On Thu, 14 Nov 2024 22:53:06 GMT, Kevin Rushforth <k...@openjdk.org> wrote:
>> This PR removes all remaining uses of `AccessController` and
>> `AccessControlContext`, which represent the last remaining uses of the
>> terminally deprecated security APIs except for those in the `/ios/` or
>> `/android/` directories.
>>
>> With the removal of doPrivileged and the `if (System.getSecurityManager() !=
>> null)` code paths, the ACC is no longer used, so can be completely
>> eliminated. Along with this, I removed all unused imports of
>> security-related APIs and all related `@SuppressWarnings("removal")
>> annotations.
>>
>> ### Notes to reviewers
>>
>> * Most of the changes were straight-forward removals of methods and fields
>> to save, retrieve and pass around the `AccessControlContext`.
>> * The Toolkit class stores a collection of listeners in a `WeakHashMap` with
>> the listener as the key (thus weakly held) and the ACC as the value. We no
>> longer need or want the ACC, but I kept the use of `WeakHashMap` and changed
>> the value type to `Object`, storing a singleton dummy object as the value
>> for each entry. This minimizes the changes, while preserving the behavior of
>> reclaiming the entries when they are garbage collected.
>
> Kevin Rushforth has updated the pull request incrementally with one
> additional commit since the last revision:
>
> additional comments
looks good, with some minor comments.
I don't know if @mstr2 's idea of converting weak hash maps to sets would help,
since it does not save any memory (might actually eat more memory). putting a
`null` value might reduce the memory footprint, but needs to be checked if it
would work, and putting a Boolea.TRUE would work without adding a dummy object.
modules/javafx.graphics/src/main/java/com/sun/javafx/tk/Toolkit.java line 98:
> 96:
> 97: // Used as the (dummy) value for the various listener maps
> 98: private static final Object dummyObj = new Object();
minor: maybe use all-uppercase name for this static final object?
we could also be using Boolean.TRUE (or null), since there is a comment down
below explaining the purpose of these maps.
curiously, there is no WeakHashSet.
modules/javafx.web/src/main/java/com/sun/webkit/dom/JSObject.java line 45:
> 43: // We do this, rather than removing the parameter, in order to keep
> the
> 44: // native WebKit code the same across different release families.
> 45: private static final Object dummyAcc = new Object();
very minor: all uppercase name
modules/javafx.web/src/main/java/com/sun/webkit/dom/JSObject.java line 120:
> 118: public void setSlot(int index, Object value) throws JSException {
> 119: Invoker.getInvoker().checkEventThread();
> 120: setSlotImpl(peer, peer_type, index, value,dummyAcc);
missing space after `,`
modules/javafx.web/src/main/java/com/sun/webkit/dom/JSObject.java line 129:
> 127: public Object call(String methodName, Object... args) throws
> JSException {
> 128: Invoker.getInvoker().checkEventThread();
> 129: return callImpl(peer, peer_type, methodName, args,dummyAcc);
missing space after `,`
-------------
PR Review: https://git.openjdk.org/jfx/pull/1638#pullrequestreview-2437376185
PR Review Comment: https://git.openjdk.org/jfx/pull/1638#discussion_r1842997206
PR Review Comment: https://git.openjdk.org/jfx/pull/1638#discussion_r1843010983
PR Review Comment: https://git.openjdk.org/jfx/pull/1638#discussion_r1843011173
PR Review Comment: https://git.openjdk.org/jfx/pull/1638#discussion_r1843011305
