Skip to content
Find file
Fetching contributors…
Cannot retrieve contributors at this time
133 lines (94 sloc) 5.05 KB



This fork is an experimental fork of Soulmate to use

For simple testing, use a haproxy config similar to the one included. Then run this code by doing this:

ruby server.rb -sv

In another terminal, run the original sinatra version (make sure you have the Thin app server gem). Then you can compare them with a test like this:

ab -n 2000 -c 200 '[]=WORD&term=abo'
ab -n 2000 -c 200 '[]=WORD&term=abo'


Soulmate is a tool to help solve the common problem of developing a fast autocomplete feature. It uses Redis's sorted sets to build an index of partially completed words and the corresponding top matching items, and provides a simple sinatra app to query them. Soulmate finishes your sentences.

Soulmate was designed to be simple and fast, and offers the following:

  • Provide suggestions for multiple types of items in a single query (at SeatGeek we're autocompleting for performers, events, and venues)
  • Results are ordered by a user-specified score
  • Arbitrary metadata for each item (at SeatGeek we're storing both a url and a subtitle)

An item is a simple JSON object that looks like:

  "id": 3,
  "term": "Citi Field",
  "score": 81,
  "data": {
    "url": "/citi-field-tickets/",
    "subtitle": "Flushing, NY"

Where id is a unique identifier (within the specific type), term is the phrase you wish to provide completions for, score is a user-specified ranking metric (redis will order things lexicographically for items with the same score), and data is an optional container for metadata you'd like to return when this item is matched (at SeatGeek we're including a url for the item as well as a subtitle for when we present it in an autocomplete dropdown).

See Soulmate in action at SeatGeek.

Getting Started

As always, kick things off with a gem install:

gem install soulmate

Loading Items

You can load data into Soulmate by piping items in the JSON lines format into soulmate load TYPE.

Here's a sample venues.json (one JSON item per line):

{"id":1,"term":"Dodger Stadium","score":85,"data":{"url":"\/dodger-stadium-tickets\/","subtitle":"Los Angeles, CA"}}
{"id":28,"term":"Angel Stadium","score":85,"data":{"url":"\/angel-stadium-tickets\/","subtitle":"Anaheim, CA"}}
{"id":30,"term":"Chase Field ","score":85,"data":{"url":"\/chase-field-tickets\/","subtitle":"Phoenix, AZ"}}
{"id":29,"term":"Sun Life Stadium","score":84,"data":{"url":"\/sun-life-stadium-tickets\/","subtitle":"Miami, FL"}}
{"id":2,"term":"Turner Field","score":83,"data":{"url":"\/turner-field-tickets\/","subtitle":"Atlanta, GA"}}

And here's the load command (Soulmate assumes redis is running locally on the default port, or you can specify a redis connection string with the --redis argument):

$ soulmate load venue --redis=redis://localhost:6379/0 < venues.json

Querying for Data

Once it's loaded, we can query this data by starting soulmate-web:

$ soulmate-web --foreground --no-launch --redis=redis://localhost:6379/0

And viewing the service in your browser: http://localhost:5678/search?types[]=venue&term=stad. You should see something like:

  "term": "stad",
  "results": {
    "venue": [
        "id": 28,
        "term": "Angel Stadium",
        "score": 85,
        "data": {
          "url": "/angel-stadium-tickets/",
          "subtitle": "Anaheim, CA"
        "id": 1,
        "term": "Dodger Stadium",
        "score": 85,
        "data": {
          "url": "/dodger-stadium-tickets/",
          "subtitle": "Los Angeles, CA"
        "id": 29,
        "term": "Sun Life Stadium",
        "score": 84,
        "data": {
          "url": "/sun-life-stadium-tickets/",
          "subtitle": "Miami, FL"

The /search method supports multiple types as well as an optional limit. For example: http://localhost:5678/search?types[]=event&types[]=venue&types[]=performer&limit=3&term=yank.

Contributing to soulmate

  • Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet
  • Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it
  • Fork the project
  • Start a feature/bugfix branch
  • Commit and push until you are happy with your contribution
  • Please try not to mess with the Rakefile, version, or history. If you want to have your own version, or is otherwise necessary, that is fine, but please isolate to its own commit so I can cherry-pick around it.


Copyright (c) 2011 Eric Waller. See LICENSE.txt for further details.

Jump to Line
Something went wrong with that request. Please try again.