Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
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.