CODEWITHSUNDEEP
scalacsvreadfile
Joey Kim
Joey Kim

Reputation: 179

read csv file(numeric type) in scala

I would like to read csv file in scala.

enter image description here

  val rows = ArrayBuffer[Array[String]]()

      using(io.Source.fromFile("test.csv")) { source =>
        for (line <- source.getLines) {
          rows += line.split(",").map(_.trim)
          }
        }     
      def using[A <: { def close(): Unit }, B](resource: A)(f: A => B): B =
         try {
            f(resource)
         }finally {
            resource.close()
            }

I used above code to read csv file.

I get ArrayBuffer(Array[String]) now, but I want to get ArrayBuffer(Array[Double]).

toDouble method is wrong

What's the ideal way to do?

Thank you for your time.

Upvotes: 0

Views: 773

Answers (2)

Nicolas Rinaudo
Nicolas Rinaudo

Reputation: 6168

I would recommend against parsing CSV manually, you'll most likely eventually encounter one of the annoying edge cases (values with lines or columns separators, for instance). It's often better and safer to use dedicated libraries.

I'm going to recommend kantan.csv because I'm the author, but there are plenty other quality libraries out there (product-collections, pure csv...).

With kantan.csv, your problem would be solved thusly:

import kantan.csv.ops._

// For an immutable structure.
new java.io.File("test.csv").readCsv[List, List[Double]](',', false)

// For an Array, as in your example.
new java.io.File("test.csv").readCsv[Array, Array[Double]](',', false)

Upvotes: 0

Łukasz
Łukasz

Reputation: 8663

toDouble is right, just call it on strings.

rows.map(_.map(_.toDouble))

Or right in your code

using(io.Source.fromFile("test.csv")) { source =>
  for (line <- source.getLines) {
    rows += line.split(",").map(_.trim).map(_.toDouble)
  }
}

To avoid mutable collections you could use lists, like this:

val rows: List[List[Double]] = using(io.Source.fromFile("test.csv")) { source =>
  source.getLines.toList map { line =>
    line.split(",").map(_.trim.toDouble).toList
  }
}

Upvotes: 3

Related Questions