Fix bytes hashing in source_hash

Author: msimacek

graalpython/com.oracle.graal.python/src/com/oracle/graal/python/builtins/modules/ImpModuleBuiltins.java

@@ -49,15 +49,16 @@
 import org.graalvm.nativeimage.ImageInfo;
 
 import com.oracle.graal.python.PythonLanguage;
+import com.oracle.graal.python.annotations.ArgumentClinic;
 import com.oracle.graal.python.builtins.Builtin;
 import com.oracle.graal.python.builtins.CoreFunctions;
 import com.oracle.graal.python.builtins.Python3Core;
 import com.oracle.graal.python.builtins.PythonBuiltinClassType;
 import com.oracle.graal.python.builtins.PythonBuiltins;
 import com.oracle.graal.python.builtins.objects.PNone;
 import com.oracle.graal.python.builtins.objects.buffer.PythonBufferAccessLibrary;
+import com.oracle.graal.python.builtins.objects.bytes.BytesNodes;
 import com.oracle.graal.python.builtins.objects.bytes.PBytes;
-import com.oracle.graal.python.builtins.objects.bytes.PBytesLike;
 import com.oracle.graal.python.builtins.objects.cext.capi.CExtNodes.ExecModuleNode;
 import com.oracle.graal.python.builtins.objects.cext.capi.DynamicObjectNativeWrapper;
 import com.oracle.graal.python.builtins.objects.cext.capi.ExternalFunctionNodesFactory.DefaultCheckFunctionResultNodeGen;
@@ -72,7 +73,6 @@
 import com.oracle.graal.python.builtins.objects.code.PCode;
 import com.oracle.graal.python.builtins.objects.common.HashingStorageLibrary;
 import com.oracle.graal.python.builtins.objects.ints.IntBuiltins;
-import com.oracle.graal.python.builtins.objects.ints.PInt;
 import com.oracle.graal.python.builtins.objects.module.PythonModule;
 import com.oracle.graal.python.builtins.objects.object.PythonObject;
 import com.oracle.graal.python.builtins.objects.str.PString;
@@ -82,6 +82,8 @@
 import com.oracle.graal.python.nodes.function.PythonBuiltinBaseNode;
 import com.oracle.graal.python.nodes.function.PythonBuiltinNode;
 import com.oracle.graal.python.nodes.function.builtins.PythonBinaryBuiltinNode;
+import com.oracle.graal.python.nodes.function.builtins.PythonBinaryClinicBuiltinNode;
+import com.oracle.graal.python.nodes.function.builtins.clinic.ArgumentClinicProvider;
 import com.oracle.graal.python.nodes.util.CastToJavaStringNode;
 import com.oracle.graal.python.parser.sst.SerializationUtils;
 import com.oracle.graal.python.runtime.ExecutionContext.IndirectCallContext;
@@ -378,30 +380,25 @@ private static void doPostInit(Python3Core core, PythonBuiltins builtins) {
         }
     }
 
-    @Builtin(name = "source_hash", minNumOfPositionalArgs = 2)
+    @Builtin(name = "source_hash", minNumOfPositionalArgs = 2, parameterNames = {"key", "source"})
+    @ArgumentClinic(name = "key", conversion = ArgumentClinic.ClinicConversion.Long)
+    @ArgumentClinic(name = "source", conversion = ArgumentClinic.ClinicConversion.ReadableBuffer)
     @GenerateNodeFactory
-    public abstract static class SourceHashNode extends PythonBinaryBuiltinNode {
+    public abstract static class SourceHashNode extends PythonBinaryClinicBuiltinNode {
         @Specialization
-        @TruffleBoundary
-        PBytes run(long magicNumber, PBytesLike source) {
+        PBytes run(long magicNumber, Object sourceBuffer,
+                        @Cached BytesNodes.HashBufferNode hashBufferNode) {
             byte[] hash = new byte[Long.BYTES];
-            long hashCode = magicNumber ^ source.hashCode();
+            long hashCode = magicNumber ^ hashBufferNode.execute(sourceBuffer);
             for (int i = 0; i < hash.length; i++) {
                 hash[i] = (byte) (hashCode << (8 * i));
             }
             return factory().createBytes(hash);
         }
 
