Batch processing for the Pure Research Information System.
Add this line to your application's Gemfile:
gem 'research_metadata_batch'And then execute:
$ bundle
Or install it yourself as:
$ gem install research_metadata_batch
Uses the default gem behaviour which merely inspects the metadata models using STDOUT.
pure_config = {
url: ENV['PURE_URL'],
username: ENV['PURE_USERNAME'],
password: ENV['PURE_PASSWORD'],
api_key: ENV['PURE_API_KEY']
}
ResearchMetadataBatch::Dataset.new(pure_config: pure_config).processEither open up classes or create subclasses to implement application-specific behaviour.
This example creates subclasses and uses Amazon Web Services.
Implement methods from {ResearchMetadataBatch::Shared}.
# require aws sdk
module App
module Shared
def init(aws_config:)
# Do something with :aws_config
end
def act(model)
# Do something with Amazon Web Services
return {key1: 'some_value', key2: 'another_value', msg: 'what_happened'}
end
end
endrequire_relative 'shared'
module App
class ResearchOutput < ResearchMetadataBatch::ResearchOutput
include App::Shared
end
endrequire 'research_metadata_batch'
require_relative 'research_output'
pure_config = {
url: ENV['PURE_URL'],
username: ENV['PURE_USERNAME'],
password: ENV['PURE_PASSWORD'],
api_key: ENV['PURE_API_KEY']
}
aws_config = {
# details
}
log_file = '/path/to/your/log/file'
config = {
pure_config: pure_config,
log_file: log_file
}
batch = App::ResearchOutput.new config
batch.init aws_config: aws_config
params = {
size: 50,
typeUri: [
'/dk/atira/pure/researchoutput/researchoutputtypes/contributiontojournal/article',
'/dk/atira/pure/researchoutput/researchoutputtypes/contributiontoconference/paper'
]
}
batch.process params: params