Error when reading CSV files.

[FedX-sudo]

Whenever I read a CSV file via CSV.read("data.csv") I run into the error

ERROR: LoadError: UndefVarError: CSV not defined

I ran the Pkg test CSV command and this is the result:

Test Summary:                  | Pass  Fail  Total
     CSV                       | 1830     1   1831
     CSV.File                  | 1628         1628
     CSV.write                 |   92     1     93
     PooledArrays              |   14           14
     LazyStringVector          |    7            7
     CSV.Rows                  |   37           37
     CSV.detect                |    8            8
     CSV.findrowstarts!        |    4            4
     CSV.promote_typecode      |    2            2
     CSV.File with select/drop |   38           38

ERROR: LoadError: Some tests did not pass: 1830 passed, 1 failed, 0 errored, 0 broken.
in expression starting at /home/fedx/.julia/packages/CSV/CJfFO/test/runtests.jl:13
ERROR: Package CSV errored during testing

[quinnj]

You have to make sure you do using CSV, DataFrames before you call CSV.read("data.csv", DataFrame).

[FedX-sudo]

I did make sure I was using DataFrames however, I did not realize that CSV reading required the DataFrame part, is that a predefined variable?

[quinnj]

CSV doesn't require DataFrames.jl. When call CSV.read(filename, sink), however, you're required to provide a 2nd argument of the kind of thing you want to read the csv file into. Alternatively, you can just call f = CSV.File(filename) and you get a CSV.File object returned, which can be useful for most people. You can iterate rows like:

for row in CSV.File(file)
    # so something with row.col1, for example
end

Or access entire columns like:

col1 = f.col1

[FedX-sudo]

Ok, I am sending the output into an empty data frame, via

df = DataFrame()
CSV.read("data.csv", df)

however, I am still reciving the "csv not defined" error.

[quinnj]

You need to load the CSV.jl package first. When you start the julia program, the CSV.jl package isn't loaded yet, so you need to do something like:

using CSV, DataFrames
df = CSV.read("data.csv", DataFrame)

Note that for the 2nd argument to CSV.read, you pass DataFrame, which is a function that tells the CSV package how to read the csv file into.

[FedX-sudo]

Ok, I did load the package in the last program, but the issue was I needed to define the df variable using CSV.read. Sorry for taking up your time and thank you so much for your help!

[quinnj]

No problem; happy to help