class RDF::Util::Cache
A ‘Hash`-like cache that holds only weak references to the values it caches, meaning that values contained in the cache can be garbage collected. This allows the cache to dynamically adjust to changing memory conditions, caching more objects when memory is plentiful, but evicting most objects if memory pressure increases to the point of scarcity.
While this cache is something of an internal implementation detail of RDF.rb, some external libraries do currently make use of it as well, including [SPARQL](github.com/ruby-rdf/sparql/) and [Spira](github.com/ruby-rdf/spira). Do be sure to include any changes here in the RDF.rb changelog.
@see RDF::URI.intern
@see en.wikipedia.org/wiki/Weak_reference @since 0.2.0
Attributes
The configured cache capacity.
Public Class Methods
Source
# File lib/rdf/util/cache.rb, line 25 def self.new(*args) # JRuby doesn't support `ObjectSpace#_id2ref` unless the `-X+O` # startup option is given. In addition, ObjectSpaceCache is very slow # on Rubinius. On those platforms we'll default to using # the WeakRef-based cache: if RUBY_PLATFORM == 'java' || (defined?(RUBY_ENGINE) && RUBY_ENGINE == 'rbx') klass = WeakRefCache else klass = ObjectSpaceCache end cache = klass.allocate cache.send(:initialize, *args) cache end
@private
Source
# File lib/rdf/util/cache.rb, line 42 def initialize(capacity = nil) @capacity = capacity || RDF.config.cache_size @cache ||= {} @index ||= {} end
@param [Integer] capacity
Public Instance Methods
Source
# File lib/rdf/util/cache.rb, line 56 def capacity? @capacity.equal?(-1) || @capacity > @cache.size end
@return [Boolean]