[SPARK-20871][SQL] log Janino code at info level (but include a max of 1000 lines)

pj.fanning · pj.fanning · commit 52b20f38f550 · 2017-07-17T23:14:44.000+01:00
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeFormatter.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeFormatter.scala
@@ -28,14 +28,22 @@ import java.util.regex.Matcher
 object CodeFormatter {
   val commentHolder = """\/\*(.+?)\*\/""".r
 
-  def format(code: CodeAndComment): String = {
+  def format(code: CodeAndComment): String = format(code, -1)
+
+  def format(code: CodeAndComment, maxLines: Int): String = {
     val formatter = new CodeFormatter
-    code.body.split("\n").foreach { line =>
+    val lines = code.body.split("\n")
+    val needToTruncate = maxLines >= 0 && lines.length > maxLines
+    val filteredLines = if (needToTruncate) lines.take(maxLines) else lines
+    filteredLines.foreach { line =>
       val commentReplaced = commentHolder.replaceAllIn(
         line.trim,
         m => code.comment.get(m.group(1)).map(Matcher.quoteReplacement).getOrElse(m.group(0)))
       formatter.addLine(commentReplaced)
     }
+    if (needToTruncate) {
+      formatter.addLine(s"[truncated to $maxLines lines (total lines is ${lines.length})]")
+    }
     formatter.result()
   }
 
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala
@@ -1050,10 +1050,12 @@ object CodeGenerator extends Logging {
       case e: JaninoRuntimeException =>
         val msg = s"failed to compile: $e"
         logError(msg, e)
+        logInfo(s"\n${CodeFormatter.format(code, 1000)}")
         throw new JaninoRuntimeException(msg, e)
       case e: CompileException =>
         val msg = s"failed to compile: $e"
         logError(msg, e)
+        logInfo(s"\n${CodeFormatter.format(code, 1000)}")
         throw new CompileException(msg, e.getLocation)
     }
     evaluator.getClazz().newInstance().asInstanceOf[GeneratedClass]
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeFormatterSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeFormatterSuite.scala
@@ -20,18 +20,18 @@ package org.apache.spark.sql.catalyst.expressions.codegen
 import org.apache.spark.SparkFunSuite
 import org.apache.spark.sql.catalyst.util._
 
-
 class CodeFormatterSuite extends SparkFunSuite {
 
-  def testCase(name: String)(
-      input: String, comment: Map[String, String] = Map.empty)(expected: String): Unit = {
+  def testCase(name: String)(input: String,
+      comment: Map[String, String] = Map.empty, maxLines: Int = -1)(expected: String): Unit = {
     test(name) {
       val sourceCode = new CodeAndComment(input.trim, comment)
-      if (CodeFormatter.format(sourceCode).trim !== expected.trim) {
+      if (CodeFormatter.format(sourceCode, maxLines).trim !== expected.trim) {
         fail(
           s"""
              |== FAIL: Formatted code doesn't match ===
-             |${sideBySide(CodeFormatter.format(sourceCode).trim, expected.trim).mkString("\n")}
+             |${sideBySide(CodeFormatter.format(sourceCode, maxLines).trim,
+                 expected.trim).mkString("\n")}
            """.stripMargin)
       }
     }
@@ -129,6 +129,22 @@ class CodeFormatterSuite extends SparkFunSuite {
     """.stripMargin
   }
 
+  testCase("function calls with maxLines=2") (
+    """
+      |foo(
+      |a,
+      |b,
+      |c)
+    """.stripMargin,
+    maxLines = 2
+  ) {
+    """
+      |/* 001 */ foo(
+      |/* 002 */   a,
+      |/* 003 */   [truncated to 2 lines (total lines is 4)]
+    """.stripMargin
+  }
+
   testCase("single line comments") {
     """
       |// This is a comment about class A { { { ( (
