linked_research_metadata 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 7c1a8e70e56d0e27192ea504c47611805d45d4b9
+  data.tar.gz: 496d1fb8a316cc68e05ee2a310bd94a83b9a600e
+SHA512:
+  metadata.gz: d907f9dd473fe7ea2248c2e28bddd92f66be1fce737da20ae75f5a9a11facc108d9c3ed203ec78b7c11ff0544973398ac6f5ed2c11d705e8f95835b529c8202e
+  data.tar.gz: b71ef79fbe3cfcf6355991948669119b886b6ab896a60258a86b1d7c3981d7aa6b79e5b2f778e3484d05f5f55626fd313594b9e11077ee1c8b0f978926e465ae

data/.gitignore

@@ -0,0 +1,18 @@
+*.gem
+*.rbc
+.bundle
+.config
+.yardoc
+Gemfile.lock
+InstalledFiles
+_yardoc
+coverage
+doc/
+lib/bundler/man
+pkg
+rdoc
+spec/reports
+test/tmp
+test/version_tmp
+tmp
+.idea

data/CHANGELOG.md

@@ -0,0 +1,9 @@
+# Change Log
+All notable changes to this project will be documented in this file.
+This project adheres to [Semantic Versioning](http://semver.org/).
+
+## Unreleased
+
+## 0.1.0 - 2017-06-30
+### Added
+- Working product supports datasets.

data/Gemfile

@@ -0,0 +1,4 @@
+source "https://rubygems.org"
+
+# Specify your gem's dependencies in linked_research_metadata.gemspec
+gemspec

data/MODEL.md

@@ -0,0 +1,69 @@
+# Data Model
+Version 0.1.0
+
+## Prefixes
+Prefix | URI
+--- | ---
+dcterms | http://purl.org/dc/terms/
+foaf | http://xmlns.com/foaf/0.1/
+mads | http://www.loc.gov/mads/rdf/v1#
+owl | http://www.w3.org/2002/07/owl#
+premis | http://www.loc.gov/premis/rdf/v1#
+rdf | http://www.w3.org/1999/02/22-rdf-syntax-ns#
+swpo | http://sw-portal.deri.org/ontologies/swportal#
+vivo | http://vivoweb.org/ontology/core#
+xsd | http://www.w3.org/2001/XMLSchema#
+
+## Resources
+
+### Dataset
+Property | Value | Cardinality
+--- | --- | ---
+dcterms:available | xsd:date | 1
+dcterms:created | xsd:date | 1
+dcterms:creator | URI | 1..n
+dcterms:contributor | URI | 0..n
+dcterms:description | Literal (String) | 0..1
+dcterms:hasPart | URI | 0..n
+dcterms:identifier | URI. DOI. | 0..1
+dcterms:relation | URI | 0..n
+dcterms:spatial | Literal (String) | 0..n
+dcterms:subject | Literal (String) | 0..n
+dcterms:temporal | Literal (String). RKMS‐ISO8601 form. | 0..1
+dcterms:title | Literal (String) | 1
+rdf:type | vivo:Dataset | 1
+
+### File
+Property | Value | Cardinality
+--- | --- | ---
+dcterms:extent | Literal (Integer) | 1
+dcterms:format | Literal (String) | 1
+dcterms:license | URI | 0..1
+dcterms:title | Literal (String) | 1
+rdf:type | premis:File | 1
+
+### Organisation
+Property | Value | Cardinality
+--- | --- | ---
+dcterms:title | Literal (String) | 1
+rdf:type | foaf:Organization | 1
+
+### Person
+Property | Value | Cardinality
+--- | --- | ---
+foaf:name | Literal (String) | 1
+mads:hasAffiliation | URI | 0..n
+rdf:type | foaf:Person | 1
+vivo:OrcidId | URI | 0..1
+
+### Project
+Property | Value | Cardinality
+--- | --- | ---
+dcterms:title | Literal (String) | 1
+rdf:type | vivo:Project | 1
+
+### Publication
+Property | Value | Cardinality
+--- | --- | ---
+dcterms:title | Literal (String) | 1
+rdf:type | swpo:Publication | 1

