Quandl Ruby

require 'quandl'

Quandl.configure do |config|
  config.auth_token = 'your quandl api key'
end

# Replace `Class` below with the relevant class from above.

# Query is the specific query for that class
# Options is an options hash containing data manipulation options

# Data manipulation docs:
#   http://www.quandl.com/help/api#Data-Manipulation

# Creating an instance doesn't retrieve the data
# Data format options:
#   `:json` => parsed JSON (default)
#   `:csv`  => parsed CSV into hashes
#   `:xml`  => a string of unparsed xml
# Set the format by providing a `:format` key in the options hash
# Quandl::Class.get is an alias for Quandl::Class.new
query = Quandl::Class.new(query, options)

# Use instance#data to retrieve data from Quandl
data = query.data

# Instantiating and then calling the `data` method memoizes the response
# Pass in `true` to instance#data to clear the cache
data = query.data(true)

# This will make more sense if you read through:
#   http://www.quandl.com/help/api#Data-Manipulation
gdp = Quandl::Dataset.get('FRED/GDP')

# Corresponds to the `rows` parameter. Aliased as #rows.
gdp.limit(20)

# Corresponds to the `sort_order` parameter. Aliased as #sort.
gdp.order(:asc) # Only accepts :asc or :desc

# Corresponds to the `column` parameter.
gdp.column(4)

# Corresponds to the `collapse` parameter. Aliased as #frequency
# Accepts: :none, :daily, :weekly, :monthly, :quarterly, :annual
gdp.collapse(:annual)

# Corresponds to the `exclude_headers` parameter.
gdp.headers(false) # (sets `exclude_headers` to `true`)

# Corresponds to the `trim_start` parameter. Also accepts an instance of Date.
gdp.start('2010-01-01')

# Corresponds to the `trim_end` parameter. Also accepts an instance of Date.
gdp.end('2014-01-01')

# Corresponds to the `transformation` parameter.
# Accepts: :diff, :rdiff, :cumul, :normalize
gdp.transform(:normalize)

# After all the above, call #metadata or #data to retrieve the respective data
gdp.metadata # => returns metadata for 'FRED/GDP'
gdp.data # => returns metadata for 'FRED/GDP'

# If you were to change an option, like below:
gdp.transform(:rdiff)

# You'll need to call #reload! or pass `true` as an argument to #data or #metadata
gdp.reload!

gdp.data

# Get the US GDP in JSON
#   Example from: http://www.quandl.com/help/api#A-Simple-Example
gdp = Quandl::Dataset.get('FRED/GDP')

# Annual percentage changes of US GDP, crude oil prices, and Apple stock for the last 10 years
#   Example from: http://www.quandl.com/help/api#A-Multiset-Example

datasets = ['FRED/GDP/1', 'DOE/RWTC/1', 'WIKI/AAPL/4']
comparison = Quandl::Multiset.get(datasets).
               collapse(:annual).
               transform(:rdiff).
               limit(10)
# Unlike the official Quandl REST API, separate source, table, and column numbers
# With `/` as opposed to `.` to maintain consistency with Quandl::Dataset

# Retrieve only the metadata for a dataset
#   Example from: http://www.quandl.com/help/api#Getting-Metadata
oil_metadata = Quandl::Metadata.get('NSE/OIL')

# Retrieve a user's favorites
#   Example from: http://www.quandl.com/help/api#Getting-Favourites

# Defaults to getting the favorites of the user who's API key you've configured
my_favorites = Quandl::Favorites.get

# Or pass in another API key as the first argument to override your own
another_guys_favorites = Quandl::Favorites.get('that_guys_api_key')

# For Brits and Canucks, Quandl::Favorites is aliased as Quandl::Favourites