saved

NVIDIA · May 24, 2022 · 4063aaf · 4063aaf
1 parent 72d3ffd
commit 4063aaf
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 7 deletions.
diff --git a/sql-plugin/src/main/scala/com/nvidia/spark/rapids/RegexParser.scala b/sql-plugin/src/main/scala/com/nvidia/spark/rapids/RegexParser.scala
@@ -767,7 +767,7 @@ class CudfRegexTranspiler(mode: RegexMode) {
 //      println("everything is fine")
     }
 
-    def foo(regex: RegexAST): Unit = {
+    def checkEndAnchorNearNewline(regex: RegexAST): Unit = {
       regex match {
         case RegexSequence(parts) =>
           parts.indices.foreach { i =>
@@ -779,16 +779,16 @@ class CudfRegexTranspiler(mode: RegexMode) {
             }
           }
         case RegexChoice(l, r) =>
-          foo(l)
-          foo(r)
-        case RegexGroup(_, term) => foo(term)
-        case RegexRepetition(ast, _) => foo(ast)
+          checkEndAnchorNearNewline(l)
+          checkEndAnchorNearNewline(r)
+        case RegexGroup(_, term) => checkEndAnchorNearNewline(term)
+        case RegexRepetition(ast, _) => checkEndAnchorNearNewline(ast)
         case _ =>
         // ignore
       }
     }
 
-    foo(regex)
+    checkEndAnchorNearNewline(regex)
 
     rewrite(regex, replacement, previous)
   }

diff --git a/tests/src/test/scala/com/nvidia/spark/rapids/RegularExpressionTranspilerSuite.scala b/tests/src/test/scala/com/nvidia/spark/rapids/RegularExpressionTranspilerSuite.scala
@@ -398,10 +398,11 @@ class RegularExpressionTranspilerSuite extends FunSuite with Arm {
   }
 
   test("fall back to CPU for newline next to line or string anchor") {
+    // these patterns were discovered during fuzz testing and resulted in different
+    // results between CPU and GPU
     val patterns = Seq(raw"\w[\r,B]\Z", raw"\s\Z\Z", "^$\\s", "$x*\\r", "$\\r")
     for (mode <- Seq(RegexFindMode, RegexReplaceMode)) {
       patterns.foreach(pattern => {
-        println(pattern)
         assertUnsupported(pattern, mode,
           "End of line/string anchor is not supported in this context")
       })