Merge pull request #7 from MITLibraries/3_mets

Initial Mets generation

Author: JPrevost

Gemfile

@@ -7,6 +7,7 @@ gem 'devise'
 gem 'http_logger'
 gem 'jquery-rails'
 gem 'less-rails'
+gem 'nokogiri'
 gem 'omniauth-mit-oauth2'
 gem 'omniauth-oauth2', '~> 1.3.1'
 gem 'passenger'
@@ -35,4 +36,6 @@ group :test do
   gem 'minitest-reporters'
   gem 'minitest-rails'
   gem 'minitest-rails-capybara'
+  gem 'vcr'
+  gem 'webmock'
 end

Gemfile.lock

@@ -36,6 +36,7 @@ GEM
       minitest (~> 5.1)
       thread_safe (~> 0.3, >= 0.3.4)
       tzinfo (~> 1.1)
+    addressable (2.3.8)
     annotate (2.6.10)
       activerecord (>= 3.2, <= 4.3)
       rake (~> 10.4)
@@ -64,6 +65,8 @@ GEM
       term-ansicolor (~> 1.3)
       thor (~> 0.19.1)
       tins (~> 1.6.0)
+    crack (0.4.2)
+      safe_yaml (~> 1.0.0)
     debug_inspector (0.0.2)
     devise (3.5.2)
       bcrypt (~> 3.0)
@@ -81,6 +84,7 @@ GEM
       multipart-post (>= 1.2, < 3)
     globalid (0.3.6)
       activesupport (>= 4.1.0)
+    hashdiff (0.2.3)
     hashie (3.4.3)
     http-cookie (1.0.2)
       domain_name (~> 0.5)
@@ -203,6 +207,7 @@ GEM
       ruby-progressbar (~> 1.7)
       tins (<= 1.6.0)
     ruby-progressbar (1.7.5)
+    safe_yaml (1.0.4)
     simplecov (0.10.0)
       docile (~> 1.1.0)
       json (~> 1.8)
@@ -237,13 +242,18 @@ GEM
     unf (0.1.4)
       unf_ext
     unf_ext (0.0.7.1)
+    vcr (3.0.0)
     warden (1.2.3)
       rack (>= 1.0)
     web-console (2.2.1)
       activemodel (>= 4.0)
       binding_of_caller (>= 0.7.2)
       railties (>= 4.0)
       sprockets-rails (>= 2.0, < 4.0)
+    webmock (1.22.3)
+      addressable (>= 2.3.6)
+      crack (>= 0.3.2)
+      hashdiff
     xpath (2.0.0)
       nokogiri (~> 1.3)
 
@@ -262,6 +272,7 @@ DEPENDENCIES
   minitest-rails
   minitest-rails-capybara
   minitest-reporters
+  nokogiri
   omniauth-mit-oauth2
   omniauth-oauth2 (~> 1.3.1)
   passenger
@@ -273,7 +284,9 @@ DEPENDENCIES
   therubyracer
   twitter-bootstrap-rails
   uglifier (>= 1.3.0)
+  vcr
   web-console (~> 2.0)
+  webmock
 
 BUNDLED WITH
    1.10.6

app/models/epdcx.rb

@@ -0,0 +1,88 @@
+class Epdcx
+  def initialize(xml, submission)
+    @xml = xml
+    @xml['mets'].xmlData do
+      @xml['epdcx'].descriptionSet(
+        'xmlns:epdcx' => 'http://purl.org/eprint/epdcx/2006-11-16/',
+        'xsi:schemaLocation' =>
+          'http://purl.org/eprint/epdcx/xsd/2006-11-16/epdcx.xsd') do
+        work(submission)
+        entity(submission)
+      end
+    end
+  end
+
+  private
+
+  def entity(submission)
+    @xml['epdcx'].description('epdcx:resourceId' => 'sword-mets-expr-1') do
+      expression
+      lang
+      journal_article
+      statement('http://purl.org/eprint/terms/bibliographicCitation',
+                submission.journal)
+      doi(submission)
+    end
+  end
+
+  def work(submission)
+    @xml['epdcx'].description('epdcx:resourceId' => 'sword-mets-epdcx-1') do
+      scholarly_work
+      statement('http://purl.org/dc/elements/1.1/title', submission.title)
+      statement('http://purl.org/dc/elements/1.1/creator', submission.author)
+
+      @xml['epdcx'].statement('epdcx:propertyURI' =>
+                             'http://purl.org/eprint/terms/isExpressedAs') do
+        @xml['epdcx'].valueRef('sword-mets-expr-1')
+      end
+    end
+  end
+
+  def doi(submission)
+    @xml['epdcx'].statement('epdcx:propertyURI' =>
+                            'http://purl.org/dc/elements/1.1/identifier') do
+      @xml['epdcx'].valueString('epdcx:sesURI' =>
+                                'http://purl.org/dc/terms/URI') do
+        @xml.text(submission.doi)
+      end
+    end
+  end
+
+  def expression
+    @xml['epdcx'].statement('epdcx:propertyURI' =>
+                            'http://purl.org/dc/elements/1.1/type',
+                            'epdcx:valueURI' =>
+                            'http://purl.org/eprint/entityType/Expression')
+  end
+
+  def journal_article
+    @xml['epdcx'].statement('epdcx:propertyURI' =>
+                            'http://purl.org/dc/elements/1.1/type',
+                            'epdcx:vesURI' =>
+                            'http://purl.org/eprint/terms/Type',
+                            'epdcx:valueURI' =>
+                            'http://purl.org/eprint/type/JournalArticle')
+  end
+
+  def lang
+    @xml['epdcx'].statement('epdcx:propertyURI' =>
+                            'http://purl.org/dc/elements/1.1/language',
+                            'epdcx:vesURI' =>
+                            'http://purl.org/dc/terms/RFC3066') do
+      @xml['epdcx'].valueString('en')
+    end
+  end
+
+  def scholarly_work
+    @xml['epdcx'].statement('epdcx:propertyURI' =>
+                            'http://purl.org/dc/elements/1.1/type',
+                            'epdcx:valueURI' =>
+                            'http://purl.org/eprint/entityType/ScholarlyWork')
+  end
+
+  def statement(prop_uri, value)
+    @xml['epdcx'].statement('epdcx:propertyURI' => prop_uri) do
+      @xml['epdcx'].valueString(value)
+    end
+  end
+end

