Support user taint analysis config

Author: mmvpm

utbot-framework/src/main/kotlin/org/utbot/engine/UtBotSymbolicEngine.kt

@@ -108,7 +108,8 @@ class UtBotSymbolicEngine(
     val mockStrategy: MockStrategy = NO_MOCKS,
     chosenClassesToMockAlways: Set<ClassId>,
     applicationContext: ApplicationContext,
-    private val solverTimeoutInMillis: Int = checkSolverTimeoutMillis
+    private val solverTimeoutInMillis: Int = checkSolverTimeoutMillis,
+    taintUserConfigPath: String? = null,
 ) : UtContextInitializer() {
     private val graph = methodUnderTest.sootMethod.jimpleBody().apply {
         logger.trace { "JIMPLE for $methodUnderTest:\n$this" }
@@ -144,7 +145,10 @@ class UtBotSymbolicEngine(
     private val taintMarkRegistry: TaintMarkRegistry = TaintMarkRegistry()
     private val taintMarkManager: TaintMarkManager = TaintMarkManager(taintMarkRegistry)
 
-    private val taintConfigProvider: TaintConfigurationProvider = TaintConfigurationProviderResources("taint/config.yaml")
+    private val taintConfigProvider: TaintConfigurationProvider = TaintConfigurationProviderCombiner(listOf(
+        TaintConfigurationProviderUserRules(taintUserConfigPath ?: ""),
+        TaintConfigurationProviderResources("taint/config.yaml")
+    ))
     private val taintConfiguration: TaintConfiguration = taintConfigProvider.getConfiguration()
 
     private val taintContext: TaintContext = TaintContext(taintMarkManager, taintConfiguration)

utbot-framework/src/main/kotlin/org/utbot/framework/plugin/api/TestCaseGenerator.kt

@@ -135,6 +135,7 @@ open class TestCaseGenerator(
         mockStrategy: MockStrategyApi,
         chosenClassesToMockAlways: Set<ClassId> = Mocker.javaDefaultClasses.mapTo(mutableSetOf()) { it.id },
         methodsGenerationTimeout: Long = utBotGenerationTimeoutInMillis,
+        taintConfigPath: String? = null,
         generate: (engine: UtBotSymbolicEngine) -> Flow<UtResult> = defaultTestFlow(methodsGenerationTimeout)
     ): List<UtMethodTestSet> {
         if (isCanceled()) return methods.map { UtMethodTestSet(it) }
@@ -168,7 +169,8 @@ open class TestCaseGenerator(
                                 mockStrategy,
                                 chosenClassesToMockAlways,
                                 applicationContext,
-                                executionTimeEstimator
+                                executionTimeEstimator,
+                                taintConfigPath
                             )
 
                             engineActions.map { engine.apply(it) }
@@ -258,7 +260,8 @@ open class TestCaseGenerator(
         mockStrategyApi: MockStrategyApi,
         chosenClassesToMockAlways: Set<ClassId>,
         applicationContext: ApplicationContext,
-        executionTimeEstimator: ExecutionTimeEstimator
+        executionTimeEstimator: ExecutionTimeEstimator,
+        taintConfigPath: String? = null,
     ): UtBotSymbolicEngine {
         logger.debug("Starting symbolic execution for $method  --$mockStrategyApi--")
         return UtBotSymbolicEngine(
@@ -269,7 +272,8 @@ open class TestCaseGenerator(
             mockStrategy = mockStrategyApi.toModel(),
             chosenClassesToMockAlways = chosenClassesToMockAlways,
             applicationContext = applicationContext,
-            solverTimeoutInMillis = executionTimeEstimator.updatedSolverCheckTimeoutMillis
+            solverTimeoutInMillis = executionTimeEstimator.updatedSolverCheckTimeoutMillis,
+            taintUserConfigPath = taintConfigPath,
         )
     }
 

utbot-framework/src/main/kotlin/org/utbot/framework/process/EngineProcessMain.kt

@@ -113,6 +113,7 @@ private fun EngineProcessModel.setup(kryoHelper: KryoHelper, watchdog: IdleWatch
                 MockStrategyApi.valueOf(params.mockStrategy),
                 kryoHelper.readObject(params.chosenClassesToMockAlways),
                 params.timeout,
+                params.taintConfigPath,
                 generate = generateFlow,
             )
                 .summarizeAll(Paths.get(params.searchDirectory), null)

utbot-framework/src/main/kotlin/org/utbot/framework/process/generated/EngineProcessModel.Generated.kt

@@ -72,7 +72,7 @@ class EngineProcessModel private constructor(
         }
 
 
-        const val serializationHash = -120710112541549600L
+        const val serializationHash = 1663132864271860294L
 
     }
     override val serializersOwner: ISerializersOwner get() = EngineProcessModel
@@ -173,7 +173,7 @@ val IProtocol.engineProcessModel get() = getOrCreateExtension(EngineProcessModel
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:98]
+ * #### Generated from [EngineProcessModel.kt:100]
  */
 data class FindMethodParamNamesArguments (
     val classId: ByteArray,
@@ -236,7 +236,7 @@ data class FindMethodParamNamesArguments (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:102]
+ * #### Generated from [EngineProcessModel.kt:104]
  */
 data class FindMethodParamNamesResult (
     val paramNames: ByteArray
@@ -293,7 +293,7 @@ data class FindMethodParamNamesResult (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:91]
+ * #### Generated from [EngineProcessModel.kt:93]
  */
 data class FindMethodsInClassMatchingSelectedArguments (
     val classId: ByteArray,
@@ -356,7 +356,7 @@ data class FindMethodsInClassMatchingSelectedArguments (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:95]
+ * #### Generated from [EngineProcessModel.kt:97]
  */
 data class FindMethodsInClassMatchingSelectedResult (
     val executableIds: ByteArray
@@ -424,7 +424,8 @@ data class GenerateParams (
     val isSymbolicEngineEnabled: Boolean,
     val isFuzzingEnabled: Boolean,
     val fuzzingValue: Double,
-    val searchDirectory: String
+    val searchDirectory: String,
+    val taintConfigPath: String?
 ) : IPrintable {
     //companion
 
@@ -442,7 +443,8 @@ data class GenerateParams (
             val isFuzzingEnabled = buffer.readBool()
             val fuzzingValue = buffer.readDouble()
             val searchDirectory = buffer.readString()
-            return GenerateParams(methods, mockStrategy, chosenClassesToMockAlways, timeout, generationTimeout, isSymbolicEngineEnabled, isFuzzingEnabled, fuzzingValue, searchDirectory)
+            val taintConfigPath = buffer.readNullable { buffer.readString() }
+            return GenerateParams(methods, mockStrategy, chosenClassesToMockAlways, timeout, generationTimeout, isSymbolicEngineEnabled, isFuzzingEnabled, fuzzingValue, searchDirectory, taintConfigPath)
         }
 
         override fun write(ctx: SerializationCtx, buffer: AbstractBuffer, value: GenerateParams)  {
@@ -455,6 +457,7 @@ data class GenerateParams (
             buffer.writeBool(value.isFuzzingEnabled)
             buffer.writeDouble(value.fuzzingValue)
             buffer.writeString(value.searchDirectory)
+            buffer.writeNullable(value.taintConfigPath) { buffer.writeString(it) }
         }
 
 
@@ -479,6 +482,7 @@ data class GenerateParams (
         if (isFuzzingEnabled != other.isFuzzingEnabled) return false
         if (fuzzingValue != other.fuzzingValue) return false
         if (searchDirectory != other.searchDirectory) return false
+        if (taintConfigPath != other.taintConfigPath) return false
 
