[semanticcpg] feat: add validation level

joern-cli/frontends/c2cpg/src/test/scala/io/joern/c2cpg/testfixtures/AstC2CpgFrontend.scala

@@ -11,7 +11,7 @@ import io.joern.x2cpg.ValidationMode
 import io.joern.x2cpg.X2Cpg.newEmptyCpg
 import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.semanticcpg.utils.FileUtil
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 trait AstC2CpgFrontend extends LanguageFrontend {
   override type ConfigType = Config
@@ -49,7 +49,7 @@ trait AstC2CpgFrontend extends LanguageFrontend {
       )
     headerPass.createAndApply()
     new FunctionDeclNodePass(cpg, headerPass.unhandledMethodDeclarations(), config).createAndApply()
-    new PostFrontendValidator(cpg, false).run()
+    new PostFrontendValidator(cpg, ValidationLevel.V0).run()
     cpg
   }
 }

joern-cli/frontends/c2cpg/src/test/scala/io/joern/c2cpg/testfixtures/C2CpgFrontend.scala

@@ -5,7 +5,7 @@ import io.joern.c2cpg.Config
 import io.joern.x2cpg.testfixtures.LanguageFrontend
 import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.semanticcpg.utils.FileUtil
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 trait C2CpgFrontend extends LanguageFrontend {
   override type ConfigType = Config
@@ -16,7 +16,7 @@ trait C2CpgFrontend extends LanguageFrontend {
       .fold(Config())(_.asInstanceOf[Config])
       .withInputPath(sourceCodePath.getAbsolutePath)
     val res = c2cpg.createCpg(config).get
-    new PostFrontendValidator(res, false).run()
+    new PostFrontendValidator(res, ValidationLevel.V0).run()
     res
   }
 }

joern-cli/frontends/csharpsrc2cpg/src/test/scala/io/joern/csharpsrc2cpg/testfixtures/CSharpCode2CpgFixture.scala

@@ -11,7 +11,7 @@ import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.semanticcpg.utils.FileUtil
 import io.shiftleft.semanticcpg.language.{ICallResolver, NoResolve}
 import io.shiftleft.semanticcpg.layers.LayerCreatorContext
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 import org.scalatest.Inside
 
 import java.io.File
@@ -87,7 +87,7 @@ trait CSharpFrontend extends LanguageFrontend {
   override def execute(sourceCodeFile: File): Cpg = {
     val config = defaultConfig.withInputPath(sourceCodeFile.getAbsolutePath)
     val tmp    = new CSharpSrc2Cpg().createCpg(config).get
-    new PostFrontendValidator(tmp, false).createAndApply()
+    new PostFrontendValidator(tmp, ValidationLevel.V0).createAndApply()
     tmp
   }
 

joern-cli/frontends/ghidra2cpg/src/test/scala/io/joern/ghidra2cpg/fixtures/BinToCpgFixture.scala

@@ -3,7 +3,7 @@ package io.joern.ghidra2cpg.fixtures
 import io.joern.x2cpg.X2Cpg.applyDefaultOverlays
 import io.joern.x2cpg.testfixtures.LanguageFrontend
 import io.shiftleft.codepropertygraph.generated.Cpg
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 import io.shiftleft.utils.ProjectRoot
 import org.scalatest.BeforeAndAfterAll
 import org.scalatest.matchers.should.Matchers
@@ -24,7 +24,7 @@ class BinToCpgFixture(val frontend: LanguageFrontend) extends AnyWordSpec with M
     val bin = new File(binDirectory, binName)
     cpg = frontend.execute(bin)
     passes(cpg)
-    new PostFrontendValidator(cpg, false).run()
+    new PostFrontendValidator(cpg, ValidationLevel.V0).run()
   }
 
   override def afterAll(): Unit = {

joern-cli/frontends/ghidra2cpg/src/test/scala/io/joern/ghidra2cpg/fixtures/GhidraBinToCpgSuite.scala

