Restforce is a ruby gem for the Salesforce REST api.
Features include:
- A clean and modular architecture using Faraday middleware and Hashie::Mash'd responses.
- Support for interacting with multiple users from different organizations.
- Support for parent-to-child relationships.
- Support for aggregate queries.
- Support for the Streaming API
- Support for the Composite API
- Support for the Composite Batch API
- Support for the GetUpdated API
- Support for blob data types.
- Support for GZIP compression.
- Support for custom Apex REST endpoints.
- Support for dependent picklists.
- Support for decoding Force.com Canvas signed requests. (NEW!)
Official Website | Documentation | Changelog
Add this line to your application's Gemfile:
gem 'restforce', '~> 7.3.1'
And then execute:
$ bundle
Or install it yourself as:
$ gem install restforce
As of version 7.0.0, this gem is only compatible with Ruby 3.0.0 and later. If you're using an earlier Ruby version:
- for Ruby 2.7, use version 6.2.4 or earlier
- for Ruby 2.6, use version 5.3.1 or earlier
- for Ruby 2.5, use version 5.0.6 or earlier
- for Ruby 2.4, use version 4.3.0 or earlier
- for Ruby 2.3, use version 3.2.0 or earlier
- for Ruby versions 2.2, 2.1 and 2.0, use version 2.5.3 or earlier
- for Ruby 1.9.3, use version 2.4.2
This gem is versioned using Semantic Versioning, so you can be confident when updating that there will not be breaking changes outside of a major version (following format MAJOR.MINOR.PATCH, so for instance moving from 3.1.0 to 4.0.0 would be allowed to include incompatible API changes). See the changelog for details on what has changed in each version.
Restforce is designed with flexibility and ease of use in mind. By default, all API calls will
return Hashie::Mash objects,
so you can do things like client.query('select Id, (select Name from Children__r) from Account').first.Children__r.first.Name.
Which authentication method you use really depends on your use case. If you're building an application where many users from different organizations are authenticated through OAuth and you need to interact with data in their org on their behalf, you should use the OAuth token authentication method.
If you're using the gem to interact with a single org (maybe you're building some salesforce integration internally?) then you should use the username/password authentication method.
It is also important to note that the client object should not be reused across different threads, otherwise you may encounter thread-safety issues.
client = Restforce.new(oauth_token: 'access_token',
instance_url: 'instance url',
api_version: '41.0')Although the above will work, you'll probably want to take advantage of the (re)authentication middleware by specifying refresh_token, client_id, client_secret, and authentication_callback:
client = Restforce.new(oauth_token: 'access_token',
refresh_token: 'refresh token',
instance_url: 'instance url',
client_id: 'client_id',
client_secret: 'client_secret',
authentication_callback: Proc.new { |x| Rails.logger.debug x.to_s },
api_version: '41.0')The middleware will use the refresh_token automatically to acquire a new access_token if the existing access_token is invalid. The refresh process uses the host option so make sure that is set correctly for sandbox organizations.
authentication_callback is a proc that handles the response from Salesforce when the refresh_token is used to obtain a new access_token. This allows the access_token to be saved for re-use later - otherwise subsequent API calls will continue the cycle of "auth failure/issue new access_token/auth success".
The proc is passed one argument, a Hashie::Mash of the response from the Salesforce API:
{
"access_token" => "00Dx0000000BV7z!AR8AQP0jITN80ESEsj5EbaZTFG0RNBaT1cyWk7T5rqoDjoNIWQ2ME_sTZzBjfmOE6zMHq6y8PIW4eWze9JksNEkWUl.Cju7m4",
"signature" => "SSSbLO/gBhmmyNUvN18ODBDFYHzakxOMgqYtu+hDPsc=",
"scope" => "refresh_token full",
"instance_url" => "https://na1.salesforce.com",
"id" => "https://login.salesforce.com/id/00Dx0000000BV7z/005x00000012Q9P",
"token_type" => "Bearer",
"issued_at" => "1278448384422"
}The id field can be used to uniquely identify the user that the access_token and refresh_token belong to.
If you prefer to use a username and password to authenticate:
client = Restforce.new(username: config['username'],
password: config['password'],
instance_url: config['instance_url'],
host: config['host'], # https://test.salesforce.com for sandbox (optional)
client_id: config['client_key'], # Salesforce Client Key
client_secret: config['client_secret'], # Salesforce Client Secret
api_version: '55.0')If you prefer to use a JWT Bearer Token to authenticate:
client = Restforce.new(username: 'foo',
client_id: 'client_id',
instance_url: 'instance_url',
jwt_key: 'certificate_private_key',
api_version: '38.0')The jwt_key option is the private key of the certificate uploaded to your Connected App in Salesforce.
