March 2020 - core-libs-dev

[15] RFR: 8214694: cleanup rawtypes warnings in open jndi tests
by Chris Yin 01 Apr '20

01 Apr '20

Hello Please review following simple changes to cleanup raw types warning for open jndi tests (under test/jdk/com/sun/jndi and test/jdk/javax/naming), thanks Bug: https://bugs.openjdk.java.net/browse/JDK-8214694 Webrev: http://cr.openjdk.java.net/~xyin/8214694/webrev.00/ The changes should be straightforward, only fix raw types warnings, no test logic change, no code optimization or cleanup. Minor change to each test file, just a little surprised about the affected tests count, hope this covers all. Run related jndi tests on 4 platforms for total 200 times, all passed. Thanks, Chris

3 8

(XS) Review Request JDK-8241964: Clean up java.lang.Class javadoc
by Mandy Chung 31 Mar '20

31 Mar '20

This patch cleans up the javadoc in java.lang.Class and replaces "this object" and "this class object" with "this {@code Class} object" Webrev at: http://cr.openjdk.java.net/~mchung/jdk15/webrevs/8241964/webrev.00/ thanks Mandy

2 2

RFR [15] 8241921: Remove leftover diagnostic from test/jdk/java/io/Serializable/records/SerialPersistentFieldsTest.java
by Chris Hegarty 31 Mar '20

31 Mar '20

Remove leftover diagnostic that dumps generated class from test/jdk/java/io/Serializable/records/SerialPersistentFieldsTest.java At one point it was useful to be able to inspect the generated bytecode using javap, but it was noticed recently for a different reason that the output file is always R1.class, which is clearly wrong. Anyway, this can simply be removed. --- a/test/jdk/java/io/Serializable/records/SerialPersistentFieldsTest.java +++ b/test/jdk/java/io/Serializable/records/SerialPersistentFieldsTest.java @@ -25,28 +25,26 @@ * @test * @summary Basic tests for prohibited magic serialPersistentFields * @library /test/lib * @modules java.base/jdk.internal.org.objectweb.asm * @compile --enable-preview -source ${jdk.version} SerialPersistentFieldsTest.java * @run testng/othervm --enable-preview SerialPersistentFieldsTest */ import java.io.ByteArrayInputStream; import java.io.ByteArrayOutputStream; -import java.io.FileOutputStream; import java.io.IOException; import java.io.ObjectInputStream; import java.io.ObjectOutputStream; import java.io.ObjectStreamClass; import java.io.ObjectStreamField; import java.io.Serializable; -import java.io.UncheckedIOException; import java.lang.reflect.Field; import java.lang.reflect.Modifier; import java.math.BigDecimal; import jdk.internal.org.objectweb.asm.ClassReader; import jdk.internal.org.objectweb.asm.ClassVisitor; import jdk.internal.org.objectweb.asm.ClassWriter; import jdk.internal.org.objectweb.asm.FieldVisitor; import jdk.internal.org.objectweb.asm.MethodVisitor; import jdk.internal.org.objectweb.asm.Type; import jdk.test.lib.ByteCodeLoader; @@ -222,27 +220,20 @@ return deserialize(serialize(obj)); } // -- machinery for augmenting a record class with prohibited serial field -- static byte[] addSerialPersistentFields(byte[] classBytes, ObjectStreamField[] spf) { ClassReader reader = new ClassReader(classBytes); ClassWriter writer = new ClassWriter(reader, COMPUTE_MAXS | COMPUTE_FRAMES); reader.accept(new SerialPersistentFieldsVisitor(writer, spf), 0); - try { - FileOutputStream fos = new FileOutputStream("R1.class"); - fos.write(writer.toByteArray()); - fos.close(); - } catch (IOException ioe) { - throw new UncheckedIOException(ioe); - } return writer.toByteArray(); } /** A visitor that adds a serialPersistentFields field, and assigns it in clinit. */ static final class SerialPersistentFieldsVisitor extends ClassVisitor { static final String FIELD_NAME = "serialPersistentFields"; static final String FIELD_DESC = "[Ljava/io/ObjectStreamField;"; final ObjectStreamField[] spf; String className; SerialPersistentFieldsVisitor(ClassVisitor cv, ObjectStreamField[] spf) { -Chris.