@@ -8,7 +8,7 @@ import org.apache.commons.io.FileUtils
 import io.shiftleft.codepropertygraph.generated.nodes
 import io.joern.dataflowengineoss.language.*
 import io.shiftleft.semanticcpg.language.*
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 import java.nio.file.Files
 
@@ -30,7 +30,7 @@ class GhidraFrontend extends LanguageFrontend {
     val cpgBin = dir.getAbsolutePath
     val config = Config().withInputPath(inputFile.getAbsolutePath).withOutputPath(cpgBin)
     val cpg    = new Ghidra2Cpg().createCpg(config).get
-    new PostFrontendValidator(cpg, false).run()
+    new PostFrontendValidator(cpg, ValidationLevel.V0).run()
     cpg
   }
 

joern-cli/frontends/gosrc2cpg/src/test/scala/io/joern/go2cpg/testfixtures/GoCodeToCpgSuite.scala

@@ -10,7 +10,7 @@ import io.joern.x2cpg.testfixtures.{Code2CpgFixture, DefaultTestCpg}
 import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.semanticcpg.language.{ICallResolver, NoResolve}
 import io.shiftleft.semanticcpg.utils.FileUtil
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 import org.scalatest.Inside
 class DefaultTestCpgWithGo(val fileSuffix: String) extends DefaultTestCpg with SemanticTestCpg {
   override type ConfigType = Config
@@ -41,7 +41,7 @@ class DefaultTestCpgWithGo(val fileSuffix: String) extends DefaultTestCpg with S
       .getOrElse(Config())
       .withInputPath(sourceCodePath.getAbsolutePath)
     val res = goSrc2Cpg.get.createCpg(config).get
-    new PostFrontendValidator(res, false).run()
+    new PostFrontendValidator(res, ValidationLevel.V0).run()
     res
   }
 

joern-cli/frontends/javasrc2cpg/src/test/scala/io/joern/javasrc2cpg/testfixtures/JavaSrcCodeToCpgFixture.scala

@@ -13,7 +13,7 @@ import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.codepropertygraph.generated.nodes.{Expression, Literal}
 import io.shiftleft.semanticcpg.language.*
 import io.shiftleft.semanticcpg.utils.FileUtil
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 import java.io.File
 import java.nio.file.Path as JavaPath
@@ -37,7 +37,7 @@ trait JavaSrcFrontend extends LanguageFrontend {
       .getOrElse(JavaSrc2Cpg.DefaultConfig.withDelombokMode("no-delombok"))
       .withCacheJdkTypeSolver(true)
     val res = new JavaSrc2Cpg().createCpg(config.withInputPath(sourceCodeFile.getAbsolutePath)).get
-    new PostFrontendValidator(res, false).run()
+    new PostFrontendValidator(res, ValidationLevel.V0).run()
     res
   }
 }

joern-cli/frontends/jimple2cpg/src/test/scala/io/joern/jimple2cpg/testfixtures/JimpleCodeToCpgFixture.scala

@@ -8,7 +8,7 @@ import io.joern.x2cpg.X2Cpg
 import io.joern.x2cpg.testfixtures.{Code2CpgFixture, DefaultTestCpg, LanguageFrontend, TestCpg}
 import io.shiftleft.semanticcpg.utils.FileUtil.*
 import io.shiftleft.codepropertygraph.generated.Cpg
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 import java.io.File
 import java.nio.file.Path
@@ -24,7 +24,7 @@ trait Jimple2CpgFrontend extends LanguageFrontend {
   override def execute(sourceCodeFile: File): Cpg = {
     val config = getConfig().getOrElse(Config())
     val cpg    = new Jimple2Cpg().createCpg(config.withInputPath(sourceCodeFile.getAbsolutePath)).get
-    new PostFrontendValidator(cpg, false).run()
+    new PostFrontendValidator(cpg, ValidationLevel.V0).run()
     cpg
   }
 }

joern-cli/frontends/jssrc2cpg/src/test/scala/io/joern/jssrc2cpg/testfixtures/JsSrc2CpgFrontend.scala

