[ruby] loop Do Block & Break Expressions (#4524)

* Handles `break` statements on an expression level
* Lowering `loop` do-block statements as `do-while` loops
* Fixed missing transform match & `loop` block handling

Resolves #4508

joern-cli/frontends/rubysrc2cpg/src/main/scala/io/joern/rubysrc2cpg/astcreation/AstForExpressionsCreator.scala

@@ -49,6 +49,7 @@ trait AstForExpressionsCreator(implicit withSchemaValidation: ValidationMode) {
     case node: ProcOrLambdaExpr         => astForProcOrLambdaExpr(node)
     case node: RubyCallWithBlock[_]     => astsForCallWithBlockInExpr(node)
     case node: SelfIdentifier           => astForSelfIdentifier(node)
+    case node: BreakStatement           => astForBreakStatement(node)
     case node: StatementList            => astForStatementList(node)
     case node: DummyNode                => Ast(node.node)
     case _                              => astForUnknown(node)

joern-cli/frontends/rubysrc2cpg/src/main/scala/io/joern/rubysrc2cpg/astcreation/AstForStatementsCreator.scala

@@ -319,7 +319,7 @@ trait AstForStatementsCreator(implicit withSchemaValidation: ValidationMode) { t
     returnAst(returnNode(node, code(node)), List(astForMemberCall(node)))
   }
 
-  private def astForBreakStatement(node: BreakStatement): Ast = {
+  protected def astForBreakStatement(node: BreakStatement): Ast = {
     val _node = NewControlStructure()
       .controlStructureType(ControlStructureTypes.BREAK)
       .lineNumber(line(node))
@@ -386,8 +386,9 @@ trait AstForStatementsCreator(implicit withSchemaValidation: ValidationMode) { t
           elseClause.map(transform).orElse(defaultElseBranch(node.span)),
           ensureClause
         )(node.span)
-      case WhileExpression(condition, body) => WhileExpression(condition, transform(body))(node.span)
-      case UntilExpression(condition, body) => UntilExpression(condition, transform(body))(node.span)
+      case WhileExpression(condition, body)   => WhileExpression(condition, transform(body))(node.span)
+      case DoWhileExpression(condition, body) => DoWhileExpression(condition, transform(body))(node.span)
+      case UntilExpression(condition, body)   => UntilExpression(condition, transform(body))(node.span)
       case IfExpression(condition, thenClause, elsifClauses, elseClause) =>
         IfExpression(
           condition,

joern-cli/frontends/rubysrc2cpg/src/main/scala/io/joern/rubysrc2cpg/astcreation/RubyIntermediateAst.scala

@@ -3,6 +3,8 @@ package io.joern.rubysrc2cpg.astcreation
 import io.joern.rubysrc2cpg.passes.Defines
 import io.shiftleft.codepropertygraph.generated.nodes.NewNode
 
+import scala.annotation.tailrec
+
 object RubyIntermediateAst {
 
   case class TextSpan(

joern-cli/frontends/rubysrc2cpg/src/main/scala/io/joern/rubysrc2cpg/parser/RubyNodeCreator.scala

@@ -550,6 +550,19 @@ class RubyNodeCreator extends RubyParserBaseVisitor[RubyNode] {
   override def visitMethodCallWithBlockExpression(ctx: RubyParser.MethodCallWithBlockExpressionContext): RubyNode = {
     ctx.methodIdentifier().getText match {
       case Defines.Proc | Defines.Lambda => ProcOrLambdaExpr(visit(ctx.block()).asInstanceOf[Block])(ctx.toTextSpan)
+      case Defines.Loop =>
+        DoWhileExpression(
+          SimpleIdentifier(Option(Defines.getBuiltInType(Defines.TrueClass)))(
+            ctx.methodIdentifier().toTextSpan.spanStart("true")
+          ),
+          ctx.block() match {
+            case b: RubyParser.DoBlockBlockContext =>
+              visit(b.doBlock().bodyStatement())
+            case y =>
+              logger.warn(s"Unexpected loop block body ${y.getClass}")
+              visit(ctx.block())
+          }
+        )(ctx.toTextSpan)
       case _ =>
         SimpleCallWithBlock(visit(ctx.methodIdentifier()), List(), visit(ctx.block()).asInstanceOf[Block])(
           ctx.toTextSpan

joern-cli/frontends/rubysrc2cpg/src/main/scala/io/joern/rubysrc2cpg/passes/Defines.scala

@@ -22,6 +22,7 @@ object Defines {
   val Lambda: String     = "lambda"
   val Proc: String       = "proc"
   val This: String       = "this"
+  val Loop: String       = "loop"
 
   val Program: String = ":program"
 

joern-cli/frontends/rubysrc2cpg/src/test/scala/io/joern/rubysrc2cpg/dataflow/MethodTests.scala

@@ -25,8 +25,7 @@ class MethodTests extends RubyCode2CpgFixture(withPostProcessing = true, withDat
     sink.reachableByFlows(src).l.size shouldBe 2
   }
 
-  // Works in deprecated
-  "Data flow through do-while loop" ignore {
+  "Data flow through do-while loop" in {
     val cpg = code("""
                      |x = 0
                      |num = -1
@@ -42,7 +41,7 @@ class MethodTests extends RubyCode2CpgFixture(withPostProcessing = true, withDat
 
     val source = cpg.identifier.name("x").l
     val sink   = cpg.call.name("puts").l
-    sink.reachableByFlows(source).l.size shouldBe 2
+    sink.reachableByFlows(source).size shouldBe 5
   }
 
   "Data flow through methodOnlyIdentifier usage" in {

joern-cli/frontends/rubysrc2cpg/src/test/scala/io/joern/rubysrc2cpg/querying/ControlStructureTests.scala

@@ -76,6 +76,25 @@ class ControlStructureTests extends RubyCode2CpgFixture {
     assignment.lineNumber shouldBe Some(4)
   }
 
+  "a break expression nested in a control structure should be represented" in {
+    val cpg = code("""
+        |x = 0
+        |num  = -1
+        |loop do
+        |  num = x + 1
+        |  x = x + 1
+        |  if x > 10
+        |    break
+        |  end
+        |end
+        |puts num
+        |""".stripMargin)
+
+    val List(breakNode) = cpg.break.l
+    breakNode.code shouldBe "break"
+    breakNode.lineNumber shouldBe Some(8)
+  }
+
   "`if-end` statement is represented by an `IF` CONTROL_STRUCTURE node" in {
     val cpg = code("""
         |if __LINE__ > 1 then

joern-cli/frontends/rubysrc2cpg/src/test/scala/io/joern/rubysrc2cpg/querying/MethodTests.scala

@@ -419,7 +419,7 @@ class MethodTests extends RubyCode2CpgFixture {
   "break unless statement" should {
     val cpg = code("""
         |  def foo
-        |    loop do
+        |    bar do
         |      break unless 1 < 2
         |    end
         |  end