FRED API Library and Command Line

Command line interface and a Ruby library for the Federal Reserve Economic Data (FRED) API.

Install

$ gem install fredric

Or with bundler:

gem 'fredric'

Features

Easy to use interface.
Use as a library or through the command line.
Access any FRED endpoint and option directly.
Display output as JSON, XML or CSV.
Save output to a file as JSON, XML or CSV.
Includes a built in file cache (disabled by default).

Usage

First, require and initialize with your FRED API Key.

require 'fredric'
fredric = Fredric::API.new 'your-api-key'

Now, you can access any FRED endpoint with any optional parameter, like this:

result = fredric.get "series/observations", series_id: 'GNPCA'

In addition, for convenience, you can use the first part of the endpoint as a method name, like this:

result = fredric.series 'observations', series_id: 'GNPCA'

In other words, these calls are the same:

fredric.get 'endpoint', param: value
fredric.endpoint, param: value

as well as these two:

fredric.get 'endpoint/sub', param: value
fredric.endpoint 'sub', param: value

By default, you will get a ruby hash in return. If you wish to have more control over the response, use the get! method instead:

result = fredric.get! "series/ovservations", series_id: 'GNPCA'

# Request Object
p payload.request.class
# => HTTParty::Request

# Response Object
p payload.response.class
# => Net::HTTPOK

p payload.response.body
# => JSON string

p payload.response.code
# => 200

p payload.response.msg
# => OK

# Headers Object
p payload.headers
# => Hash with headers

# Parsed Response Object
p payload.parsed_response
# => Hash with HTTParty parsed response 
#    (this is the content returned with #get)

You can get the response as CSV by calling get_csv:

result = fredric.get_csv "series/overvations", series_id: 'GNPCA'
# => CSV string

Fredric automatically decides which part of the data to convert to CSV. When there is an array in the response (for example, in the case of observations), it will be used as the CSV data. Otherwise, the entire response will be treated as a single-row CSV.

To save the output directly to a file, use the save method:

fredric.save "filename.json", "series/overvations", series_id: 'GNPCA'

Or, to save CSV, use the save_csv method:

fredric.save_csv "filename.csv", "series/overvations", series_id: 'GNPCA'

Command Line

The command line utility fred acts in a similar way. To set your FRED API Key, simply set it in the environment variable FRED_KEY:

$ export FRED_KEY=y0urAP1k3y

These commands are available:

$ fred get [--csv] PATH [PARAMS...] - print the output.
$ fred pretty PATH [PARAMS...] - print a pretty JSON.
$ fred see PATH [PARAMS...] - print a colored output.
$ fred url PATH [PARAMS...] - show the constructed URL.
$ fred save [--csv] FILE PATH [PARAMS...] - save the output to a file.

Run fred --help for more information, or view the full usage help.

Examples:

# Shows details about a data series
$ fred see series series_id:GNPCA

# Shows data as CSV
$ fred get --csv series/observations series_id:GNPCA

# Pass arguments that require spaces
$ fred see series/search "search_text:consumer price index" limit:3

# Saves a file
$ fred save --csv result.csv series/search search_text:cpi limit:3

# Shows the URL that Fredric has constructed, good for debugging
$ fred url series/observations query:interest limit:5
# => https://api.stlouisfed.org/fred/series/observations?api_key=ce1e45e6551de5db555a09b88d23682f&file_type=json&query=interest&limit=5

Caching

We are using the Lightly gem for automatic HTTP caching. To take the path of least surprises, caching is disabled by default.

You can enable and customize it by either passing options on initialization, or by accessing the Lightly object directly at a later stage.

fredric = Fredric::API.new 'apikey', use_cache: true
fredric = Fredric::API.new 'apikey', use_cache: true, cache_dir: 'tmp'
fredric = Fredric::API.new 'apikey', use_cache: true, cache_life: 120

# or 

fredric = Fredric::API.new 'apikey'
fredric.cache.enable
fredric.cache.dir = 'tmp/cache'   # Change cache folder
fredric.cache.life = 120          # Change cache life to 2 minutes

To enable caching for the command line, simply set one or both of these environment variables:

$ export FRED_CACHE_DIR=cache   # default: 'cache'
$ export FRED_CACHE_LIFE=120    # default: 3600 (1 hour)
$ fred see category/children
# => This call will be cached