-        @Specialization
-        @TruffleBoundary
-        PBytes run(PInt magicNumber, PBytesLike source) {
-            return run(magicNumber.longValue(), source);
-        }
-
-        @Specialization
-        @TruffleBoundary
-        PBytes run(int magicNumber, PBytesLike source) {
-            return run((long) magicNumber, source);
+        @Override
+        protected ArgumentClinicProvider getArgumentClinic() {
+            return ImpModuleBuiltinsClinicProviders.SourceHashNodeClinicProviderGen.INSTANCE;
         }
     }
 

graalpython/com.oracle.graal.python/src/com/oracle/graal/python/builtins/objects/bytes/BytesBuiltins.java

@@ -443,28 +443,10 @@ public Object mul(@SuppressWarnings("unused") Object self, Object other) {
     @Builtin(name = __HASH__, minNumOfPositionalArgs = 1)
     @GenerateNodeFactory
     abstract static class HashNode extends PythonUnaryBuiltinNode {
-        @Specialization(guards = "isByteStorage(self)")
-        @TruffleBoundary
-        static long hash(PBytes self) {
-            ByteSequenceStorage storage = (ByteSequenceStorage) self.getSequenceStorage();
-            byte[] array = storage.getInternalByteArray();
-            int len = storage.length();
-            int result = 1;
-            for (int i = 0; i < len; i++) {
-                result = 31 * result + array[i];
-            }
-            return result;
-        }
-
-        @Specialization(guards = "!isByteStorage(self)", limit = "1")
-        static long hashNative(PBytes self,
-                        @CachedLibrary("self") PythonBufferAccessLibrary bufferLib) {
-            int result = 1;
-            int len = bufferLib.getBufferLength(self);
-            for (int i = 0; i < len; i++) {
-                result = 31 * result + bufferLib.readByte(self, i);
-            }
-            return result;
+        @Specialization
+        long hash(PBytes bytes,
+                        @Cached BytesNodes.HashBufferNode hashBufferNode) {
+            return hashBufferNode.execute(bytes);
         }
     }
 

graalpython/com.oracle.graal.python/src/com/oracle/graal/python/builtins/objects/bytes/BytesNodes.java

@@ -858,4 +858,42 @@ private static String encodeFSDefault(byte[] path) {
             return createUTF8String(path);
         }
     }
+
+    @GenerateUncached
+    public abstract static class HashBufferNode extends PNodeWithContext {
+        public abstract long execute(Object buffer);
+
+        @Specialization(guards = "bufferLib.hasInternalByteArray(buffer)", limit = "2")
+        static long hashDirect(Object buffer,
+                        @CachedLibrary("buffer") PythonBufferAccessLibrary bufferLib) {
+            PythonBufferAccessLibrary.assertIsBuffer(buffer);
+            int len = bufferLib.getBufferLength(buffer);
+            byte[] array = bufferLib.getInternalByteArray(buffer);
+            int result = 1;
+            for (int i = 0; i < len; i++) {
+                result = 31 * result + array[i];
+            }
+            return result;
+        }
+
+        @Specialization(guards = "!bufferLib.hasInternalByteArray(buffer)", limit = "2")
+        static long hashIndirect(Object buffer,
+                        @CachedLibrary("buffer") PythonBufferAccessLibrary bufferLib) {
+            PythonBufferAccessLibrary.assertIsBuffer(buffer);
+            int len = bufferLib.getBufferLength(buffer);
+            int result = 1;
+            for (int i = 0; i < len; i++) {
+                result = 31 * result + bufferLib.readByte(buffer, i);
+            }
+            return result;
+        }
+
+        public static HashBufferNode create() {
+            return BytesNodesFactory.HashBufferNodeGen.create();
+        }
+
+        public static HashBufferNode getUncached() {
+            return BytesNodesFactory.HashBufferNodeGen.getUncached();
+        }
+    }
 }