2 1

RFR [15/java.xml] 8238183: SAX2StAXStreamWriter cannot deal with comments prior to the root element
by Joe Wang 30 Mar '20

30 Mar '20

Hi, Please review a fix for the StAXResult impl. The issue was that it output comment prior to the declaration, resulting in an invalid XML document. The patch focuses on fixing this issue, but it does not cover other issues the StAXResult impl may have (e.g. JDK-8241711). JBS: https://bugs.openjdk.java.net/browse/JDK-8238183 webrev: http://cr.openjdk.java.net/~joehw/jdk15/8238183/webrev/ Thanks, Joe

3 7

[PING] RFR(s): 8176894 Provide specialized implementation for default methods putIfAbsent, computeIfAbsent, computeIfPresent, compute, merge in TreeMap
by Tagir Valeev 30 Mar '20

30 Mar '20

Hello! A gentle reminder to review the specialized implementation of TreeMap methods. The latest webrev is here: http://cr.openjdk.java.net/~tvaleev/webrev/8176894/r4/ Issue: https://bugs.openjdk.java.net/browse/JDK-8176894 CSR is already approved (thanks to Joe Darcy and Stuart Marks): https://bugs.openjdk.java.net/browse/JDK-8227666 So only code review is necessary. Also, no sponsorship is necessary. Here's the previous discussion http://mail.openjdk.java.net/pipermail/core-libs-dev/2019-October/062859.ht… Here's benchmarking results http://mail.openjdk.java.net/pipermail/core-libs-dev/2020-February/064901.h… Let's push it into Java 15! With best regards, Tagir Valeev.

5 9

[15] RFR 8241727 : Typos: empty lines in javadoc, inconsistent indents, etc. (core-libs only)
by Ivan Gerasimov 29 Mar '20

29 Mar '20

Hello! This is a javadoc-only fix. There are a few changes that will actually be visible (for example [1], [2]), but the vast majority of changes are to remove redundant empty lines, correct indentation, or otherwise restore harmony. Would you please help review this rather technical fix? BUGURL: https://bugs.openjdk.java.net/browse/JDK-8241727 WEBREV: http://cr.openjdk.java.net/~igerasim/8241727/00/webrev/ Thank in advance! [1] Extra '+' in text: https://docs.oracle.com/en/java/javase/13/docs/api/java.base/java/io/Object… [2] Extra 'I' in text: https://docs.oracle.com/en/java/javase/13/docs/api/java.base/java/lang/anno… -- With kind regards, Ivan Gerasimov

2 2

ASM 8 is available
by Remi Forax 28 Mar '20

28 Mar '20

Hi all, it's a little bit OT even if the JDK itself as a dependency on ASM. I'm proud to announce that ASM 8 (nicknamed the Quarantine Edition) was released today ! It fully supports Java 14 (the Record attribute) and even support sealed types (the attribute PermittedSubtypes) as an experimental feature. be safe, Rémi

1 0

JDK 15 RFR of JDK-8241789: Make citations of JLS and JVMS consistent in java.lang.Class
by Joe Darcy 28 Mar '20