data/README.md

@@ -0,0 +1,63 @@
+# LinkedResearchMetadata
+
+Metadata extraction from the Pure Research Information System and transformation of the metadata into RDF.
+
+## Status
+
+[![Gem Version](https://badge.fury.io/rb/linked_research_metadata.svg)](https://badge.fury.io/rb/linked_research_metadata)
+[![Build Status](https://semaphoreci.com/api/v1/aalbinclark/linked_research_metadata/branches/master/badge.svg)](https://semaphoreci.com/aalbinclark/linked_research_metadata)
+[![Code Climate](https://codeclimate.com/github/lulibrary/linked_research_metadata/badges/gpa.svg)](https://codeclimate.com/github/lulibrary/linked_research_metadata)
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+    gem 'linked_research_metadata'
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install linked_research_metadata
+
+## Usage
+
+### Configuration
+
+Create a hash for passing to a transformer.
+
+```ruby
+# Pure host with authentication.
+config = {
+  url:      ENV['PURE_URL'],
+  username: ENV['PURE_USERNAME'],
+  password: ENV['PURE_PASSWORD'],
+  minting_uri: 'http://data.example.com'
+}
+```
+
+```ruby
+# Pure host without authentication.
+config = {
+  url: ENV['PURE_URL'],
+  minting_uri: 'http://data.example.com'
+}
+```
+
+### Transformation
+
+Create a metadata transformer for a Pure dataset.
+
+```ruby
+transformer = LinkedResearchMetadata::Transformer::Dataset.new config
+```
+
+Give it a Pure identifier...
+
+```ruby
+graph = transformer.transform uuid: 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx'
+```
+
+...and get an RDF graph.

data/Rakefile

@@ -0,0 +1,2 @@
+require "bundler/gem_tasks"
+task :default => :spec

data/lib/linked_research_metadata/transformer/base.rb

@@ -0,0 +1,44 @@
+module LinkedResearchMetadata
+  module Transformer
+
+    # Base transformer
+    #
+    class Base
+
+      # @param config [Hash]
+      # @option config [String] :url The URL of the Pure host.
+      # @option config [String] :username The username of the Pure host account.
+      # @option config [String] :password The password of the Pure host account.
+      # @option config [String] :minting_uri The URI at which to mint a resource.
+      def initialize(config)
+        @config = config
+        raise 'Minting URI missing' if @config[:minting_uri].empty?
+        @graph = RDF::Graph.new
+      end
+
+      private
+
+      def vocab(v)
+        vocabulary_map =
+        {
+            swpo: 'http://sw-portal.deri.org/ontologies/swportal#',
+            vivo: 'http://vivoweb.org/ontology/core#'
+        }
+        RDF::Vocabulary.new vocabulary_map[v]
+      end
+
+      def mint_uri(uuid, resource)
+        uri_resource_map = {
+            dataset: 'datasets',
+            file: 'files',
+            organisation: 'organisations',
+            person: 'people',
+            project: 'projects',
+            publication: 'publications'
+        }
+        File.join @config[:minting_uri], uri_resource_map[resource], uuid
+      end
+
+    end
+  end
+end

