Encoding collections as rows

CSV is often used to store rows of homogeneous data - each cell could be an Int, for instance. We’ll see in this post how kantan.csv supports this kind of scenario.

In kantan.csv, all CSV serialization is done through CsvWriter, which you can think of as a highly specialised version of Writer. One retrieves an instance of CsvWriter as follows:

import kantan.csv._
import kantan.csv.ops._

// File in which we'll be writing the CSV data.
val out = java.io.File.createTempFile("kantan.csv", "csv")

val writer = out.asCsvWriter[List[Int]](rfc.withHeader("Column 1", "Column 2", "Column 3"))

Note the type parameter to asCsvWriter: this is what the returned instance of CsvWriter will know to encode. Since we’re trying to write lists of integers, we requested a CsvWriter[List[Int]], but we could have requested any subtype of TraversableOnce of any primitive Scala type.

The value parameters are the column separator and optional header row.

Now that we have a CsvWriter that knows how to encode List[Int], we can just call its write method repeatedly:

// Writes a couple of rows.
writer.write(List(0, 1, 2))
writer.write(List(3, 4, 5))

// Makes sure resources are freed.
writer.close()

Note that CsvWriter supports a fluent style of chaining operations (or, in less pretentious english, each write returns the CsvWriter itself), which I could have used to simplify the above code. I feel that spelling things out helps, though, especially the explicit call to close at the end.

Let’s make sure that we got the expected output:

scala.io.Source.fromFile(out).mkString
// res3: String = """Column 1,Column 2,Column 3
// 0,1,2
// 3,4,5
// """

What to read next

If you want to learn more about: