Skip to content
Snippets Groups Projects
Commit 030b5801 authored by dogfooter's avatar dogfooter
Browse files

add: preprocessing 400

parent 40609468
No related branches found
No related tags found
No related merge requests found
import scala.io.Source
import java.io._
val timestamp = System.currentTimeMillis();
val input500 = sc.textFile("/source/nt/Universities400.nt")
var prefix = Source.fromFile("./3rd-dep/prefix-list.csv").getLines().toList;
var prefixPairList = prefix.map(line => line.split(",").toList)
val parse500 = input500.map(line => line.replace(prefixPairList(0)(1), prefixPairList(0)(0))).map(line => line.replace(prefixPairList(1)(1), prefixPairList(1)(0))).map(line => line.replace(prefixPairList(2)(1), prefixPairList(2)(0))).map(line => line.replace(prefixPairList(3)(1), prefixPairList(3)(0))).map(line => line.replace(prefixPairList(4)(1), prefixPairList(4)(0))).map(line => line.replace(" <", "\t<")).map(line => line.replace(" _", "\t_")).map(line => line.replace(" \"", "\t\"")).map(line => line.replace(" \\.$", ""))
val union500 = parse500.map(line => line.replaceAll(" \\.$", ""))
union500.saveAsTextFile("/source/input/instance/400")
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment