Make language frontends configurable

README.md

@@ -52,7 +52,7 @@ import de.fraunhofer.aisec.cpg.TranslationManager;
 import de.fraunhofer.aisec.cpg.graph.declarations.FunctionDeclaration;
 
 var path = Paths.get("src/test/resources/openssl/client.cpp");
-var config = TranslationConfiguration.builder().sourceLocations(path.toFile()).defaultPasses().debugParser(true).build();
+var config = TranslationConfiguration.builder().sourceLocations(path.toFile()).defaultPasses().defaultLanguages().debugParser(true).build();
 var analyzer = TranslationManager.builder().config(config).build();
 var result = analyzer.analyze().get();
 var tu = result.getTranslationUnits().get(0);

build.gradle.kts

@@ -196,6 +196,9 @@ spotless {
         )
         googleJavaFormat()
     }
+    kotlin {
+        ktfmt().kotlinlangStyle()
+    }
 }
 
 if (project.hasProperty("experimental")) {

gradle.properties

@@ -0,0 +1 @@
+org.gradle.jvmargs=-Xmx4096m -XX:MaxMetaspaceSize=512m

src/main/java/de/fraunhofer/aisec/cpg/ExperimentalGolang.kt

@@ -1,7 +1,4 @@
 package de.fraunhofer.aisec.cpg
 
-/**
- * This annotation marks the use an experimental Golang language feature.
- */
-@RequiresOptIn
-annotation class ExperimentalGolang
+/** This annotation marks the use an experimental Golang language feature. */
+@RequiresOptIn annotation class ExperimentalGolang

src/main/java/de/fraunhofer/aisec/cpg/ExperimentalGraph.kt

@@ -1,7 +1,4 @@
 package de.fraunhofer.aisec.cpg
 
-/**
- * This annotation marks the use an experimental graph feature.
- */
-@RequiresOptIn
-annotation class ExperimentalGraph
+/** This annotation marks the use an experimental graph feature. */
+@RequiresOptIn annotation class ExperimentalGraph

src/main/java/de/fraunhofer/aisec/cpg/TranslationConfiguration.java

@@ -26,6 +26,14 @@
 
 package de.fraunhofer.aisec.cpg;
 
+import static de.fraunhofer.aisec.cpg.frontends.cpp.CXXLanguageFrontend.CXX_EXTENSIONS;
+import static de.fraunhofer.aisec.cpg.frontends.cpp.CXXLanguageFrontend.CXX_HEADER_EXTENSIONS;
+
+import com.google.common.collect.Iterables;
+import com.google.common.collect.Lists;
+import de.fraunhofer.aisec.cpg.frontends.LanguageFrontend;
+import de.fraunhofer.aisec.cpg.frontends.cpp.CXXLanguageFrontend;
+import de.fraunhofer.aisec.cpg.frontends.java.JavaLanguageFrontend;
 import de.fraunhofer.aisec.cpg.passes.CallResolver;
 import de.fraunhofer.aisec.cpg.passes.ControlFlowSensitiveDFGPass;
 import de.fraunhofer.aisec.cpg.passes.EvaluationOrderGraphPass;
@@ -87,6 +95,12 @@ public class TranslationConfiguration {
    */
   public final List<String> includeBlacklist;
 
+  /**
+   * This map contains a list of language frontends classes and the file types they are registered
+   * for.
+   */
+  private Map<Class<? extends LanguageFrontend>, List<String>> frontends;
+
   /**
    * Switch off cleaning up TypeManager memory after analysis.
    *
@@ -134,6 +148,7 @@ private TranslationConfiguration(
       List<String> includeWhitelist,
       List<String> includeBlacklist,
       List<Pass> passes,
+      Map<Class<? extends LanguageFrontend>, List<String>> frontends,
       boolean codeInNodes,
       boolean processAnnotations,
       boolean disableCleanup,
@@ -148,6 +163,7 @@ private TranslationConfiguration(
     this.includeWhitelist = includeWhitelist;
     this.includeBlacklist = includeBlacklist;
     this.passes = passes != null ? passes : new ArrayList<>();
+    this.frontends = frontends;
     // Make sure to init this AFTER sourceLocations has been set
     this.codeInNodes = codeInNodes;
     this.processAnnotations = processAnnotations;
@@ -175,6 +191,10 @@ public List<Pass> getRegisteredPasses() {
     return this.passes;
   }
 
+  public Map<Class<? extends LanguageFrontend>, List<String>> getFrontends() {
+    return this.frontends;
+  }
+
   /**
    * Builds a {@link TranslationConfiguration}.
    *
@@ -192,6 +212,7 @@ public List<Pass> getRegisteredPasses() {
    */
   public static class Builder {
     private List<File> sourceLocations = new ArrayList<>();
+    private Map<Class<? extends LanguageFrontend>, List<String>> frontends = new HashMap<>();
     private File topLevel = null;
     private boolean debugParser = false;
     private boolean failOnError = false;
@@ -322,6 +343,13 @@ public Builder registerPass(@NonNull Pass pass) {
       return this;
     }
 
+    /** Registers an additional {@link de.fraunhofer.aisec.cpg.frontends.LanguageFrontend}. */
+    public Builder registerLanguage(
+        @NonNull Class<? extends LanguageFrontend> frontend, List<String> fileTypes) {
+      this.frontends.put(frontend, fileTypes);
+      return this;
+    }
+
     /**
      * Register all default {@link Pass}es.
      *
@@ -354,6 +382,22 @@ public Builder defaultPasses() {
       return this;
     }
 
+    /**
+     * Register all default languages.
+     *
+     * @return
+     */
+    public Builder defaultLanguages() {
+      registerLanguage(
+          CXXLanguageFrontend.class,
+          Lists.newArrayList(Iterables.concat(CXX_EXTENSIONS, CXX_HEADER_EXTENSIONS)));
+      registerLanguage(JavaLanguageFrontend.class, JavaLanguageFrontend.JAVA_EXTENSIONS);
+
+      // do not register experimental languages by default until we have a release strategy
+      // registerLanguage(GoLanguageFrontend.class, GoLanguageFrontend.GOLANG_EXTENSIONS);
+      return this;
+    }
+
     public Builder codeInNodes(boolean b) {
       this.codeInNodes = b;
       return this;
@@ -391,6 +435,7 @@ public TranslationConfiguration build() {
           includeWhitelist,
           includeBlacklist,
           passes,
+          frontends,
           codeInNodes,
           processAnnotations,
           disableCleanup,

src/main/java/de/fraunhofer/aisec/cpg/TranslationManager.java

@@ -26,10 +26,9 @@
 
 package de.fraunhofer.aisec.cpg;
 
-import static de.fraunhofer.aisec.cpg.frontends.LanguageFrontendFactory.CXX_EXTENSIONS;
+import static de.fraunhofer.aisec.cpg.frontends.cpp.CXXLanguageFrontend.CXX_EXTENSIONS;
 
 import de.fraunhofer.aisec.cpg.frontends.LanguageFrontend;
-import de.fraunhofer.aisec.cpg.frontends.LanguageFrontendFactory;
 import de.fraunhofer.aisec.cpg.frontends.TranslationException;
 import de.fraunhofer.aisec.cpg.graph.TypeManager;
 import de.fraunhofer.aisec.cpg.helpers.Benchmark;
@@ -39,6 +38,7 @@
 import java.io.File;
 import java.io.IOException;
 import java.io.PrintWriter;
+import java.lang.reflect.InvocationTargetException;
 import java.nio.file.Files;
 import java.nio.file.Path;
 import java.util.ArrayList;
@@ -53,6 +53,7 @@
 import java.util.stream.Collectors;
 import java.util.stream.Stream;
 import org.checkerframework.checker.nullness.qual.NonNull;
+import org.checkerframework.checker.nullness.qual.Nullable;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
@@ -208,9 +209,7 @@ private HashSet<LanguageFrontend> runFrontends(
       log.info("Parsing {}", sourceLocation.getAbsolutePath());
       LanguageFrontend frontend = null;
       try {
-        frontend =
-            LanguageFrontendFactory.getFrontend(
-                Util.getExtension(sourceLocation), config, scopeManager);
+        frontend = getFrontend(Util.getExtension(sourceLocation), scopeManager);
         if (frontend == null) {
           log.error("Found no parser frontend for {}", sourceLocation.getName());
 
@@ -258,6 +257,32 @@ private HashSet<LanguageFrontend> runFrontends(
     return usedFrontends;
   }
 
+  @Nullable
+  private LanguageFrontend getFrontend(String extension, ScopeManager scopeManager) {
+    var clazz =
+        this.config.getFrontends().entrySet().stream()
+            .filter(entry -> entry.getValue().contains(extension))
+            .map(entry -> entry.getKey())
+            .findAny()
+            .orElse(null);
+
+    if (clazz != null) {
+      try {
+        return clazz
+            .getConstructor(TranslationConfiguration.class, ScopeManager.class)
+            .newInstance(this.config, scopeManager);
+      } catch (InstantiationException
+          | IllegalAccessException
+          | InvocationTargetException
+          | NoSuchMethodException e) {
+        log.error("Could not instantiate language frontend {}", clazz.getName(), e);
+        return null;
+      }
+    }
+
+    return null;
+  }
+
   /**
    * Returns the current (immutable) configuration of this TranslationManager.
    *

src/main/java/de/fraunhofer/aisec/cpg/frontends/LanguageFrontend.java

@@ -50,7 +50,7 @@
 public abstract class LanguageFrontend {
 
   protected static final Logger log = LoggerFactory.getLogger(LanguageFrontend.class);
-  protected final TranslationConfiguration config;
+  protected TranslationConfiguration config;
   protected ScopeManager scopeManager;
   /**
    * Two data structures used to associate Objects input to a pass to results of a pass, e.g.

src/main/java/de/fraunhofer/aisec/cpg/frontends/cpp/CXXLanguageFrontend.java

@@ -96,6 +96,9 @@
  */
 public class CXXLanguageFrontend extends LanguageFrontend {
 
+  public static final List<String> CXX_EXTENSIONS = List.of(".c", ".cpp", ".cc");
+  public static final List<String> CXX_HEADER_EXTENSIONS = List.of(".h", ".hpp");
+
   private static final Logger LOGGER = LoggerFactory.getLogger(CXXLanguageFrontend.class);
   private final IncludeFileContentProvider includeFileContentProvider =
       new InternalFileContentProvider() {

src/main/java/de/fraunhofer/aisec/cpg/frontends/golang/GoLanguageFrontend.kt

@@ -2,21 +2,20 @@ package de.fraunhofer.aisec.cpg.frontends.golang
 
 import de.fraunhofer.aisec.cpg.ExperimentalGolang
 import de.fraunhofer.aisec.cpg.TranslationConfiguration
-import de.fraunhofer.aisec.cpg.passes.scopes.ScopeManager
 import de.fraunhofer.aisec.cpg.frontends.LanguageFrontend
-import kotlin.Throws
 import de.fraunhofer.aisec.cpg.frontends.TranslationException
-import java.io.File
 import de.fraunhofer.aisec.cpg.graph.declarations.TranslationUnitDeclaration
+import de.fraunhofer.aisec.cpg.passes.scopes.ScopeManager
 import de.fraunhofer.aisec.cpg.sarif.PhysicalLocation
-
-class RawGolangNode {
-
-}
+import java.io.File
+import kotlin.Throws
 
 @ExperimentalGolang
-class GoLanguageFrontend(config: TranslationConfiguration, scopeManager: ScopeManager?) : LanguageFrontend(config, scopeManager, ".") {
+class GoLanguageFrontend(config: TranslationConfiguration, scopeManager: ScopeManager?) :
+    LanguageFrontend(config, scopeManager, ".") {
     companion object {
+        @kotlin.jvm.JvmField var GOLANG_EXTENSIONS: List<String> = listOf(".go")
+
         init {
             System.loadLibrary("cpgo")
         }
@@ -38,6 +37,6 @@ class GoLanguageFrontend(config: TranslationConfiguration, scopeManager: ScopeMa
     }
 
     override fun <S, T> setComment(s: S, ctx: T) {}
-    
+
     private external fun parseInternal(s: String?): TranslationUnitDeclaration
-}
+}

src/main/java/de/fraunhofer/aisec/cpg/frontends/java/JavaLanguageFrontend.java

@@ -82,6 +82,8 @@
 /** Main parser for ONE Java files. */
 public class JavaLanguageFrontend extends LanguageFrontend {
 
+  public static final List<String> JAVA_EXTENSIONS = List.of(".java");
+
   public static final String THIS = "this";
   public static final String ANNOTATION_MEMBER_VALUE = "value";