SalesforceBulk is an easy to use Ruby gem for connecting to and using the Salesforce Bulk API. This is a rewrite and separate release of Jorge Valdivia's salesforce_bulk gem (renamed salesforcebulk
) with full unit tests and full API capability (e.g. adding multiple batches per job).
Install SalesforceBulk from RubyGems:
gem install salesforcebulk
Or include it in your project's Gemfile
with Bundler:
gem 'salesforcebulk'
To contribute, fork this repo, create a topic branch, add changes and tests, then send a pull request. To setup the project and run tests in your fork, just do:
bundle install
bundle exec rake
When retrieving a password you will also be given a security token. Combine the two into a single value as the API treats this as your real password.
require 'salesforce_bulk'
client = SalesforceBulk::Client.new(username: 'MyUsername', password: 'MyPasswordWithSecurtyToken')
client.authenticate
Optional keys include login_host
(default is 'login.salesforce.com') and version
(default is '24.0').
An important note about the data in any of the examples below: each hash in a data set must have the same set of keys. If you need to have logic to not include certain values simply specify a nil value for a key rather than not including the key-value pair.
data1 = [{:Name__c => 'Test 1'}, {:Name__c => 'Test 2'}]
data2 = [{:Name__c => 'Test 3'}, {:Name__c => 'Test 4'}]
job = client.add_job(:insert, :MyObject__c)
# easily add multiple batches to a job
batch = client.add_batch(job.id, data1)
batch = client.add_batch(job.id, data2)
job = client.close_job(job.id) # or use the abort_job(id) method
When adding a job you can specify the following operations for the first argument:
:delete
:insert
:update
:upsert
:query
When using the :upsert operation you must specify an external ID field name:
job = client.add_job(:upsert, :MyObject__c, :external_id_field_name => :MyId__c)
For any operation you should be able to specify a concurrency mode. The default is Parallel
. The only other choice is Serial
.
job = client.add_job(:upsert, :MyObject__c, :concurrency_mode => :Serial, :external_id_field_name => :MyId__c)
The Job object has various properties such as status, created time, number of completed and failed batches and various other values.
job = client.job_info(jobId) # returns a Job object
puts "Job #{job.id} is closed." if job.closed? # other: open?, aborted?
The Batch object has various properties such as status, created time, number of processed and failed records and various other values.
batch = client.batch_info(jobId, batchId) # returns a Batch object
puts "Batch #{batch.id} is in progress." if batch.in_progress?
batches = client.batch_info_list(jobId) # returns an Array of Batch objects
batches.each do |batch|
puts "Batch #{batch.id} completed." if batch.completed? # other: failed?, in_progress?, queued?
end
To verify that a batch completed successfully or failed call the batch_info
or batch_info_list
methods first, otherwise if you call batch_result
without verifying and the batch failed the method will raise an error.
The object returned from the following example only applies to the operations: delete
, insert
, update
and upsert
. Query results are handled differently.
results = client.batch_result(jobId, batchId) # returns an Array of BatchResult objects
results.each do |result|
puts "Item #{result.id} had an error of: #{result.error}" if result.error?
end
To verify that a batch completed successfully or failed call the batch_info
or batch_info_list
methods first, otherwise if you call batch_result
without verifying and the batch failed the method will raise an error.
Query results are handled differently as its possible that a single batch could return multiple results if objects returned are large enough. Note: I haven't been able to replicate this behavior but in a fork by @WWJacob has discovered that multiple results can be returned.
# returns a QueryResultCollection object (an Array)
results = client.batch_result(jobId, batchId)
while results.any?
# Assuming query was: SELECT Id, Name, CustomField__c FROM Account
results.each do |result|
puts result[:Id], result[:Name], result[:CustomField__c]
end
puts "Another set is available." if results.next?
results.next
end
Note: By reviewing the API docs and response format my understanding was that the API would return multiple results sets for a single batch if the query was to large but this does not seem to be the case in my live testing. It seems to be capped at 10000 records (as it when inserting data) but I haven't been able to verify through the documentation. If you know anything about that your input is appreciated. In the meantime the gem was built to support multiple result sets for a query batch but seems that will change which will simplify that method.
To release a new version, update the version number in version.rb
, and then run bundle exec rake release
, which will create a git tag for the version, push git commits and tags, and push the .gem
file to rubygems.org.
- Support for other Ruby platforms
- Clean up/reorganize tests better
- Rdocs
To release a new version, update the version number in version.rb
, and then run bundle exec rake release
, which will create a git tag for the version, push git commits and tags, and push the .gem
file to rubygems.org.
Bug reports and pull requests are welcome on GitHub at https://github.com/javierjulio/salesforce_bulk. 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.