[
https://issues.apache.org/jira/browse/COLLECTIONS-728?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16945717#comment-16945717
]
Claude Warren commented on COLLECTIONS-728:
-------------------------------------------
I am working to clean up the issues that [~erans] raised with the following
notes:
The BloomCollection can be used. It works as described but it is limited to
wrapping Java Collection implementations. I expect that this will be
sufficient for many implementations but for very large scale implementations it
may be desirable to have the BloomFilterGated<T> implement on top of a larger
datastore. So the BloomCollection is not _just_ an example.
I can for see other implementations of BloomFilter. This contribution contains
2 (though Counting is derived from Standard) there are a number of other
implementations identified in https://en.wikipedia.org/wiki/Bloom_filter . I
would expect to see some of those contributed in the future.
With respect to duplicate build() methods: They are not "necessary". They are
convenience methods akin to the Java MessageDigest update() and digest()
methods. No, all the digest() methods in MessageDigest are not necessary but
they are convenient. Yes, MessageDigest could have been implemented with only
the no argument digest() method. The same can be said for the build() methods
of the ProtoBloomFilter.Builder. The use cases for both are much the same:
Having a single method to build a digest/ProtoBloomFilter from a single data
object (e.g. messageDigest..digest( "foo" ) or builder.build("foo"))
> BloomFilter contribution
> ------------------------
>
> Key: COLLECTIONS-728
> URL: https://issues.apache.org/jira/browse/COLLECTIONS-728
> Project: Commons Collections
> Issue Type: Task
> Reporter: Claude Warren
> Priority: Minor
>
> Contribution of BloomFilter library comprising base implementation and gated
> collections.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