data/lib/linked_research_metadata/transformer/dataset.rb

@@ -0,0 +1,223 @@
+module LinkedResearchMetadata
+  module Transformer
+
+    # Dataset transformer
+    #
+    class Dataset < Base
+
+      # @param config [Hash]
+      # @option config [String] :url The URL of the Pure host.
+      # @option config [String] :username The username of the Pure host account.
+      # @option config [String] :password The password of the Pure host account.
+      # @option config [String] :minting_uri The URI at which to mint a resource.
+      def initialize(config)
+        super
+      end
+
+      # Dataset transformation
+      #
+      # @param uuid [String]
+      # @return [RDF::Graph]
+      def transform(uuid:)
+        dataset_extractor = Puree::Extractor::Dataset.new @config
+        @resource = dataset_extractor.find uuid: uuid
+        raise 'No metadata for ' + uuid if !@resource
+        dataset_uri = mint_uri uuid, :dataset
+        @resource_uri = RDF::URI.new dataset_uri
+        build_graph
+        @graph
+      end
+
+      private
+
+      def meta
+        @graph << [ @resource_uri, RDF.type, "#{vocab(:vivo)}Dataset" ]
+      end
+
+      def available
+        object = @resource.available
+        if object
+          object_literal = RDF::Literal.new(object.strftime("%F"), datatype: RDF::XSD.date)
+          @graph << [ @resource_uri, RDF::Vocab::DC.available, object_literal ]
+        end
+      end
+
+      def created
+        object =  @resource.created
+        if object
+          object_literal = RDF::Literal.new(object.strftime("%F"), datatype: RDF::XSD.date)
+          @graph << [ @resource_uri, RDF::Vocab::DC.created, object_literal ]
+        end
+      end
+
+      def doi
+        if @resource.doi
+          doi_uri = RDF::URI.new @resource.doi
+          doi_predicate_uri = RDF::Vocab::DC.identifier
+          @graph << [ @resource_uri, doi_predicate_uri, doi_uri ]
+        end
+      end
+
+      def description
+        object = @resource.description
+        if object
+          @graph << [ @resource_uri, RDF::Vocab::DC.description, object ]
+        end
+      end
+
+      def files
+        @resource.files.each do |i|
+          file_uri = RDF::URI.new mint_uri(SecureRandom.uuid, :file)
+
+          @graph << [ @resource_uri, RDF::Vocab::DC.hasPart, file_uri ]
+
+          # license
+          if i.license && i.license.url
+              uri = RDF::URI.new i.license.url
+              @graph << [ file_uri, RDF::Vocab::DC.license, uri ]
+          end
+
+          # mime
+          @graph << [ file_uri, RDF::Vocab::DC.format, i.mime ]
+
+          # size
+          size_predicate = RDF::Vocab::DC.extent
+          @graph << [ file_uri, size_predicate, i.size ]
+
+          #name
+          @graph << [ file_uri, RDF::Vocab::DC.title, i.name ]
+
+          #type
+          @graph << [ file_uri, RDF.type, RDF::Vocab::PREMIS.File ]
+        end
+      end
+
+      def keywords
+        @resource.keywords.each do |i|
+          @graph << [ @resource_uri, RDF::Vocab::DC.subject, i ]
+        end
+      end
+
+      def person(person_uri, uuid, name)
+        @graph << [ person_uri, RDF.type, RDF::Vocab::FOAF.Person ]
+        @graph << [ person_uri, RDF::Vocab::FOAF.name, name ]
+        person_extractor = Puree::Extractor::Person.new @config
+        person = person_extractor.find uuid: uuid
+        if person
+          person.affiliations.each do |i|
+            organisation_uri = RDF::URI.new mint_uri(i.uuid, :organisation)
+            @graph << [ person_uri, RDF::Vocab::MADS.hasAffiliation, organisation_uri ]
+            @graph << [ organisation_uri, RDF::Vocab::DC.title, i.name ]
+            @graph << [ organisation_uri, RDF.type, RDF::Vocab::FOAF.Organization ]
+          end
+          if person.orcid
+            orcid_uri = RDF::URI.new "http://orcid.org/#{person.orcid}"
+            orcid_predicate_uri = RDF::URI.new "#{vocab(:vivo)}OrcidId"
+            @graph << [ person_uri, orcid_predicate_uri, orcid_uri ]
+          end
+        end
+      end
+
+      def projects
+        @resource.projects.each do |i|
+          project_uri = RDF::URI.new mint_uri(i.uuid, :project)
+          @graph << [ @resource_uri, RDF::Vocab::DC.relation, project_uri ]
+          @graph << [ project_uri, RDF::Vocab::DC.title, i.title ]
+          @graph << [ project_uri, RDF.type, "#{vocab(:vivo)}Project" ]
+        end
+      end
+
+      def publications
+        @resource.publications.each do |i|
+          if i.type == 'Dataset'
+            publication_uri = RDF::URI.new mint_uri(i.uuid, :dataset)
+            @graph << [ publication_uri, RDF.type, "#{vocab(:vivo)}Dataset" ]
+          else
+            publication_uri = RDF::URI.new mint_uri(i.uuid, :publication)
+            @graph << [ publication_uri, RDF.type, "#{vocab(:swpo)}Publication" ]
+          end
+          @graph << [ @resource_uri, RDF::Vocab::DC.relation, publication_uri ]
+          @graph << [ publication_uri, RDF::Vocab::DC.title, i.title ]
+        end
+      end
+
+      def publisher
+        # should be URI
+      end
+
+      def roles
+        all_persons = []
+        all_persons << @resource.persons_internal
+        all_persons << @resource.persons_external
+        all_persons << @resource.persons_other
+        all_persons.each do |person_type|
+          person_type.each do |i|
+            name = i.name.first_last
+            if i.uuid
+              uuid = i.uuid
+            else
+              uuid = SecureRandom.uuid
+            end
+            person_uri = RDF::URI.new mint_uri(uuid, :person)
+            if i.role == 'Creator'
+              @graph << [ @resource_uri, RDF::Vocab::DC.creator, person_uri ]
+              person person_uri, uuid, name
+            end
+            if i.role == 'Contributor'
+              @graph << [ @resource_uri, RDF::Vocab::DC.contributor, person_uri ]
+              person person_uri, uuid, name
+            end
+          end
+        end
+
+      end
+
+      def spatial
+        @resource.spatial_places.each do |i|
+          @graph << [ @resource_uri, RDF::Vocab::DC.spatial, i ]
+        end
+      end
+
+      def temporal
+        t = @resource.temporal
+        temporal_range = ''
+        if t
+          if t.start
+            temporal_range << t.start.strftime("%F")
+            if t.end
+              temporal_range << '/'
+              temporal_range << t.end.strftime("%F")
+            end
+            object = temporal_range
+            @graph << [ @resource_uri, RDF::Vocab::DC.temporal, object ]
+          end
+        end
+      end
+
+      def title
+        object = @resource.title
+        if object
+          @graph << [ @resource_uri, RDF::Vocab::DC.title, object ]
+        end
+      end
+
+      def build_graph
+        meta
+        available
+        created
+        description
+        doi
+        files
+        keywords
+        projects
+        publications
+        # publisher
+        roles
+        spatial
+        temporal
+        title
+      end
+
+    end
+  end
+end

