Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
branch: master
Fetching contributors…

Cannot retrieve contributors at this time

100 lines (78 sloc) 4.412 kb

Riak-Client/Ripple Release Notes

0.7.1 Patch Release – 2010-06-08

This release has no new features but includes bug fixes and some internal refactoring of the Ripple::Document hierarchy.

  • The Net::HTTP backend should handle streamed keys better (although
    still not perfectly).
  • The Riak::MapReduce#timeout method now returns self, allowing
    chaining.
  • The Ripple::Document and Ripple::EmbeddedDocument are less coupled
    from one another so numerous internal confusions about calling order
    are fixed.
  • When using Riak::RObject#to_link, a blank tag is no longer allowed.

0.7.0 Feature Release – 2010-05-06

This release includes a number of new features. The largest change is that the library is now split into two gems, ‘riak-client’ and ‘ripple’. ‘riak-client’ supports ActiveSupport 2.3.5, ‘ripple’ only supports 3.0.0.beta3.

A big kudos goes to Adam Hunter who contributed the majority of the new associations code.

In addition, these changes were made:

  • Keys should stream properly now from Bucket#keys (the “stream” option was left off).
  • Deletes can be issued directly from a Bucket without instantiating an RObject.
  • Added a ActiveSupport 3.0-compatible Cache Store. [Shay Frendt]
  • Added Bucket#exists?
  • A provisionally complete implementation of embedded document associations. [Adam Hunter]
  • Ripple::Document passes ActiveModel::Lint tests.
  • Updated Rails 3 dependencies to beta3 and RSpec to 2.0.0.beta6
  • Ripple::Document handles nil keys better. [John Lynch]

0.6.1 Patch Release – 2010-03-17

This is a minor release with fixes for a few issues:

  • Riak::Link objects will now be unique when added to RObject#links
    Set. [John Lynch]
  • Attributes on Ripple::Document classes are no longer clone, which had
    prevented non-scalar properties from being modified directly (e.g. Array).
  • Buckets, keys, and walk specs are properly escaped in URLs
    (including slashes).
  • Time-related properties properly convert to string formats in JSON that
    they will work in the context of MapReduce jobs.

0.6.0 Feature Release – 2010-03-05

This release contains enhancements and bugfixes in preparation for the
Riak 0.9 release.

  • The CurbBackend now uses fibers to prevent curl-handle corruption when
    a block is given to streaming operations.
  • The default prefix is now “/riak/” to match the latest version of Riak.
  • The client configuration for Ripple is now used.
  • Added Bucket#new and Bucket#get_or_new for easily creating new objects.
  • Added Bucket#allow_mult and Bucket#n_value accessors for more easily setting
    bucket properties.
  • Added timestamps! method for easily adding created_at/updated_at to documents.
    [Adam Hunter]
  • The ‘links’ collection on RObject is now a Set instead of an Array.
  • All literal messages are now stored in YAML localization files.
  • Object siblings (caused by concurrent updates when allow_mult is true) can now
    be accessed directly.
  • Map-reduce jobs now have timeouts (in parity with Riak).

0.5.1 Patch Release – 2010-02-22

This is a minor release with fixes for Ruby 1.9, bundler/edge Rails,
and a minor feature addition. Changes:

  • Qualify namespaces for Ruby 1.9.
  • Decoupled a few specs that gave the appearance of failure.
  • Added “bucket” and “key” properties on Riak::Link objects. [John Lynch]
  • Fully-qualify the JSON constant, using ActiveSupport::JSON instead.
  • Adjusted gem specification to accommodate edge Rails. [Preston Marshall]

0.5 Initial Release – 2010-02-10

This is the first release of Ripple, which would not have been possible
without the generous support of Sonian and Basho Technologies. Many thanks.
It includes:

  • A robust basic client, Riak, with:
    • multiple HTTP backends (curb, net/http)
    • sensible client defaults (local, default port)
    • bucket access and manipulation, including key-streaming
    • object reading, storing, deleting and reloading
    • automatic de-serialization of JSON, YAML, and Marshal (when given the right content type)
    • streaming POST/PUT bodies (when given an IO)
    • method-chained map-reduce job construction
  • A document-style modeling library, Ripple, with:
    • ActiveModel 3.0 compatibility
    • Property/attribute definition with automatic type-casting
    • Bucket selection based on class name, with single-bucket inheritance (configurable)
    • Validations
    • Dirty-tracking
    • Simple finders – all documents, by key
    • Reloading
Jump to Line
Something went wrong with that request. Please try again.