A JVM library for reading CSV files without damaging content in Unicode characters beyond the basic multilingual plane.

Some widely used libraries for working with .csv files read arrays of bytes or streams of characters, and fail to preserve Unicode codepoints outside the basic multilingual plane. safecsv reads .csv sources by Unicode codepoint, so it is reliable rather than fast.

More information: http://neelsmith.github.io/safecsv/

This script (in groovy) reads a .csv file and prints the first column of every row:

import edu.holycross.shot.safecsv.SafeCsvReader

SafeCsvReader srcReader = new SafeCsvReader("CSVFILE")
srcReader.readAll().each { record ->
    println("First column is " + record[0])
}

This script (in groovy) reads a .csv file and prints the first column of every row labelled with the value from the corresponding header column:

import edu.holycross.shot.safecsv.SafeCsvReader

boolean hasHeader = true
SafeCsvReader srcReader = new SafeCsvReader("CSVFILE", true)
def headers = srcReader.readHeader()
srcReader.readAll().each { record ->
    println("Column " + headers[0] + ": " + record[0])
}