From b1d8a72a5cc469b5139d9a976ccfa9b4b3eea61d Mon Sep 17 00:00:00 2001
From: Ellen Spertus <ellen.spertus@gmail.com>
Date: Mon, 6 Aug 2018 14:16:19 -0700
Subject: Partly refactored HBaseCrossrefScoreJob. Everything compiles.

---
 .../main/scala/sandcrawler/StringUtilities.scala   | 59 ++++++++++++++++++++++
 1 file changed, 59 insertions(+)
 create mode 100644 scalding/src/main/scala/sandcrawler/StringUtilities.scala

(limited to 'scalding/src/main/scala/sandcrawler/StringUtilities.scala')

diff --git a/scalding/src/main/scala/sandcrawler/StringUtilities.scala b/scalding/src/main/scala/sandcrawler/StringUtilities.scala
new file mode 100644
index 0000000..290b03f
--- /dev/null
+++ b/scalding/src/main/scala/sandcrawler/StringUtilities.scala
@@ -0,0 +1,59 @@
+import java.text.Normalizer
+import java.util.regex.Pattern
+
+object StringUtilities {
+  // Adapted from https://git-wip-us.apache.org/repos/asf?p=commons-lang.git;a=blob;f=src/main/java/org/apache/commons/lang3/StringUtils.java;h=1d7b9b99335865a88c509339f700ce71ce2c71f2;hb=HEAD#l934
+  def removeAccents(s : String) : String = {
+    val replacements = Map(
+      '\u0141' -> 'L',
+      '\u0142' -> 'l',  // Letter ell
+      '\u00d8' -> 'O',
+      '\u00f8' -> 'o'
+    )
+    val sb = new StringBuilder(Normalizer.normalize(s, Normalizer.Form.NFD))
+    for (i <- 0 to sb.length - 1) {
+      for (key <- replacements.keys) {
+        if (sb(i) == key) {
+          sb.deleteCharAt(i);
+          sb.insert(i, replacements(key))
+        }
+      }
+    }
+    val pattern = Pattern.compile("\\p{InCombiningDiacriticalMarks}+")
+    pattern.matcher(sb).replaceAll("")
+  }
+
+  // Adapted from: https://stackoverflow.com/a/16018452/631051
+  def similarity(s1a : String, s2a : String) : Double = {
+    val (s1, s2) = (removeAccents(s1a), removeAccents(s2a))
+    val longer : String = if (s1.length > s2.length) s1 else s2
+    val shorter : String = if (s1.length > s2.length) s2 else s1
+    if (longer.length == 0) {
+      // Both strings are empty.
+      1
+    } else {
+      (longer.length - stringDistance(longer, shorter)) / longer.length.toDouble
+    }
+  }
+
+  // Source: https://oldfashionedsoftware.com/2009/11/19/string-distance-and-refactoring-in-scala/
+  def stringDistance(s1: String, s2: String): Int = {
+    val memo = scala.collection.mutable.Map[(List[Char],List[Char]),Int]()
+    def min(a:Int, b:Int, c:Int) = Math.min( Math.min( a, b ), c)
+    def sd(s1: List[Char], s2: List[Char]): Int = {
+      if (!memo.contains((s1, s2))) {
+        memo((s1,s2)) = (s1, s2) match {
+          case (_, Nil) => s1.length
+          case (Nil, _) => s2.length
+          case (c1::t1, c2::t2)  =>
+            min( sd(t1,s2) + 1, sd(s1,t2) + 1,
+              sd(t1,t2) + (if (c1==c2) 0 else 1) )
+        }
+      }
+      memo((s1,s2))
+    }
+
+    sd( s1.toList, s2.toList )
+  }
+}
+
-- 
cgit v1.2.3