stream_to_s3.rb

require'looker-sdk'
require'aws-sdk-s3'
require'securerandom'

sdk=LookerSDK::Client.new(
:client_id=>ENV['LOOKERSDK_CLIENT_ID'],
:client_secret=>ENV['LOOKERSDK_CLIENT_SECRET'],
:api_endpoint=>ENV['LOOKERSDK_BASE_URL']
)

#define all the variables for Looker and S3 credentials
$query_id_for_look =14964# update the query id for the query of the Look to use
$access_key_id ="REPLACE_WITH_YOUR_IAM_KEY_ID"
$secret_access_key ="REPLACE_WITH_YOUR_IAM_SECRET_ACCESS"
$unique_name =SecureRandom.hex

schedule_hash={
:name=>$unique_name,
:query_id=> $query_id_for_look,
:run_once=>true,
:send_all_results=>true,
:scheduled_plan_destination=>[{
:format=>"csv",
:apply_formatting=>false,
:address=>"s3://looker-api-schedules",
:type=>"s3",
:parameters=>"{\"region\":\"us-east-1\",\"access_key_id\":\"#{access_key_id}\"",
:secret_parameters=>"{\"secret_access_key\":\"#{secret_access_key}\"}"
}]
}

#create the schedule
schedule=sdk.scheduled_plan_run_once(schedule_hash).inspect

#wait to run the next code to make sure that schedule has been created
sleep(3)

#get all the running queries and then loop through them to find the task assocaited with our schedule
running_queries=sdk.all_running_queries()

running_queries.eachdo |query|
ifquery[:query_id] == $query_id_for_look
 $query_task =query[:query_task_id]
end
end


$status =sdk.query_task($query_task)

#poll the query task until it is done before we go get it from s3
whiletrue
if $status[:status] == "complete"
break
end
 $status =sdk.query_task($query_task)
end


s3=Aws::S3::Resource.new(region: 'us-east-1')
bucket=s3.bucket('bucket_name')


#loop through the bucket and find the file that we created, we check to see if the unique name is in the filename
bucket.objects.limit(1000).eachdo |item|
ifitem.key.include? $unique_name
 $file_name =item.key
end
end


# Create the object to retrieve
obj=s3.bucket('looker-api-schedules').object($file_name)

# Get the item's content and save it to a file
obj.get(response_target: './output.csv')