RFR: 8334714: Implement JEP 484: Class-File API [v9]

[Chen Liang]

On Fri, 15 Nov 2024 11:53:00 GMT, Adam Sotona <asotona at openjdk.org> wrote:

>> Class-File API is leaving preview.
>> This is a removal of all `@PreviewFeature` annotations from Class-File API.
>> It also bumps all `@since` tags and removes `jdk.internal.javac.PreviewFeature.Feature.CLASSFILE_API`.
>> 
>> Please review.
>> 
>> Thanks,
>> Adam
>
> Adam Sotona has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains 12 commits:
> 
>  - Merge remote-tracking branch 'openjdk/master' into JDK-8334714-final
>    
>    # Conflicts:
>    #	src/java.base/share/classes/java/lang/classfile/CustomAttribute.java
>  - Merge remote-tracking branch 'openjdk/master' into JDK-8334714-final
>  - Merge remote-tracking branch 'openjdk/master' into JDK-8334714-final
>    
>    # Conflicts:
>    #	src/java.base/share/classes/jdk/internal/javac/PreviewFeature.java
>  - Merge remote-tracking branch 'openjdk/master' into JDK-8334714-final
>    
>    # Conflicts:
>    #	src/java.base/share/classes/java/lang/classfile/AccessFlags.java
>    #	src/java.base/share/classes/java/lang/classfile/ClassBuilder.java
>    #	src/java.base/share/classes/java/lang/classfile/ClassElement.java
>    #	src/java.base/share/classes/java/lang/classfile/ClassFileTransform.java
>    #	src/java.base/share/classes/java/lang/classfile/ClassHierarchyResolver.java
>    #	src/java.base/share/classes/java/lang/classfile/ClassModel.java
>    #	src/java.base/share/classes/java/lang/classfile/ClassReader.java
>    #	src/java.base/share/classes/java/lang/classfile/ClassSignature.java
>    #	src/java.base/share/classes/java/lang/classfile/CodeBuilder.java
>    #	src/java.base/share/classes/java/lang/classfile/CodeElement.java
>    #	src/java.base/share/classes/java/lang/classfile/CodeModel.java
>    #	src/java.base/share/classes/java/lang/classfile/CompoundElement.java
>    #	src/java.base/share/classes/java/lang/classfile/FieldBuilder.java
>    #	src/java.base/share/classes/java/lang/classfile/FieldElement.java
>    #	src/java.base/share/classes/java/lang/classfile/Instruction.java
>    #	src/java.base/share/classes/java/lang/classfile/MethodBuilder.java
>    #	src/java.base/share/classes/java/lang/classfile/MethodElement.java
>    #	src/java.base/share/classes/java/lang/classfile/TypeKind.java
>    #	src/java.base/share/classes/java/lang/classfile/attribute/LocalVariableTableAttribute.java
>    #	src/java.base/share/classes/java/lang/classfile/attribute/LocalVariableTypeTableAttribute.java
>    #	src/java.base/share/classes/java/lang/classfile/attribute/RuntimeInvisibleAnnotationsAttribute.java
>    #	src/java.base/share/classes/java/lang/classfile/attribute/RuntimeVisibleAnnotationsAttribute.java
>    #	src/java.base/share/classes/java/lang/classfile/constantpool/AnnotationConstantValueEntry.java
>    #	src/java.base/share/classes/java/lang/classfile/constantpool/ConstantDynami...

I fear that promoting creation of unbound `PoolEntry` may be a bad move - they are there only for temporary components of attributes. We should promote creating a local pool entry whenever possible - this is one of the tricks for CodeBuilder optimization, where the usage of CodeBuilder convenience factories elide the creation of intermediate instruction and non-local pool entry objects.

Also for `TemporaryConstantPool` - it has a few weird behaviors, especially that its elements have bad indices.  The `TemporaryConstantPool` itself is never used as a `ConstantPoolBuilder`, and it can be refitted to only implement `ConstantPool` with minimal impact. (Well, aside from these blunt cast-and use cases you've shown)

-------------

PR Comment: https://git.openjdk.org/jdk/pull/19826#issuecomment-2529060857