A caching plugin for graphql-ruby
NOTE: this project is looking for maintainers. If you need caching to work for newer versions of graphql-ruby or for connection types, you may want to have a look at the graphql-fragment_cache gem.
A custom caching plugin for graphql-ruby
At StackShare we’ve been rolling out graphql-ruby for several of our new features and found ourselves in need of a caching solution. We could have simply used Rails.cache
in our resolvers, but this creates very verbose types or resolver classes. It also means that each and every resolver must define it’s own expiration and key. GraphQL Cache solves that problem by integrating caching functionality into the graphql-ruby resolution process making caching transparent on most fields except for a metadata flag denoting the field as cached. More details on our motivation for creating this here.
Add this line to your application’s Gemfile:
gem 'graphql-cache'
And then execute:
$ bundle
Or install it yourself as:
$ gem install graphql-cache
class MySchema < GraphQL::Schema
query Types::Query
use GraphQL::Cache
end
cache
metadata key when defining fields.module Types
class Base < GraphQL::Schema::Object
field_class GraphQL::Cache::Field
end
end
Also note that if you want access to the cache
keyword param in interface fields, the field_class directive must be added to your base interface module as well
GraphQL Cache can be configured in an initializer:
# config/initializers/graphql_cache.rb
GraphQL::Cache.configure do |config|
config.namespace = 'GraphQL::Cache' # Cache key prefix for keys generated by graphql-cache
config.cache = Rails.cache # The cache object to use for caching
config.logger = Rails.logger # Logger to receive cache-related log messages
config.expiry = 5400 # 90 minutes (in seconds)
end
Any object, list, or connection field can be cached by simply adding cache: true
to the field definition:
field :calculated_field, Int, cache: true
By default all keys will have an expiration of GraphQL::Cache.expiry
which defaults to 90 minutes. If you want to set a field-specific expiration time pass a hash to the cache
parameter like this:
field :calculated_field, Int, cache: { expiry: 10800 } # expires key after 180 minutes
GraphQL Cache generates a cache key using the context of a query during execution. A custom key can be included to implement versioned caching as well. By providing a :key
value to the cache config hash on a field definition. For example, to use a custom method that returns the cache key for an object use:
field :calculated_field, Int, cache: { key: :custom_cache_key }
With this configuration the cache key used for this resolved value will use the result of the method custom_cache_key
called on the parent object.
It is possible to force graphql-cache to resolve and write all cached fields to cache regardless of the presence of a given key in the cache store. This will effectively “renew” any existing cached expirations and warm any that don’t exist. To use forced caching, add a value to :force_cache
in the query context:
MySchema.execute('{ company(id: 123) { cachedField }}', context: { force_cache: true })
This will resolve all cached fields using the field’s resolver and write them to cache without first reading the value at their respective cache keys. This is useful for structured cache warming strategies where the cache expiration needs to be updated when a warming query is made.
After checking out the repo, run bin/setup
to install dependencies. Then, run rake spec
to run the tests. You can also run bin/console
for an interactive prompt that will allow you to experiment.
To install this gem onto your local machine, run bundle exec rake install
.
Bug reports and pull requests are welcome on GitHub at https://github.com/stackshareio/graphql-cache. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant code of conduct.
The gem is available as open source under the terms of the MIT License.
Everyone interacting in the graphql-cache project’s codebases, issue trackers, chat rooms and mailing lists is expected to follow the code of conduct…