Choose "use digital signatures" in the Connected App configuration screen to upload your certificate.
You can also set the username, password, security token, client ID, client secret and API version in environment variables:
export SALESFORCE_USERNAME="username"
export SALESFORCE_PASSWORD="password"
export SALESFORCE_SECURITY_TOKEN="security token"
export SALESFORCE_CLIENT_ID="client id"
export SALESFORCE_CLIENT_SECRET="client secret"
export SALESFORCE_API_VERSION="41.0"client = Restforce.newNote: Restforce library does not cache JWT Bearer tokens automatically. This means that every instantiation of the Restforce class will be treated as a new login by Salesforce. Remember that Salesforce enforces rate limits on login requests. If you are building an application that will instantiate the Restforce class more than this specified rate limit, you might want to consider caching the Bearer token either in-memory or in your own storage by leveraging the authentication_callback method.
If you want to authenticate as an application, you can use the Client Credentials flow:
client = Restforce.new(client_id: 'client_id',
client_secret: 'client_secret',
api_version: '55.0',
host: 'MYDOMAIN.my.salesforce.com')You can connect to sandbox organizations by specifying a host. The default host is 'login.salesforce.com':
client = Restforce.new(host: 'test.salesforce.com')The host can also be set with the environment variable SALESFORCE_HOST.
You can specify a HTTP proxy using the proxy_uri option, as follows, or by setting the SALESFORCE_PROXY_URI environment variable:
client = Restforce.new(username: 'foo',
password: 'bar',
security_token: 'security token',
client_id: 'client_id',
client_secret: 'client_secret',
proxy_uri: 'http://proxy.example.com:123',
api_version: '41.0')You may specify a username and password for the proxy with a URL along the lines of 'http://user:password@proxy.example.com:123'.
You can set any of the options passed into Restforce.new globally:
Restforce.configure do |config|
config.client_id = 'foo'
config.client_secret = 'bar'
endBy default, the gem defaults to using Version 26.0 (Winter '13) of the Salesforce API. This maintains backwards compatibility for existing users.
We strongly suggest configuring Restforce to use the most recent API version, to get the best Salesforce API experience - for example, some more recently-added API endpoints will not be available without moving to a more recent
version. If you're trying to use a method that is unavailable with your API version,
Restforce will raise an APIVersionError.
There are three ways to set the API version:
- Passing in an
api_versionoption when instantiatingRestforce(i.e.Restforce.new(api_version: '41.0')) - Setting the
SALESFORCE_API_VERSIONenvironment variable (i.e.export SALESFORCE_API_VERSION="41.0") - Configuring the version globally with
Restforce.configure:
Restforce.configure do |config|
config.api_version = '41.0'
# ...
endAll the CRUD methods (create, update, upsert, destroy) have equivalent methods with
a ! at the end (create!, update!, upsert!, destroy!), which can be used if you need
to do some custom error handling. The bang methods will raise exceptions, while the
non-bang methods will return false in the event that an exception is raised. This
works similarly to ActiveRecord.
Salesforce allows the addition of
custom headers
in REST API requests to trigger specific logic. In order to pass any custom headers along with API requests,
you can specify a hash of :request_headers upon client initialization. The example below demonstrates how
to include the sforce-auto-assign header in all client HTTP requests:
client = Restforce.new(oauth_token: 'access_token',
instance_url: 'instance url',
api_version: '41.0',
request_headers: { 'sforce-auto-assign' => 'FALSE' })accounts = client.query("select Id, Something__c from Account where Id = 'someid'")
# => #<Restforce::Collection >
account = accounts.first
# => #<Restforce::SObject >
account.sobject_type
# => 'Account'
account.Id
# => "someid"
account.Name = 'Foobar'
account.save
# => true
account.destroy
# => trueaccounts = client.query_all("select Id, Something__c from Account where isDeleted = true")
# => #<Restforce::Collection >query_all allows you to include results from your query that Salesforce hides in the default "query" method. These include soft-deleted records and archived records (e.g. Task and Event records which are usually archived automatically after they are a year old).
Only available in version 29.0 and later of the Salesforce API.
explain takes the same parameters as query and returns a query plan in JSON format.
