matrix{Float64}:
 1.0  2.0  3.0

But the string behavior still seems odd to me:

julia> using DelimitedFiles

julia> readdlm(IOBuffer("\"hello, again\",\"goodbye\""), ',')
1×2 Matrix{Any}:
 "hello, again"  "goodbye"

julia> readdlm(IOBuffer(" \"hello, again\",\"goodbye\""), ',')
1×3 Matrix{Any}:
 " \"hello"  " again\""  "goodbye"

julia> readdlm(IOBuffer("\"hello, again\" ,\"goodbye\""), ',')
ERROR: unexpected character ' ' after quoted field at row 1 column 1
Stacktrace: