CSV

RFC 4180 compliant CSV parsing and encoding for Elixir. Allows to specify other separators, so it could also be named: TSV. Why it is not idk, because of defaults I think.

Why do we want it?

It parses files which contain rows (in utf-8) separated by either commas or other separators.

If that's not enough reason to absolutely ❤️ 💚 💕 ❤️ 💞 💖 it, it also parses a CSV file in order about 2x times as fast as an unparallelized stream implementation 🚀

When do we want it?

Now.

How do I get it?

Add

{:csv, "~> 2.3"}

to your deps in mix.exs like so:

defp deps do
  [
    {:csv, "~> 2.3"}
  ]
end

Note: Elixir 1.1.0 is required for all versions above 1.1.5.

From 1.x to 2.x - the tasty ™️ update.

2.x has some nice new features like the separation between hair- and error-raising decode! and the zen of decode, better error messages and an easier to understand codebase for you to contribute.

The only thing you have to do to upgrade to 2.x is to change your calls to decode to decode!, and adjust your exceptions-catching code to catch the right exceptions still. But why not take full advantage and convert to decode and the new tuple stream?

You know you want it.

Great! How do I use it right now?

There are two interesting things you want to do regarding csvs - encoding end decoding.

Decoding

Do this to decode:

File.stream!("data.csv") |> CSV.decode

And you'll get a stream of row tuples:

[ok: ["a", "b"], ok: ["c", "d"]]

And, potentially error tuples:

[error: "Row has length 3 - expected length 2 on line 1", ok: ["c", "d"]]

Use the bang to decode! into a two-dimensional list, raising errors as they occur:

File.stream!("data.csv") |> CSV.decode!

Encoding

Do this to encode a table (two-dimensional list):

table_data |> CSV.encode

And you'll get a stream of lines ready to be written to an IO. So, this is writing to a file:

file = File.open!("test.csv", [:write, :utf8])
table_data |> CSV.encode |> Enum.each(&IO.write(file, &1))

I have this file, but it's tab-separated ⁉️

Pass in another separator to the decoder:

File.stream!("data.csv") |> CSV.decode(separator: ?\t)

If you want to take revenge on whoever did this to you, encode with semicolons like this:

your_data |> CSV.encode(separator: ?;)

You can also specify headers when encoding, which will encode map values into the right place:

[%{"a" => "value!"}] |> CSV.encode(headers: ["z", "a"])
# ["z,a\\r\\n", ",value!\\r\\n"]

You'll surely appreciate some more info on encode

Polymorphic encoding

Make sure your data gets encoded the way you want - implement the CSV.Encode protocol for whatever strange you wish to encode:

defimpl CSV.Encode, for: MyData do
  def encode(%MyData{has: fun}, env \\ []) do
    "so much #{fun}" |> CSV.Encode.encode(env)
  end
end

Or similar.

Ensure performant encoding

The encoding protocol implements a fallback to Any for types where a simple call o to_string will provide unambiguous results. Protocol dispatch for the fallback to Any is very slow when protocols are not consolidated, so make sure you have consolidate_protocols: true in your mix.exs or you consolidate protocols manually for production in order to get good performance.

There is more to know about everything ™️ - Check the doc

License

MIT

Contributions & Bugfixes are most welcome!

Please make sure to add tests. I will not look at PRs that are either failing or lowering coverage. Also, solve one problem at a time.

TylerTemp/csv

CSV