Merge pull request #370 from zhuowang-linkedin/zhuowang/patternMatchB…

…ugFix

Fix PatternMatch hashCode bug

src/main/scala/com/amazon/deequ/analyzers/PatternMatch.scala

@@ -60,6 +60,23 @@ case class PatternMatch(column: String, pattern: Regex, where: Option[String] =
   override protected def additionalPreconditions(): Seq[StructType => Unit] = {
     hasColumn(column) :: isString(column) :: Nil
   }
+
+  // PatternMatch hasCode is different with the same-parameter objects
+  // because Regex compares by address
+  // fix this by tuple with pattern string
+  private val internalObj = (column, pattern.toString(), where)
+
+  override def hashCode(): Int = {
+    internalObj.hashCode()
+  }
+
+  override def equals(obj: Any): Boolean = {
+    obj match {
+      case o: PatternMatch => internalObj.equals(o.asInstanceOf[PatternMatch].internalObj)
+      case _ => false
+    }
+  }
+
 }
 
 object Patterns {

src/test/scala/com/amazon/deequ/analyzers/AnalyzerTests.scala

@@ -679,6 +679,13 @@ class AnalyzerTests extends AnyWordSpec with Matchers with SparkContextSpec with
   "Pattern compliance analyzer" should {
     val someColumnName = "some"
 
+    "PatternMatch hashCode should equal for the same pattern" in {
+      val p1 = PatternMatch("col1", "[a-z]".r)
+      val p2 = PatternMatch("col1", "[a-z]".r)
+      p1.hashCode() should equal(p2.hashCode())
+      p1 should equal(p2)
+    }
+
     "not match doubles in nullable column" in withSparkSession { sparkSession =>
 
        val df = dataFrameWithColumn(someColumnName, DoubleType, sparkSession, Row(1.1),