data/lib/linked_research_metadata/transformer/transformer.rb

@@ -0,0 +1,10 @@
+module LinkedResearchMetadata
+
+  # Extracts metadata from the Pure Research Information System and
+  # converts it into RDF.
+  #
+  module Transformer
+
+  end
+
+end

data/lib/linked_research_metadata/version.rb

@@ -0,0 +1,5 @@
+module LinkedResearchMetadata
+  # Semantic version number
+  #
+  VERSION = "0.1.0"
+end

data/lib/linked_research_metadata.rb

@@ -0,0 +1,12 @@
+require 'puree'
+require 'linkeddata'
+require 'linked_research_metadata/transformer/base'
+require 'linked_research_metadata/transformer/dataset'
+require 'linked_research_metadata/version'
+
+# Metadata extraction from the Pure Research Information System and
+# transformation of the metadata into RDF.
+#
+module LinkedResearchMetadata
+
+end

data/linked_research_metadata.gemspec

@@ -0,0 +1,29 @@
+# coding: utf-8
+lib = File.expand_path("../lib", __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require "linked_research_metadata/version"
+
+Gem::Specification.new do |spec|
+  spec.name          = "linked_research_metadata"
+  spec.version       = LinkedResearchMetadata::VERSION
+  spec.authors       = ["Adrian Albin-Clark"]
+  spec.email         = ["a.albin-clark@lancaster.ac.uk"]
+
+  spec.summary       = %q{Metadata extraction from the Pure Research Information System and transformation of the metadata into RDF.}
+  spec.homepage      = "https://github.com/lulibrary/linked_research_metadata"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files -z`.split("\x0").reject do |f|
+    f.match(%r{^(test|spec|features)/})
+  end
+
+  spec.require_paths = ["lib"]
+
+  spec.required_ruby_version = '~> 2.3'
+
+  spec.add_runtime_dependency "puree", "~> 1.3"
+  spec.add_runtime_dependency "linkeddata", "~> 2.2"
+
+  spec.add_development_dependency "minitest-reporters", "~> 1.1"
+
+end

metadata

@@ -0,0 +1,100 @@
+--- !ruby/object:Gem::Specification
+name: linked_research_metadata
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Adrian Albin-Clark
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2017-06-30 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: puree
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.3'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.3'
+- !ruby/object:Gem::Dependency
+  name: linkeddata
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.2'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.2'
+- !ruby/object:Gem::Dependency
+  name: minitest-reporters
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.1'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.1'
+description: 
+email:
+- a.albin-clark@lancaster.ac.uk
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- CHANGELOG.md
+- Gemfile
+- MODEL.md
+- README.md
+- Rakefile
+- lib/linked_research_metadata.rb
+- lib/linked_research_metadata/transformer/base.rb
+- lib/linked_research_metadata/transformer/dataset.rb
+- lib/linked_research_metadata/transformer/transformer.rb
+- lib/linked_research_metadata/version.rb
+- linked_research_metadata.gemspec
+homepage: https://github.com/lulibrary/linked_research_metadata
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - "~>"
+    - !ruby/object:Gem::Version
+      version: '2.3'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.5.1
+signing_key: 
+specification_version: 4
+summary: Metadata extraction from the Pure Research Information System and transformation
+  of the metadata into RDF.
+test_files: []
+has_rdoc: