error parsing a `Int64` value on column 1, row 3; encountered '.' about csv.jl HOT 4 CLOSED

Update:

I just deleted columns 3-22 keeping just the first 2 Int64 columns:

julia> f = CSV.Source("my 2 cols.csv")
[snip]
    CSV.Options:
        delim: ','
        quotechar: '"'
        escapechar: '\\'
        null: ""
        dateformat: Base.Dates.DateFormat(Base.Dates.Slot[],"","english")
123000x2 Data.Schema:
 Trial, Timestep
 Int64,    Int64

julia> ds = Data.stream!(f,Data.Table)
Data.Table:
123000x2 Data.Schema:
 Trial, Timestep
 Int64,    Int64

        Column Data:
        [1,1,1,1,1,1,1,1,1,1  …  3000,3000,3000,3000,3000,3000,3000,3000,3000,3000]
        [0,3,6,9,12,15,18,21,24,27  …  93,96,99,102,105,108,111,114,117,120]

This looks good.

Update:

Three columns also works: 2 Int64 and 1 Float64.

from csv.jl.

Update:

Hmm... if I open the CSV file in Excel and "Save as" CSV, then I can stream! the whole thing.

Update:

Trying to see any differences between the original CSV and the saved CSV.

from csv.jl.

It looks like my original CSV file had trailing commas at the end of each line I didn't notice before.

How could I make CSV.jl read a "bad" CVS file, etc, get it to handle an extra trailing comma, e.g. ignore it?

Edit: Unfortunately, these CSV files are coming from a third-party app and I can't control what comes out of it.

from csv.jl.

We now have the CSV.validate(file) function which takes the same arguments as CSV.read, but will give informative error information for improperly formatted csv data.

from csv.jl.