Skip to content

Instantly share code, notes, and snippets.

@JacobNinja
Created February 22, 2013 23:08
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save JacobNinja/5017303 to your computer and use it in GitHub Desktop.
Save JacobNinja/5017303 to your computer and use it in GitHub Desktop.
Lazy sequences in Ruby
lazy_expensive_sequence = Enumerator.new do |y|
offset = 0
amount_to_retrieve = 50
loop do
## This may not be memory safe if we try to retrieve all at once, so we safely retrieving an amount that we know will fit in memory
retrieve_memory_hogs(offset, amount_to_retrieve).each do |i|
y << i
end
offset += amount_to_retrieve
end
end
## realize the sequence without consuming too much memory
lazy_expensive_sequence.each do |i|
## I never have to worry about overloading memory
## I never have to worry about the logic behind how I chunk my memory
## I just consume
end
## Lazy composition
def select_map(collection, filter, transform)
Enumerator.new do |y|
collection.each do |i|
y << transform.call(i) if filter.call(i)
end
end
end
## Concurrency via threads
def retrieve_sources(website_urls)
Enumerator.new do |y|
website_urls.each do |url|
Thread.new do
y << get_source(url)
end
end
end
end
## Consume the concurrent lazy sequence
## the consumer doesn't know or care whether it's using threads. It just gets yielded results
retrieve_sources(%w(github.com google.com facebook.com)).each do |source|
analyze(source)
end
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment