Docs about csv

[beatrichartz]

Interesting - can you elaborate or post some example code you tried and the code that solved it with the other library so I can create an appropriate example?

Thanks!

[simonh1000]

Sure

CSVLixir.read("ignore/customers.csv")
|> Enum.to_list
|> Enum.each( fn([id, nm, csr, sal]) ->
        Customer.changeset(%Customer{},
            %{masterid: id,
              custname: nm,
              csrid: String.to_integer(csr),
              salesid: String.to_integer(sal)})
        |> Repo.insert
    end)

[beatrichartz]

Depending on the version you use and assuming the file contains no headers, these are the solutions I can think of:

v1.x

File.stream!("ignore/customers.csv")
|> CSV.decode
|> Enum.each(fn([id, nm, csr, sal]) ->
        Customer.changeset(%Customer{},
            %{masterid: id,
              custname: nm,
              csrid: String.to_integer(csr),
              salesid: String.to_integer(sal)})
        |> Repo.insert
    end)

Latest master:

File.stream!("ignore/customers.csv")
|> CSV.decode
|> Enum.each(fn
        {:ok, [id, nm, csr, sal]} ->
            Customer.changeset(%Customer{},
                %{masterid: id,
                  custname: nm,
                  csrid: String.to_integer(csr),
                  salesid: String.to_integer(sal)})
            |> Repo.insert
         {:error, message} ->
            # Whatever you want to do with invalid rows
    end)

Do these work?

[simonh1000]

How annoying - i'm sure I tried that. I only needed csv to copy across a database so absolute performance is not a necessity, so I'm going to stick with what i had. But I would recommend you add something like this to the README

[hopewise]

Hi,

As I've read the latest sample code above, I see that Enum.each is used, however, I read in the elixir docs:

Note that the functions in the Enum module are eager: they always start the enumeration of the given enumerable. The Stream module allows lazy enumeration of enumerables and provides infinite streams.

So, why not use Stream instead of Enum to iterate through the output of CSV.decode?

[beatrichartz]

@hopewise CSV.decode works with streams, but at some point you want to start enumeration, otherwise the stream will never be run. This is why you are using Enum.each, which is roughly equivalent to doing Stream.each ... |> Stream.run