         return true
     }
@@ -494,6 +498,7 @@ data class GenerateParams (
         __r = __r*31 + isFuzzingEnabled.hashCode()
         __r = __r*31 + fuzzingValue.hashCode()
         __r = __r*31 + searchDirectory.hashCode()
+        __r = __r*31 + if (taintConfigPath != null) taintConfigPath.hashCode() else 0
         return __r
     }
     //pretty print
@@ -509,6 +514,7 @@ data class GenerateParams (
             print("isFuzzingEnabled = "); isFuzzingEnabled.print(printer); println()
             print("fuzzingValue = "); fuzzingValue.print(printer); println()
             print("searchDirectory = "); searchDirectory.print(printer); println()
+            print("taintConfigPath = "); taintConfigPath.print(printer); println()
         }
         printer.print(")")
     }
@@ -518,7 +524,7 @@ data class GenerateParams (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:58]
+ * #### Generated from [EngineProcessModel.kt:60]
  */
 data class GenerateResult (
     val notEmptyCases: Int,
@@ -581,7 +587,7 @@ data class GenerateResult (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:110]
+ * #### Generated from [EngineProcessModel.kt:112]
  */
 data class GenerateTestReportArgs (
     val eventLogMessage: String?,
@@ -674,7 +680,7 @@ data class GenerateTestReportArgs (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:119]
+ * #### Generated from [EngineProcessModel.kt:121]
  */
 data class GenerateTestReportResult (
     val notifyMessage: String,
@@ -806,7 +812,7 @@ data class JdkInfo (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:86]
+ * #### Generated from [EngineProcessModel.kt:88]
  */
 data class MethodDescription (
     val name: String,
@@ -875,7 +881,7 @@ data class MethodDescription (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:62]
+ * #### Generated from [EngineProcessModel.kt:64]
  */
 data class RenderParams (
     val testSetsId: Long,
@@ -1016,7 +1022,7 @@ data class RenderParams (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:79]
+ * #### Generated from [EngineProcessModel.kt:81]
  */
 data class RenderResult (
     val generatedCode: String,
@@ -1079,7 +1085,7 @@ data class RenderResult (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:83]
+ * #### Generated from [EngineProcessModel.kt:85]
  */
 data class SetupContextParams (
     val classpathForUrlsClassloader: List<String>
@@ -1217,7 +1223,7 @@ data class TestGeneratorParams (
 
 
 /**
- * #### Generated from [EngineProcessModel.kt:105]
+ * #### Generated from [EngineProcessModel.kt:107]
  */
 data class WriteSarifReportArguments (
     val testSetsId: Long,

utbot-framework/src/main/kotlin/org/utbot/taint/TaintConfigurationProvider.kt

@@ -1,5 +1,6 @@
 package org.utbot.taint
 
+import org.utbot.common.PathUtil.toPath
 import org.utbot.taint.model.TaintConfiguration
 import org.utbot.taint.parser.TaintYamlParser
 
@@ -16,7 +17,6 @@ interface TaintConfigurationProvider {
  * @param configPath relative path to the .yaml file in resources
  */
 class TaintConfigurationProviderResources(private val configPath: String) : TaintConfigurationProvider {
-
     override fun getConfiguration(): TaintConfiguration {
         val configUrl = javaClass.classLoader.getResource(configPath)
             ?: return TaintConfiguration()
@@ -26,6 +26,28 @@ class TaintConfigurationProviderResources(private val configPath: String) : Tain
     }
 }
 
+/**
+ * Reads and parses [TaintConfiguration] from the [configPath].
+ *
+ * @param configPath relative path to the .yaml file in the user's project.
+ */
+class TaintConfigurationProviderUserRules(private val configPath: String) : TaintConfigurationProvider {
+    override fun getConfiguration(): TaintConfiguration {
+        val configFile = configPath.toPath().toFile()
+        if (!configFile.exists()) {
+            return TaintConfiguration()
+        }
+        val yamlInput = configFile.readText()
+        val dtoConfig = TaintYamlParser.parse(yamlInput)
+        return TaintConfigurationAdapter.convert(dtoConfig)
+    }
+}
+
+class TaintConfigurationProviderCombiner(private val inners: List<TaintConfigurationProvider>) : TaintConfigurationProvider {
+    override fun getConfiguration(): TaintConfiguration =
+        inners.fold(TaintConfiguration()) { resultConfig, configProvider ->
+            resultConfig + configProvider.getConfiguration()
+        }
+}
+
 // TODO: TaintConfigurationProviderCached -- stores binary config to the temp directory
-// TODO: TaintConfigurationProviderUserRules -- from .utbot/taint-config.yaml
-// TODO: TaintConfigurationProviderCombiner -- takes several providers and combines their results

utbot-framework/src/main/kotlin/org/utbot/taint/model/TaintConfiguration.kt

@@ -33,6 +33,14 @@ data class TaintConfiguration(
             equalParameters(executableId, it.signature) && equalMethodNames(executableId, it.methodFqn)
         }
 
+    operator fun plus(other: TaintConfiguration): TaintConfiguration =
+        TaintConfiguration(
+            sources = sources + other.sources,
+            passes = passes + other.passes,
+            cleaners = cleaners + other.cleaners,
+            sinks = sinks + other.sinks
+        )
+
     // internal
 
     private fun equalParameters(executableId: ExecutableId, signature: TaintSignature): Boolean =

utbot-intellij/src/main/kotlin/org/utbot/intellij/plugin/generator/UtTestsDialogProcessor.kt

@@ -12,6 +12,7 @@ import com.intellij.openapi.roots.OrderEnumerator
 import com.intellij.openapi.ui.Messages
 import com.intellij.openapi.util.Computable
 import com.intellij.openapi.util.text.StringUtil
+import com.intellij.project.stateStore
 import com.intellij.psi.PsiClass
 import com.intellij.psi.PsiMethod
 import com.intellij.refactoring.util.classMembers.MemberInfo
@@ -268,6 +269,8 @@ object UtTestsDialogProcessor {
                                     }
                                     .executeSynchronously()
 
+                                val taintConfigPath = getTaintConfigPath(project)
+
                                 withStaticsSubstitutionRequired(true) {
                                     val startTime = System.currentTimeMillis()
                                     val timerHandler =
@@ -293,7 +296,8 @@ object UtTestsDialogProcessor {
                                             true,
                                             UtSettings.useFuzzing,
                                             project.service<Settings>().fuzzingValue,
-                                            searchDirectory.pathString
+                                            searchDirectory.pathString,
+                                            taintConfigPath?.pathString
                                         )
 
                                         if (rdGenerateResult.notEmptyCases == 0) {
@@ -347,6 +351,14 @@ object UtTestsDialogProcessor {
         }
     }
 
+    /**
+     * Returns "{project}/.idea/utbot-taint-config.yaml" or null if it does not exists
+     */
+    private fun getTaintConfigPath(project: Project): Path? {
+        val path = project.stateStore.directoryStorePath?.resolve("utbot-taint-config.yaml")
+        return if (path != null && path.toFile().exists()) path else null
+    }
+
     private val PsiClass.canonicalName: String
     /*
     This method calculates exactly name that is used by compiler convention,

utbot-intellij/src/main/kotlin/org/utbot/intellij/plugin/process/EngineProcess.kt

@@ -256,7 +256,8 @@ class EngineProcess private constructor(val project: Project, private val classN
         isSymbolicEngineEnabled: Boolean,
         isFuzzingEnabled: Boolean,
         fuzzingValue: Double,
-        searchDirectory: String
+        searchDirectory: String,
+        taintConfigPath: String?
     ): RdTestGenerationResult {
         assertReadAccessNotAllowed()
         val params = GenerateParams(
@@ -268,7 +269,8 @@ class EngineProcess private constructor(val project: Project, private val classN
             isSymbolicEngineEnabled,
             isFuzzingEnabled,
             fuzzingValue,
-            searchDirectory
+            searchDirectory,
+            taintConfigPath
         )
         val result = engineModel.generate.startBlocking(params)
 

utbot-rd/src/main/rdgen/org/utbot/rd/models/EngineProcessModel.kt

@@ -54,6 +54,8 @@ object EngineProcessModel : Ext(EngineProcessRoot) {
         field("fuzzingValue", PredefinedType.double)
         // method filters
         field("searchDirectory", PredefinedType.string)
+        // taint analysis
+        field("taintConfigPath", PredefinedType.string.nullable)
     }
     val generateResult = structdef {
         field("notEmptyCases", PredefinedType.int)