28 Mar '20

Hello, As noticed during the CSR review of JDK-8238359, not all the citations of JVMS and JLS are done consistently in java.lang.Class. These citations should be done uniformly. http://cr.openjdk.java.net/~darcy/8241789.0/ Patch below. The referenced JVMS "table 4.1" had been renumbered so I replaced it with a reference to section 4.1. I did not reflow the paragraphs to make this review easier; I'll do a reflow where appropriate before pushing. Thanks, -Joe --- old/src/java.base/share/classes/java/lang/Class.java 2020-03-27 21:37:41.703808000 -0700 +++ new/src/java.base/share/classes/java/lang/Class.java 2020-03-27 21:37:41.415664000 -0700 @@ -139,7 +139,7 @@ * </pre></blockquote> * * It is also possible to get the {@code Class} object for a named - * type (or for void) using a class literal. See Section 15.8.2 of + * type (or for void) using a class literal. See Section {@jls 15.8.2} of * <cite>The Java™ Language Specification</cite>. * For example: * @@ -376,14 +376,14 @@ * * Note that this method throws errors related to loading, linking * or initializing as specified in Sections {@jls 12.2}, {@jls - * 12.3}, and {@jls 12.4} of The Java Language - * Specification. + * 12.3}, and {@jls 12.4} of <cite>The Java™ Language + * Specification</cite>. * Note that this method does not check whether the requested class * is accessible to its caller. * * @param name fully qualified name of the desired class * @param initialize if {@code true} the class will be initialized (which implies linking). - * See Section {@jls 12.4} of The Java Language Specification. + * See Section {@jls 12.4} of <cite>The Java™ Language Specification</cite>. * @param loader class loader from which the class must be loaded * @return class object representing the desired class * @@ -660,8 +660,8 @@ * Specifically, this method tests whether the type represented by the * specified {@code Class} parameter can be converted to the type * represented by this {@code Class} object via an identity conversion - * or via a widening reference conversion. See The Java Language - * Specification, sections {@jls 5.1.1} and {@jls 5.1.4}, + * or via a widening reference conversion. See <cite>The Java™ Language + * Specification</cite>, sections {@jls 5.1.1} and {@jls 5.1.4}, * for details. * * @param cls the {@code Class} object to be checked @@ -745,7 +745,7 @@ * Returns {@code true} if this class is a synthetic class; * returns {@code false} otherwise. * @return {@code true} if and only if this class is a synthetic class as - * defined by the Java Language Specification. + * defined by <cite>The Java™ Language Specification</cite>. * @jls 13.1 The Form of a Binary * @since 1.5 */ @@ -896,8 +896,8 @@ * the type variables declared by this generic declaration * @throws java.lang.reflect.GenericSignatureFormatError if the generic * signature of this generic declaration does not conform to - * the format specified in - * <cite>The Java™ Virtual Machine Specification</cite> + * the format specified in section {@jvms 4.7.9} of the + * <cite>The Java™ Virtual Machine Specification</cite>, * @since 1.5 */ @SuppressWarnings("unchecked") @@ -945,6 +945,7 @@ * * @throws java.lang.reflect.GenericSignatureFormatError if the generic * class signature does not conform to the format specified in + * section {@jvms 4.7.9} of the * <cite>The Java™ Virtual Machine Specification</cite> * @throws TypeNotPresentException if the generic superclass * refers to a non-existent type declaration @@ -1143,7 +1144,7 @@ * * @throws java.lang.reflect.GenericSignatureFormatError * if the generic class signature does not conform to the format - * specified in + * specified in section {@jvms 4.7.9} of the * <cite>The Java™ Virtual Machine Specification</cite> * @throws TypeNotPresentException if any of the generic * superinterfaces refers to a non-existent type declaration @@ -1201,8 +1202,8 @@ * {@code false}. The values of its other modifiers are not determined * by this specification. * - * The modifier encodings are defined in The Java Virtual Machine - * Specification, table 4.1. + * The modifier encodings are defined in section {@jvms 4.1} of the + * <cite>The Java™ Virtual Machine Specification</cite>. * * @return the {@code int} representing the modifiers for this class * @see java.lang.reflect.Modifier @@ -1611,7 +1612,7 @@ /** * Returns the canonical name of the underlying class as - * defined by the Java Language Specification. Returns null if + * defined by <cite>The Java™ Language Specification</cite>, section {@jls 6.7}. Returns null if * the underlying class does not have a canonical name (i.e., if * it is a local or anonymous class or an array whose component * type does not have a canonical name). @@ -1928,7 +1929,8 @@ * class has no public constructors, or if the class is an array class, or * if the class reflects a primitive type or void. * - * Note that while this method returns an array of {@code + * @apiNote + * While this method returns an array of {@code * Constructor<T>} objects (that is an array of constructors from * this class), the return type of this method is {@code * Constructor<?>[]} and not {@code Constructor<T>[]} as @@ -2401,7 +2403,7 @@ * object represents an interface, a primitive type, an array class, or * void. * - * See The Java Language Specification, section {@jls 8.2}. + * See <cite>The Java™ Language Specification</cite>, section {@jls 8.2}. * * @return the array of {@code Constructor} objects representing all the * declared constructors of this class @@ -3538,9 +3540,10 @@ * otherwise, if this class is not a system class (i.e., it has a * class loader) its class loader's default assertion status is returned; * otherwise, the system class default assertion status is returned. - * + * + * @apiNote * Few programmers will have any need for this method; it is provided - * for the benefit of the JRE itself. (It allows a class to determine at + * for the benefit of the JDK itself. (It allows a class to determine at * the time that it is initialized whether assertions should be enabled.) * Note that this method is not guaranteed to return the actual * assertion status that was (or will be) associated with the specified @@ -4025,10 +4028,10 @@ * * @apiNote A {@code class} file of version 55.0 or greater may record the * host of the nest to which it belongs by using the {@code NestHost} - * attribute (JVMS 4.7.28). Alternatively, a {@code class} file of + * attribute (JVMS {@jvms 4.7.28}). Alternatively, a {@code class} file of * version 55.0 or greater may act as a nest host by enumerating the nest's * other members with the - * {@code NestMembers} attribute (JVMS 4.7.29). + * {@code NestMembers} attribute (JVMS {@jvms 4.7.29}). * A {@code class} file of version 54.0 or lower does not use these * attributes. *