app/models/mets.rb

@@ -0,0 +1,70 @@
+class Mets
+  def initialize(submission)
+    @submission = submission
+    @builder = Nokogiri::XML::Builder.new do |xml|
+      xml['mets'].mets(
+        'xmlns:mets' => 'http://www.loc.gov/METS/',
+        'xmlns:xlink' => 'http://www.w3.org/1999/xlink',
+        'xmlns:epdcx' => 'http://purl.org/eprint/epdcx/2006-11-16/') do
+        overall_structure(xml)
+      end
+    end
+  end
+
+  def to_xml
+    @builder.to_xml
+  end
+
+  private
+
+  def overall_structure(xml)
+    header(xml)
+    dmd(xml)
+    filesec(xml)
+    structmap(xml)
+  end
+
+  def header(xml)
+    xml['mets'].metsHdr('CREATEDATE' => '2009-02-12T08:35:13.561-05:00') do
+      xml['mets'].agent('ROLE' => 'DISSEMINATOR',
+                        'TYPE' => 'ORGANIZATION') do
+        xml['mets'].name('MIT Libraries: QuickSubmit')
+      end
+    end
+  end
+
+  def dmd(xml)
+    xml['mets'].dmdSec('ID' => 'VAA4025-102-1-a01-dmd') do
+      xml['mets'].mdWrap('MDTYPE' => 'OTHER', 'OTHERMDTYPE' => 'EPDCX',
+                         'MIMETYPE' => 'text/xml') do
+        xml_data(xml)
+      end
+    end
+  end
+
+  def filesec(xml)
+    xml['mets'].fileSec('ID' => 'VAA4025-102-1-a02-dmd') do
+      xml['mets'].fileGrp('USE' => 'CONTENT') do
+        xml['mets'].file('ID' => 'JHEP09',
+                         'MIMETYPE' => 'application/pdf') do
+          xml['mets'].FLocat('LOCTYPE' => 'URL',
+                             'xlink:href' => 'JHEP09(2015)175_pdfa.pdf')
+        end
+      end
+    end
+  end
+
+  def structmap(xml)
+    xml['mets'].structMap('ID' => 'VAA4025-102-1-a03-dmd') do
+      xml['mets'].div('DMDID' => 'sword-mets-dmd-1') do
+        xml['mets'].div('TYPE' => 'File') do
+          xml['mets'].fptr('FILEID' => 'JHEP09')
+        end
+      end
+    end
+  end
+
+  def xml_data(xml)
+    Epdcx.new(xml, @submission)
+  end
+end

app/models/submission.rb

@@ -20,4 +20,8 @@ class Submission < ActiveRecord::Base
   validates :user, presence: true
   validates :title, presence: true
   validates :agreed_to_license, inclusion: { in: [true] }
+
+  def to_mets
+    Mets.new(self).to_xml
+  end
 end

test/models/mets_test.rb

@@ -0,0 +1,25 @@
+require 'test_helper'
+
+class MetsTest < ActiveSupport::TestCase
+  test 'initializing with a full submission creates valid mets xml' do
+    VCR.use_cassette('mets_dtd') do
+      dtd = Net::HTTP.get(URI('http://www.loc.gov/standards/mets/version111/mets.xsd'))
+      sub = submissions(:sub_two)
+      xsd = Nokogiri::XML::Schema(dtd)
+      xml = Mets.new(sub).to_xml
+      doc = Nokogiri::XML(xml)
+      assert_equal(true, xsd.valid?(doc))
+    end
+  end
+
+  test 'initializing with a sparse submission creates valid mets xml' do
+    VCR.use_cassette('mets_dtd') do
+      dtd = Net::HTTP.get(URI('http://www.loc.gov/standards/mets/version111/mets.xsd'))
+      sub = submissions(:sub_one)
+      xsd = Nokogiri::XML::Schema(dtd)
+      xml = Mets.new(sub).to_xml
+      doc = Nokogiri::XML(xml)
+      assert_equal(true, xsd.valid?(doc))
+    end
+  end
+end

test/models/submission_test.rb

@@ -40,4 +40,15 @@ class SubmissionTest < ActiveSupport::TestCase
     sub.agreed_to_license = false
     assert_not sub.valid?
   end
+
+  test '#mets' do
+    VCR.use_cassette('mets_dtd') do
+      dtd = Net::HTTP.get(
+        URI('http://www.loc.gov/standards/mets/version111/mets.xsd'))
+      sub = submissions(:sub_two)
+      xsd = Nokogiri::XML::Schema(dtd)
+      doc = Nokogiri::XML(sub.to_mets)
+      assert_equal(true, xsd.valid?(doc))
+    end
+  end
 end

test/test_helper.rb

@@ -12,6 +12,11 @@
 require 'minitest/reporters'
 Minitest::Reporters.use!
 
+VCR.configure do |config|
+  config.cassette_library_dir = 'test/vcr_cassettes'
+  config.hook_into :webmock
+end
+
 module ActiveSupport
   class TestCase
     # Setup all fixtures in test/fixtures/*.yml for all tests in alpha order.