For the nitty-gritty details on the response meanings visit the
Salesforce Query Explain page.
accounts = client.explain("select Id, Something__c from Account where Id = 'someid'")
# => #<Restforce::Mash >Only available in version 30.0 and later of the Salesforce API.
client.find('Account', '001D000000INjVe')
# => #<Restforce::SObject Id="001D000000INjVe" Name="Test" LastModifiedBy="005G0000002f8FHIAY" ... >
client.find('Account', '1234', 'Some_External_Id_Field__c')
# => #<Restforce::SObject Id="001D000000INjVe" Name="Test" LastModifiedBy="005G0000002f8FHIAY" ... >find raises an error if nothing is found.
select allows the fetching of a specific list of fields from a single object. It requires an external_id lookup, but is often much faster than an arbitrary query.
# Select the `Id` column from a record with `Some_External_Id_Field__c` set to '001D000000INjVe'
client.select('Account', '001D000000INjVe', ["Id"], 'Some_External_Id_Field__c')
# => {"attributes" : {"type" : "Account","url" : "/services/data/v20.0/sobjects/Account/Some_External_Id_Field__c/001D000000INjVe"}, "Id" : "003F000000BGIn3"}# Find all occurrences of 'bar'
client.search('FIND {bar}')
# => #<Restforce::Collection >
# Find accounts matching the term 'genepoint' and return the `Name` field
client.search('FIND {genepoint} RETURNING Account (Name)').map(&:Name)
# => ['GenePoint']# Add a new account
client.create('Account', Name: 'Foobar Inc.')
# => '0016000000MRatd'# Update the Account with `Id` '0016000000MRatd'
client.update('Account', Id: '0016000000MRatd', Name: 'Whizbang Corp')
# => true# Update the record with external `External__c` external ID set to '12'
client.upsert('Account', 'External__c', External__c: 12, Name: 'Foobar')
# => true or "RecordId"The upsert method will return the record Id if included in the response body from the Salesforce API; otherwise, it returns true. Currently the Salesforce API only returns the Id for newly created records.
# Delete the Account with `Id` '0016000000MRatd'
client.destroy('Account', '0016000000MRatd')
# => true# Get the global describe for all sobjects
client.describe
# => { ... }
# Get the describe for the Account object
client.describe('Account')
# => { ... }# Get layouts for an sobject type
client.describe_layouts('Account')
# => { ... }
# Get the details for a specific layout by its ID
client.describe_layouts('Account', '012E0000000RHEp')
# => { ... }Only available in version 28.0 and later of the Salesforce API.
# Fetch picklist value for Account's `Type` field
client.picklist_values('Account', 'Type')
# => [#<Restforce::Mash label="Prospect" value="Prospect">]
# Given a custom object named Automobile__c with picklist fields
# `Model__c` and `Make__c`, where options for `Model__c` depends on the value of
# `Make__c`.
client.picklist_values('Automobile__c', 'Model__c', valid_for: 'Honda')
# => [#<Restforce::Mash label="Civic" value="Civic">, ... ]# Get info about the logged-in user
client.user_info
# => #<Restforce::Mash active=true display_name="Chatty Sassy" email="user@example.com" ... >limits returns the API limits for the currently connected organization. This includes information such as Daily API calls and Daily Bulk API calls. More information can be found on the
Salesforce Limits page.
# Get the current limit info
limits = client.limits
# => #<Restforce::Mash >
limits["DailyApiRequests"]
# => {"Max"=>15000, "Remaining"=>14746}Only available in version 29.0 and later of the Salesforce API.
Retrieves the list of individual record IDs that have been updated (added or changed) within the given timespan for the specified object
# Get the ids of all accounts which have been updated in the last day
client.get_updated('Account', Time.local(2015,8,18), Time.local(2015,8,19))
# => { ... }Retrieves the list of IDs and time of deletion for records that have been deleted within the given timespan for the specified object
# Get the list of accounts which have been deleted in the last day
client.get_deleted('Account', Time.local(2015,8,18), Time.local(2015,8,19))
# => { ... }Performs an authentication and returns the response. In general, calling this directly shouldn't be required, since the client will handle authentication for you automatically. This should only be used if you want to force an authentication before using the streaming api, or you want to get some information about the user.
response = client.authenticate!
