… count. Fixes #47.
…e test tasks. Regarding: http://weblog.rubyonrails.org/2009/9/1/gem-packaging-best-practices
…y adjustments. Ability to check existence of doc with exists? class method.
…s and runtime.
…mestamping tests that actually had sleep statements in them because mongo is so freakin fast.
…cu/schema-free. Fixes #87.
… dcu/fix-dependent. Fixes #88.
…ass. otherwise it would define the new key for all objects. Example: class Thing; include MongoMapper::Document; end a = Thing.new b = Thing.new a[:foo] = nil b.save before this patch, 'foo' is a new property of b after this patch, b's schema is not changed fixes #87 Signed-off-by: David A. Cuadrado <email@example.com>
loudly (i.e. raise an exception). Some folks also prefer to handle exceptions if document creation fails. This change set satisfies both of these needs.
complexity, and really only added the benefit of checking #valid? before saving. In this change set I merge this behaviour in, remove the superfluous module, and retain all expectations without making the code gross.
are the low hanging fruits. The other instances are going to require a little leg work. The goal is to absolish alias_method_chain to make bug hunting easier.