@@ -3,7 +3,7 @@ package io.joern.jssrc2cpg.testfixtures
 import io.joern.jssrc2cpg.{Config, JsSrc2Cpg}
 import io.joern.x2cpg.testfixtures.LanguageFrontend
 import io.shiftleft.codepropertygraph.generated.Cpg
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 trait JsSrc2CpgFrontend extends LanguageFrontend {
   override type ConfigType = Config
@@ -12,7 +12,7 @@ trait JsSrc2CpgFrontend extends LanguageFrontend {
     val jssrc2cpg = new JsSrc2Cpg()
     val config    = getConfig().getOrElse(Config(tsTypes = false)).withInputPath(sourceCodePath.getAbsolutePath)
     val res       = jssrc2cpg.createCpg(config).get
-    new PostFrontendValidator(res, true).run()
+    new PostFrontendValidator(res, ValidationLevel.V1).run()
     res
   }
 }

joern-cli/frontends/kotlin2cpg/src/test/scala/io/joern/kotlin2cpg/testfixtures/KotlinCodeToCpgFixture.scala

@@ -12,7 +12,7 @@ import io.joern.x2cpg.testfixtures.DefaultTestCpg
 import io.joern.x2cpg.testfixtures.LanguageFrontend
 import io.shiftleft.semanticcpg.utils.FileUtil.*
 import io.shiftleft.codepropertygraph.generated.Cpg
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 import io.shiftleft.utils.ProjectRoot
 
 import java.io.File
@@ -41,7 +41,7 @@ trait KotlinFrontend extends LanguageFrontend {
     }
 
     val tmp = new Kotlin2Cpg().createCpg(config.withInputPath(sourceCodeFile.getAbsolutePath)).get
-    new PostFrontendValidator(tmp, true).run()
+    new PostFrontendValidator(tmp, ValidationLevel.V1).run()
     tmp
   }
 }

joern-cli/frontends/php2cpg/src/test/scala/io/joern/php2cpg/testfixtures/PhpCode2CpgFixture.scala

@@ -8,7 +8,7 @@ import io.joern.x2cpg.frontendspecific.php2cpg
 import io.joern.x2cpg.testfixtures.{Code2CpgFixture, DefaultTestCpg, LanguageFrontend}
 import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.semanticcpg.language.{ICallResolver, NoResolve}
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 import java.io.File
 
@@ -20,7 +20,7 @@ trait PhpFrontend extends LanguageFrontend {
   override def execute(sourceCodeFile: File): Cpg = {
     val defaultConfig: Config = getConfig().getOrElse(Config())
     val cpg                   = new Php2Cpg().createCpg(defaultConfig.withInputPath(sourceCodeFile.getAbsolutePath)).get
-    new PostFrontendValidator(cpg, true).run()
+    new PostFrontendValidator(cpg, ValidationLevel.V1).run()
     cpg
   }
 }

joern-cli/frontends/pysrc2cpg/src/test/scala/io/joern/pysrc2cpg/testfixtures/PySrc2CpgFixture.scala

@@ -22,7 +22,7 @@ import io.joern.x2cpg.testfixtures.LanguageFrontend
 import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.semanticcpg.language.ICallResolver
 import io.shiftleft.semanticcpg.language.NoResolve
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 trait PythonFrontend extends LanguageFrontend {
   override type ConfigType = Py2CpgOnFileSystemConfig
@@ -41,7 +41,7 @@ trait PythonFrontend extends LanguageFrontend {
           .withInputPath(sourceCodePath.getAbsolutePath)
       )
       .get
-    new PostFrontendValidator(tmp, false).run()
+    new PostFrontendValidator(tmp, ValidationLevel.V0).run()
     tmp
   }
 }

joern-cli/frontends/rubysrc2cpg/src/test/scala/io/joern/rubysrc2cpg/testfixtures/RubyCode2CpgFixture.scala

