On Sun, 1 Dec 2019 at 12:54, Claude Warren <[email protected]> wrote: > The contribution would work without the Murmur hash but losing it means no > Murmur based hashes in the factory methods. There are work arounds for > this, but all are a bit messy. > > It might be possible to clean it up a bit by creating a HashFunction > class/interface that could be used to crate the Hasher implementations > rather than the name and function that is currently used. Then it might be > much easier for integrators to implement other hash functions. > > That seems like a win all round.
Would Java Service Provider Interface be appropriate here? I could make it an optional dependency. > > +1 > On Sat, Nov 30, 2019 at 4:13 PM sebb <[email protected]> wrote: > > > On Sat, 30 Nov 2019 at 12:48, Claude Warren <[email protected]> wrote: > > > > > Greetings, > > > > > > I have a contribution[1] that requires a Murmur128 hash. The options > are > > > add the hash and associated tests to the Collections classes or use the > > > Murmur hash found in the Commons Codec project. > > > > > > In the contribution I elected to go with adding the dependency on the > > codec > > > 1.14 version.[2] However, this is currently a SNAPSHOT version and > will > > > block the release of Collections until Codec releases. > > > > > > Are there any opinions on how to proceed so that the contribution can > be > > > accepted? > > > > > > > > If the Murmur128 hash is added to Collections, it might be best if it is > > added as an internal package, so it can easily be removed if required. > > Will the contribution work at all without the Murmur128 hash? > > If so, perhaps make it an optional dependency? > > > > > > > Any opinions on the contribution as a whole? > > > > > > Thank you for your attention, > > > Claude > > > > > > [1] > > > > > > > > > https://issues.apache.org/jira/browse/COLLECTIONS-728?page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel&focusedCommentId=16985141 > > > [2] https://github.com/apache/commons-collections/pull/83 > > > > > > -- > > > I like: Like Like - The likeliest place on the web > > > <http://like-like.xenei.com> > > > LinkedIn: http://www.linkedin.com/in/claudewarren > > > > > > > > -- > I like: Like Like - The likeliest place on the web > <http://like-like.xenei.com> > LinkedIn: http://www.linkedin.com/in/claudewarren >