# => #<Restforce::Mash access_token="..." id="https://login.salesforce.com/id/00DE0000000cOGcMAM/005E0000001eM4LIAU" instance_url="https://na9.salesforce.com" issued_at="1348465359751" scope="api refresh_token" signature="3fW0pC/TEY2cjK5FCBFOZdjRtCfAuEbK1U74H/eF+Ho=">
# Get the user information
info = client.get(response.id).body
info.user_id
# => '005E0000001eM4LIAU'Using the new Blob Data api feature (500mb limit):
client.create('Document', FolderId: '00lE0000000FJ6H',
Description: 'Document test',
Name: 'My image',
Body: Restforce::FilePart.new(File.expand_path('image.jpg', __FILE__), 'image/jpeg')Using base64 encoded data (37.5mb limit):
client.create('Document', FolderId: '00lE0000000FJ6H',
Description: 'Document test',
Name: 'My image',
Body: Base64::encode64(File.read('image.jpg'))See also: Inserting or updating blob data
Restforce also makes it incredibly easy to download Attachments or Documents:
attachment = client.query('select Id, Name, Body from Attachment').first
File.open(attachment.Name, 'wb') { |f| f.write(attachment.Body) }document = client.query('select Id, Name, Body from Document').first
File.open(document.Name, 'wb') { |f| f.write(document.Body) }Note: The example above is only applicable if your SOQL query returns a single Document record. If more than one record is returned, the Body field contains an URL to retrieve the BLOB content for the first 2000 records returned. Subsequent records contain the BLOB content in the Body field. This is confusing and hard to debug. See notes in Issue #301 explaining this detail. Executive Summary: Don't retrieve the Body field in a SOQL query; instead, use the BLOB retrieval URL documented in SObject BLOB Retrieve
You can use Restforce to interact with your custom REST endpoints, by using
.get, .put, .patch, .post, and .delete.
For example, if you had the following Apex REST endpoint on Salesforce:
@RestResource(urlMapping='/FieldCase/*')
global class RESTCaseController {
@HttpGet
global static List<Case> getOpenCases() {
String companyName = RestContext.request.params.get('company');
Account company = [ Select ID, Name, Email__c, BillingState from Account where Name = :companyName];
List<Case> cases = [SELECT Id, Subject, Status, OwnerId, Owner.Name from Case WHERE AccountId = :company.Id];
return cases;
}
}Then you could query the cases using Restforce:
client.get('/services/apexrest/FieldCase', company: 'GenePoint')
# => #<Restforce::Collection ...>Restforce supports the Streaming API, and makes implementing pub/sub with Salesforce a trivial task.
Here is an example of creating and subscribing to a PushTopic:
# Restforce uses faye as the underlying implementation for CometD.
require 'faye'
# Initialize a client with your username/password/oauth token/etc.
client = Restforce.new(username: 'foo',
password: 'bar',
security_token: 'security token',
client_id: 'client_id',
client_secret: 'client_secret')
# Create a PushTopic for subscribing to Account changes.
client.create!('PushTopic',
ApiVersion: '23.0',
Name: 'AllAccounts',
Description: 'All account records',
NotifyForOperations: 'All',
NotifyForFields: 'All',
Query: "select Id from Account")
EM.run do
# Subscribe to the PushTopic.
client.subscription '/topic/AllAccounts' do |message|
puts message.inspect
end
endBoom, you're now receiving push notifications when Accounts are created/updated.
Restforce supports the Composite API. This feature permits the user to send a composite object—that is, a complex object with nested children—in a single API call. Up to 25 requests may be included in a single composite.
# build up an array of requests:
requests << {
method: :update,
sobject: sobject, # e.g. "Contact"
reference_id: reference_id,
data: data
}
# send every 25 requests as a subrequest in a single composite call
requests.each_slice(25).map do |req_slice|
client.composite do |subrequest|
req_slice.each do |r|
subrequest.send *r.values
end
end
end
# note that we're using `map` to return an array of each responses to each
# composite call; 100 requests will produce 4 responsesRestforce supports the Composite Batch API. This feature permits up to 25 subrequests in a single request, though each subrequest counts against the API limit. On the other hand, it has fewer limitations than the Composite API.
client.batch do |subrequests|
subrequests.create('Object', name: 'test')
subrequests.update('Object', id: '123', name: 'test')
subrequests.destroy('Object', '123')
end
Since API version 37.0, Salesforce stores events for 24 hours and they can be replayed if your application experienced some downtime.
In order to replay past events, all you need to do is specify the last known event ID when subscribing and you will receive all events that happened since that event ID:
EM.run {
# Subscribe to the PushTopic.
client.subscription '/topic/AllAccounts', replay: 10 do |message|
puts message.inspect
end
}In this specific case you will see events with replay ID 11, 12 and so on.
There are two magic values for the replay ID accepted by Salesforce:
-2, for getting all the events that appeared in the last 24 hours-1, for getting only newer events
Warning: Only use a replay ID of a event from the last 24 hours otherwise Salesforce will not send anything, including newer events. If in doubt, use one of the two magic replay IDs mentioned above.
You might want to store the replay ID in some sort of datastore so you can
access it, for example between application restarts. In that case, there is the
option of passing a custom replay handler which responds to [] and []=.