@@ -9,7 +9,7 @@ import io.joern.x2cpg.ValidationMode
 import io.joern.x2cpg.testfixtures.*
 import io.shiftleft.codepropertygraph.generated.Cpg
 import io.shiftleft.semanticcpg.language.{ICallResolver, NoResolve}
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 import org.scalatest.Inside
 
 import java.io.File
@@ -28,7 +28,7 @@ trait RubyFrontend(withDownloadDependencies: Boolean, disableFileContent: Boolea
 
   override def execute(sourceCodeFile: File): Cpg = {
     val tmp = Using.resource(new RubySrc2Cpg())(_.createCpg(config.withInputPath(sourceCodeFile.getAbsolutePath)).get)
-    new PostFrontendValidator(tmp, false).run()
+    new PostFrontendValidator(tmp, ValidationLevel.V0).run()
     tmp
   }
 

joern-cli/frontends/swiftsrc2cpg/src/test/scala/io/joern/swiftsrc2cpg/testfixtures/SwiftCompilerSrc2CpgFrontend.scala

@@ -3,7 +3,7 @@ package io.joern.swiftsrc2cpg.testfixtures
 import io.joern.swiftsrc2cpg.{Config, SwiftSrc2Cpg}
 import io.joern.x2cpg.testfixtures.LanguageFrontend
 import io.shiftleft.codepropertygraph.generated.Cpg
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 trait SwiftCompilerSrc2CpgFrontend extends LanguageFrontend {
   final override type ConfigType = Config
@@ -12,7 +12,7 @@ trait SwiftCompilerSrc2CpgFrontend extends LanguageFrontend {
     val pathAsString = sourceCodePath.toPath.resolve("SwiftTest").toAbsolutePath.toString
     val config       = getConfig().getOrElse(Config()).withInputPath(pathAsString)
     val cpg          = new SwiftSrc2Cpg().createCpg(config).get
-    new PostFrontendValidator(cpg, true).run()
+    new PostFrontendValidator(cpg, ValidationLevel.V1).run()
     cpg
   }
 

joern-cli/frontends/swiftsrc2cpg/src/test/scala/io/joern/swiftsrc2cpg/testfixtures/SwiftSrc2CpgFrontend.scala

@@ -3,7 +3,7 @@ package io.joern.swiftsrc2cpg.testfixtures
 import io.joern.swiftsrc2cpg.{Config, SwiftSrc2Cpg}
 import io.joern.x2cpg.testfixtures.LanguageFrontend
 import io.shiftleft.codepropertygraph.generated.Cpg
-import io.shiftleft.semanticcpg.validation.PostFrontendValidator
+import io.shiftleft.semanticcpg.validation.{PostFrontendValidator, ValidationLevel}
 
 trait SwiftSrc2CpgFrontend extends LanguageFrontend {
   override type ConfigType = Config
@@ -12,7 +12,7 @@ trait SwiftSrc2CpgFrontend extends LanguageFrontend {
     val pathAsString = sourceCodePath.getAbsolutePath
     val config       = getConfig().getOrElse(Config()).withInputPath(pathAsString)
     val tmp          = new SwiftSrc2Cpg().createCpg(config).get
-    new PostFrontendValidator(tmp, true).run()
+    new PostFrontendValidator(tmp, ValidationLevel.V1).run()
     tmp
   }
 }

semanticcpg/src/main/scala/io/shiftleft/semanticcpg/validation/validation.scala

@@ -11,6 +11,13 @@ import org.slf4j.{Logger, LoggerFactory}
 
 class ValidationError(msg: String) extends RuntimeException(msg) {}
 
+enum ValidationLevel(val value: Int) extends Ordered[ValidationLevel] {
+  case V0 extends ValidationLevel(0)
+  case V1 extends ValidationLevel(1)
+
+  override def compare(that: ValidationLevel): Int = this.value.compareTo(that.value)
+}
+
 /** Base validator pass that collects validation violations during graph traversal.
   *
   * @param cpg
@@ -26,16 +33,16 @@ abstract class AbstractValidator(cpg: Cpg) extends CpgPass(cpg) {
   }
 
   /** @return
-    *   compressed violations as (message, count) pairs in stable key order
+    *   compressed violations as (key, message, count) tuples in stable key order
     */
-  def getViolationsCompressed: mutable.ArrayBuffer[(String, Int)] = {
+  def getViolationsCompressedWithKey: mutable.ArrayBuffer[(Any, String, Int)] = {
     mutable.ArrayBuffer.from(
       violationsBuffer.iterator
         .groupByStable(_._1)
         .iterator
-        .map { (_, violations) => (violations.head._2, violations.size) }
+        .map { case (key, violations) => (key, violations.head._2, violations.size) }
         .toSeq
-        .sortBy(_._1)
+        .sortBy(_._1.toString)
     )
   }
 }
@@ -44,9 +51,17 @@ abstract class AbstractValidator(cpg: Cpg) extends CpgPass(cpg) {
 object PostFrontendValidator {
   private val logger: Logger = LoggerFactory.getLogger(getClass)
 
-  enum ErrorType {
-    case FULLNAME_UNIQUE_METHOD, FULLNAME_UNIQUE_TYPE, FULLNAME_UNIQUE_TYPEDECL, MULTI_REF, BAD_REF_TYPE, NONLOCAL_REF,
-      MULTI_AST_IN, MULTI_ARG_IN, DUPLICATE_ORDER
+  /** Newly added error types should have an `addedInLevel` one level higher than the highest */
+  enum ErrorType(val addedInLevel: ValidationLevel) {
+    case FULLNAME_UNIQUE_METHOD   extends ErrorType(ValidationLevel.V1)
+    case FULLNAME_UNIQUE_TYPE     extends ErrorType(ValidationLevel.V1)
+    case FULLNAME_UNIQUE_TYPEDECL extends ErrorType(ValidationLevel.V1)
+    case MULTI_REF                extends ErrorType(ValidationLevel.V1)
+    case BAD_REF_TYPE             extends ErrorType(ValidationLevel.V1)
+    case NONLOCAL_REF             extends ErrorType(ValidationLevel.V1)
+    case MULTI_AST_IN             extends ErrorType(ValidationLevel.V1)
+    case MULTI_ARG_IN             extends ErrorType(ValidationLevel.V1)
+    case DUPLICATE_ORDER          extends ErrorType(ValidationLevel.V1)
   }
 }
 
@@ -55,8 +70,11 @@ object PostFrontendValidator {
   *
   * Goal is to improve the pass once we fixed current violations, and then have new ideas what to check. Then plug in
   * faster checking code, then enable in sptests and prod.
+  *
+  * NOTE: All validation checks with a level lower or equal to [[fatalValidationLevel]] will result in an exception. See
+  * [[ValidationLevel]] and [[ErrorType]] for the highest validation level.
   */
-class PostFrontendValidator(cpg: Cpg, throwOnError: Boolean) extends AbstractValidator(cpg) {
+class PostFrontendValidator(cpg: Cpg, fatalValidationLevel: ValidationLevel) extends AbstractValidator(cpg) {
   import PostFrontendValidator.logger
   import PostFrontendValidator.ErrorType.*
 
@@ -221,11 +239,22 @@ class PostFrontendValidator(cpg: Cpg, throwOnError: Boolean) extends AbstractVal
 
   /** Log compressed violations and optionally throw on error. */
   private def reportViolations(): Unit = {
-    val violations = getViolationsCompressed
+    val violations = getViolationsCompressedWithKey
     if (violations.nonEmpty) {
-      val err = violations.map { (msg, n) => s"$msg ($n instances)" }.mkString("\n")
+      val err = violations.map { (_, msg, n) => s"$msg ($n instances)" }.mkString("\n")
       logger.warn(s"Graph validation failure:\n$err")
-      if (throwOnError) throw new ValidationError(err)
+
+      val fatalViolations = violations.filter { case (key, _, _) =>
+        key match {
+          case err: ErrorType => err.addedInLevel <= fatalValidationLevel
+          case _              => true
+        }
+      }
+
+      if (fatalViolations.nonEmpty) {
+        val fatalErr = fatalViolations.map { case (_, msg, n) => s"$msg ($n instances)" }.mkString("\n")
+        throw new ValidationError(s"Fatal graph validation errors (Level <= $fatalValidationLevel):\n$fatalErr")
+      }
     }
   }