RFR: 8331008: Implement JEP 478: Key Derivation Function API (Preview) [v10]

[Sean Mullan]

On Fri, 2 Aug 2024 19:19:54 GMT, Kevin Driver <kdriver at openjdk.org> wrote:

>> Introduce an API for Key Derivation Functions (KDFs), which are cryptographic algorithms for deriving additional keys from a secret key and other data. See [JEP 478](https://openjdk.org/jeps/478).
>> 
>> Work was begun in [another PR](https://github.com/openjdk/jdk/pull/18924).
>
> Kevin Driver has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains 16 additional commits since the last revision:
> 
>  - update test to include Spi updates
>  - Update with latest from master
>    
>    Merge remote-tracking branch 'origin/master' into kdf-jep-wip
>    # Please enter a commit message to explain why this merge is necessary,
>    # especially if it merges an updated upstream into a topic branch.
>    #
>    # Lines starting with '#' will be ignored, and an empty message aborts
>    # the commit.
>  - add engineGetKDFParameters to the KDFSpi
>  - code review comment fix for javadoc specification
>  - change course on null return values from derive methods
>  - code review comments
>  - threading refactor + code review comments
>  - review comments
>  - review comments
>  - update code snippet type in KDF
>  - ... and 6 more: https://git.openjdk.org/jdk/compare/4742c70e...dd2ee48f

Changes requested by mullan (Reviewer).

src/java.base/share/classes/javax/crypto/KDF.java line 458:

> 456:      * @param alg
> 457:      *     the algorithm of the resultant {@code SecretKey} object
> 458:      * @param kdfParameterSpec

I think this parameter should be renamed to `derivationParameterSpec` (or something similar) to avoid confusion with the `kdfParameterSpec` variable passed to `getInstance`.

src/java.base/share/classes/javax/crypto/KDF.java line 465:

> 463:      *     permitting a {@code null} return value here, implementations will
> 464:      *     handle cases where a value cannot be returned by throwing an
> 465:      *     appropriate {@code Exception}.

We don't need to specify that `null` is not a valid return value in APIs, it is assumed *unless* you specify that `null` is a valid return value. Otherwise, applications would have to defensively check for `null` for every method where it is not specified, which would be impractical. I think the text above is in the wrong place. If there are situations where an implementation should throw an exception, then it should be specified in the @throws blocks. Is this a case where something has gone seriously wrong and a `ProviderException` should be thrown?

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

PR Review: https://git.openjdk.org/jdk/pull/20301#pullrequestreview-2219520257
PR Review Comment: https://git.openjdk.org/jdk/pull/20301#discussion_r1704428295
PR Review Comment: https://git.openjdk.org/jdk/pull/20301#discussion_r1704440267