Below is a sample replay handler that stores the replay ID for each channel in memory using a Hash, stores a timestamp and has some rudimentary logic that will use one of the magic IDs depending on the value of the timestamp:
class SimpleReplayHandler
MAX_AGE = 86_400 # 24 hours
INIT_REPLAY_ID = -1
DEFAULT_REPLAY_ID = -2
def initialize
@channels = {}
@last_modified = nil
end
# This method is called during the initial subscribe phase
# in order to send the correct replay ID.
def [](channel)
if @last_modified.nil?
puts "[#{channel}] No timestamp defined, sending magic replay ID #{INIT_REPLAY_ID}"
INIT_REPLAY_ID
elsif old_replay_id?
puts "[#{channel}] Old timestamp, sending magic replay ID #{DEFAULT_REPLAY_ID}"
DEFAULT_REPLAY_ID
else
@channels[channel]
end
end
def []=(channel, replay_id)
puts "[#{channel}] Writing replay ID: #{replay_id}"
@last_modified = Time.now
@channels[channel] = replay_id
end
def old_replay_id?
@last_modified.is_a?(Time) && Time.now - @last_modified > MAX_AGE
end
endIn order to use it, simply pass the object as the value of the replay option
of the subscription:
EM.run {
# Subscribe to the PushTopic and use the custom replay handler to store any
# received replay ID.
client.subscription '/topic/AllAccounts', replay: SimpleReplayHandler.new do |message|
puts message.inspect
end
}See also:
Note: Restforce's streaming implementation is known to be compatible with version 0.8.9 of the faye gem.
The gem supports easy caching of GET requests (e.g. queries):
# rails example:
client = Restforce.new(cache: Rails.cache)
# or
Restforce.configure do |config|
config.cache = Rails.cache
endIf you enable caching, you can disable caching on a per-request basis by using .without_caching:
client.without_caching do
client.query('select Id from Account')
endIf you prefer to opt in to caching on a per-request, you can do so by using .with_caching and
setting the use_cache config option to false:
Restforce.configure do |config|
config.cache = Rails.cache
config.use_cache = false
endclient.with_caching do
client.query('select Id from Account')
endCaching is done based on your authentication credentials, so cached responses will not be shared between different Salesforce logins.
You can easily inspect what Restforce is sending/receiving by enabling logging, either globally (as below) or on a per-client basis.
Restforce.log = true
# Restforce will log to STDOUT with the `:debug` log level by default, or you can
# optionally set your own logger and log level
Restforce.configure do |config|
config.logger = Logger.new("/tmp/log/restforce.log")
config.log_level = :info
end
client = Restforce.new.query('select Id, Name from Account')Another awesome feature about restforce is that, because it is based on Faraday, you can insert your own middleware. For example, if you were using Restforce in a rails app, you can setup custom reporting to Librato using ActiveSupport::Notifications:
client = Restforce.new do |builder|
builder.insert_after Restforce::Middleware::InstanceURL,
FaradayMiddleware::Instrumentation, name: 'request.salesforce'
end
# config/initializers/notifications.rb
ActiveSupport::Notifications.subscribe('request.salesforce') do |*args|
event = ActiveSupport::Notifications::Event.new(*args)
Librato.increment 'api.salesforce.request.total'
Librato.timing 'api.salesforce.request.time', event.duration
endYou can use Restforce to decode signed requests from Salesforce. See the example app.
To use the Tooling API,
call Restforce.tooling instead of Restforce.new:
client = Restforce.tooling(...)You can use the Tooling API to add fields to existing objects. For example, add "Twitter Username" to the default "Account" object:
client = Restforce.tooling(...)
client.create!("CustomField", {
"FullName" => "Account.orgnamespace__twitter_username__c",
"Metadata" => { type: "Text", label: "Twitter Username", length: 15 },
})Here's the order of precedence from highest to lowest:
Arguments on new: passing configuration options directly as arguments has the highest precedence. These settings will override any other configuration.
Configuration block: using Restforce.configure to set configuration options is the next in line. They will take precedence over environment variables and defaults but will be overridden by direct arguments on instantiation.
Environment variables: has the lowest precedence. If you set options using environment variables, they will be overridden by any other configuration method.
Defaults: If none of the above methods are used, Restforce falls back to its default configuration values.
We welcome all contributions - they help us make Restforce the best gem possible.
See our CONTRIBUTING.md file for help with getting set up to work on the project locally.
- Fork it
- Create your feature branch (
git checkout -b my-new-feature) - Commit your changes (
git commit -am 'Added some feature') - Push to the branch (
git push origin my-new-feature) - Create your Pull Request