RFR: 8347471: Provide valid flags and mask in AccessFlag.Location [v8]

ExE Boss duke at openjdk.org
Sun Apr 27 16:58:47 UTC 2025


On Sat, 26 Apr 2025 19:44:02 GMT, Chen Liang <liach at openjdk.org> wrote:

>> Some AccessFlag parsing methods throw IAE because a flag mask is not valid in a location. However, there is no easy way to check what flag mask bits or what flags are valid for a location. We need such APIs to check, specific to each class file format version.
>> 
>> Also in the investigation, it's noted that `ACC_SYNTHETIC` is incorrectly represented - it is available since release 5.0 instead of release 7. This bug is fixed together for implementation simplicity.
>> 
>> The new methods are all in `AccessFlag.Location`:
>> - `Set<AccessFlag> flags()`
>> - `int flagsMask()`
>> - `Set<AccessFlag> flags(ClassFileFormatVersion)`
>> - `int flagsMask(ClassFileFormatVersion)`
>> 
>> Also there is some simplification to `AccessFlag` itself to remove the anonymous classes, which should be more startup-friendly.
>> 
>> Testing: Tier 1-3
>
> Chen Liang has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Fix iterator missing NSEE

src/java.base/share/classes/java/lang/reflect/AccessFlag.java line 722:

> 720:     }
> 721: 
> 722:     private static final class AccessFlagSet extends AbstractSet<AccessFlag> {

This should probably also define the spliterator method:

@Override
public Spliterator<AccessFlag> spliterator() {
	return Spliterators.spliterator(
		this,
		Spliterator.DISTINCT	| Spliterator.ORDERED	| Spliterator.SIZED
		| Spliterator.NONNULL	| Spliterator.IMMUTABLE	| Spliterator.SUBSIZED
	);
}

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

PR Review Comment: https://git.openjdk.org/jdk/pull/23095#discussion_r2062674913


More information about the core-libs-dev mailing list