From c3a7a86cdf888767b998b27e635782fd5b1e1df4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Pedro?= <joaop@simbioseventures.com>
Date: Sun, 1 Aug 2021 22:27:23 +0100
Subject: [PATCH] Add base impleentation for levenshtein functions

---
 .../exec/expr/fn/impl/StringFunctions.java    | 33 +++++++++++++++++++
 .../physical/impl/TestStringFunctions.java    | 11 +++++++
 2 files changed, 44 insertions(+)

diff --git a/sabot/kernel/src/main/java/com/dremio/exec/expr/fn/impl/StringFunctions.java b/sabot/kernel/src/main/java/com/dremio/exec/expr/fn/impl/StringFunctions.java
index 698ad6087f..79fbde0c19 100644
--- a/sabot/kernel/src/main/java/com/dremio/exec/expr/fn/impl/StringFunctions.java
+++ b/sabot/kernel/src/main/java/com/dremio/exec/expr/fn/impl/StringFunctions.java
@@ -1717,4 +1717,37 @@ public void eval() {
       out.end = outBytea.length;
     }
   }
+
+  /**
+   * Calculates the levenshtein distance of given strings.
+   */
+  @FunctionTemplate(name = "levenshtein", scope = FunctionScope.SIMPLE, nulls = NullHandling.NULL_IF_NULL)
+  public static class Levenshtein implements SimpleFunction {
+    @Param  VarCharHolder in1;
+    @Param  VarCharHolder in2;
+    @Output IntHolder out;
+
+    @Override
+    public void setup() {}
+
+    @Override
+    public void eval() {
+      int len1 = in1.end - in1.start;
+      int len2 = in2.end - in2.start;
+      // dist[i][j] represents the Levenstein distance between the strings
+      int[][] dist = new int[len1 + 1][len2 + 1];
+      for (int i = 0; i <= len1; i++) dist[i][0] = i;
+      for (int j = 1; j <= len2; j++) dist[0][j] = j;
+      for (int j = 0; j < len2; j++) {
+        for (int i = 0; i < len1; i++) {
+          if(in1.buffer.getByte(i) == in2.buffer.getByte(j)) {
+            dist[i + 1][j + 1] = dist[i][j];
+          } else {
+            dist[i + 1][j + 1] = Math.min(Math.min(dist[i][j + 1] + 1, dist[i + 1][j] + 1), dist[i][j] + 1);
+          }
+        }
+      }
+      out.value = dist[len1][len2];
+    }
+  }
 }
diff --git a/sabot/kernel/src/test/java/com/dremio/exec/physical/impl/TestStringFunctions.java b/sabot/kernel/src/test/java/com/dremio/exec/physical/impl/TestStringFunctions.java
index eb3b100e48..9078d9c4e0 100644
--- a/sabot/kernel/src/test/java/com/dremio/exec/physical/impl/TestStringFunctions.java
+++ b/sabot/kernel/src/test/java/com/dremio/exec/physical/impl/TestStringFunctions.java
@@ -343,4 +343,15 @@ public void stringfuncs(){
     });
   }
 
+  @Test
+  public void levenshtein(){
+    testFunctions(new Object[][]{
+      { "levenshtein('test', 'task')", 2},
+      { "levenshtein('kitten', 'sitting')", 3},
+      { "levenshtein('', 'a')", 1},
+      { "levenshtein('cat', 'coat')", 1},
+      { "levenshtein('book', 'back')", 2}
+    });
+  }
+
 }