2 1

ClassLoader used in Module.loadModuleInfoClass should be self-first
by Michael Rasmussen 27 Mar '20

27 Mar '20

The ClassLoader used to create the pseudo module-info interface is parent-first, so if there are any modules on the classpath, it will try to load the module-info.class file from there instead of the synthetic that is being generated in the method. This interface is generated if you try to read annotations from the Module. Simple example to reproduce: /* --- module-info.java --- */ @Deprecated module com.test {} /* --- com/test/Main.java --- */ package com.test; public class Main { public static void main(String[] args) throws Exception { System.out.println(Main.class.getModule().isAnnotationPresent(Deprecated.class)); } } > java -p target/classes -m com.test/com.test.Main true > java -cp target/classes -p target/classes -m com.test/com.test.Main Exception in thread "main" java.lang.NoClassDefFoundError: module-info is not a class because access_flag ACC_MODULE is set at java.base/java.lang.ClassLoader.defineClass1(Native Method) at (.....) The problem here is that the ClassLoader is parent-first, so it will search parent for a "module-info" class, and use that if it finds it. If the ClassLoader was self-first this problem would not appear, it would still delegate to parent for everything but the synthetic module-info class. /Michael This e-mail may contain information that is privileged or confidential. If you are not the intended recipient, please delete the e-mail and any attachments and notify us immediately.

2 1

REMINDER of RFR CSR JDK-8202555: Double.toString(double) sometimes produces incorrect results
by Raffaello Giulietti 27 Mar '20

27 Mar '20

Hello, here's the periodic monthly reminder that the latest patches to overcome [1] have been submitted [2] and are waiting for a review. Full background documentation is available at [3]. Greetings Raffaello ---- [1] https://bugs.openjdk.java.net/browse/JDK-8202555 [2] http://cr.openjdk.java.net/~bpb/4511638/webrev.04/ [3] https://drive.google.com/open?id=1luHhyQF9zKlM8yJ1nebU0OgVYhfC6CBN

1 0