qa 0.0.1

data/config/initializers/authorities.rb

@@ -0,0 +1 @@
+AUTHORITIES_CONFIG = YAML.load_file(File.expand_path("../../authorities.yml", __FILE__))

data/config/oclcts-authorities.yml

@@ -0,0 +1,24 @@
+url-pattern:
+  prefix-query: http://tspilot.oclc.org/{authority-id}/?query=oclcts.rootHeading+exact+%22{query}*%22&version=1.1&operation=searchRetrieve&recordSchema=http%3A%2F%2Fzthes.z3950.org%2Fxml%2F1.0%2F&maximumRecords=10&startRecord=1&resultSetTTL=300&recordPacking=xml&recordXPath=&sortKeys=
+  id-lookup: http://tspilot.oclc.org/{authority-id}/?query=dc.identifier+exact+%22{id}%22&version=1.1&operation=searchRetrieve&recordSchema=http%3A%2F%2Fzthes.z3950.org%2Fxml%2F1.0%2F&maximumRecords=10&startRecord=1&resultSetTTL=300&recordPacking=xml&recordXPath=&sortKeys=
+authorities:
+  lcgft:
+    name: Library of Congress Genre/Form Terms for Library and Archival Materials (LCGFT)
+  bisacsh:
+    name: Book Industry Study Group Subject Headings (BISAC®). Used with permission.
+  fast:
+    name: Faceted Application of Subject Terminology (FAST subject headings)
+  gsafd:
+    name: Form and genre headings for fiction and drama
+  lcshac:
+    name: Library of Congress AC Subject Headings
+  lcsh:
+    name: Library of Congress Subject Headings
+  mesh:
+    name: Medical Subject Headings (MeSH®)
+  lctgm:
+    name: "Thesaurus for graphic materials: TGM I, Subject terms"
+  gmgpc:
+    name: "Thesaurus for graphic materials: TGM II, Genre terms"
+  meta:
+    name: Controlled Vocabulary Metadata

data/config/routes.rb

@@ -0,0 +1,7 @@
+Qa::Engine.routes.draw do
+  match "/search/:vocab" => "terms#search", :via=>:get
+  match "/search/:vocab/:sub_authority" => "terms#search", :via=>:get
+  match "/terms/:vocab" => "terms#index", :via=>:get
+  match "/terms/:vocab/:sub_authority" => "terms#index", :via=>:get
+  match "/terms" => "terms#index", :via=>:get
+end

data/db/migrate/20130917200611_create_qa_subject_mesh_terms.rb

@@ -0,0 +1,11 @@
+class CreateQaSubjectMeshTerms < ActiveRecord::Migration
+  def change
+    create_table :qa_subject_mesh_terms do |t|
+      t.string :term_id
+      t.string :term
+      t.text :synonyms
+    end
+    add_index :qa_subject_mesh_terms, :term_id
+    add_index :qa_subject_mesh_terms, :term
+  end
+end

data/db/migrate/20130917201026_create_qa_mesh_tree.rb

@@ -0,0 +1,10 @@
+class CreateQaMeshTree < ActiveRecord::Migration
+  def change
+    create_table :qa_mesh_trees do |t|
+      t.string :term_id
+      t.string :tree_number
+    end
+    add_index :qa_mesh_trees, :term_id
+    add_index :qa_mesh_trees, :tree_number
+  end
+end

data/db/migrate/20130918141523_add_term_lower_to_qa_subject_mesh_terms.rb

@@ -0,0 +1,7 @@
+class AddTermLowerToQaSubjectMeshTerms < ActiveRecord::Migration
+  def change
+    add_column :qa_subject_mesh_terms, :term_lower, :string
+    add_index :qa_subject_mesh_terms, :term_lower
+    remove_index :qa_subject_mesh_terms, :term
+  end
+end

data/lib/qa.rb

@@ -0,0 +1,9 @@
+require "qa/engine"
+require "active_record"
+require "activerecord-import"
+
+module Qa
+  extend ActiveSupport::Autoload
+
+  autoload :Authorities
+end

data/lib/qa/authorities.rb

@@ -0,0 +1,12 @@
+module Qa::Authorities
+  extend ActiveSupport::Autoload
+
+  autoload :Base
+  autoload :Lcsh
+  autoload :Loc
+  autoload :Local
+  autoload :Mesh
+  autoload :MeshTools
+  autoload :Oclcts
+  autoload :Tgnlang
+end

data/lib/qa/authorities/base.rb

@@ -0,0 +1,50 @@
+require 'curl'
+
+module Qa::Authorities
+  class Base
+    attr_accessor :response, :query_url, :raw_response
+
+    def initialize(q, sub_authority='')
+      # Implement Me and set self.query_url
+
+      if self.query_url == nil
+        raise Exception 'query url in your authorities lib is not set (implement in initialize)'
+      end
+
+      #Default implementation assumed query_url is set
+      http = Curl.get(self.query_url) do |http|
+        http.headers['Accept'] = 'application/json'
+      end
+
+      self.raw_response = JSON.parse(http.body_str)
+    end
+
+    def self.authority_valid?(sub_authority)
+      sub_authority == nil || sub_authorities.include?(sub_authority)
+    end
+
+    def self.sub_authorities
+      [] #Overwrite if you have sub_authorities
+    end
+
+    def parse_authority_response
+      # Overwrite me unless your raw response needs no parsing
+      self.response = self.raw_response
+
+    end
+
+    def get_full_record(id)
+      # implement me
+      {"id"=>id}.to_json
+    end
+
+    # Parse the result from LOC, and return an JSON array of terms that match the query.
+    def results
+      self.response.to_json
+    end
+
+    # TODO: there's other info in the self.response that might be worth making access to, such as
+    # RDF links, etc.
+
+  end
+end

data/lib/qa/authorities/lcsh.rb

@@ -0,0 +1,46 @@
+require 'uri'
+
+module Qa::Authorities
+  class Lcsh < Qa::Authorities::Base
+
+    # Initialze the Lcsh class with a query and get the http response from LOC's server.
+    # This is set to a JSON object
+    def initialize(q, sub_authority='')
+      self.query_url= "http://id.loc.gov/authorities/suggest/?q=" + q
+
+      super
+    end
+
+    # Format response to the correct JSON structure
+    def parse_authority_response
+      self.response = build_response
+    end
+
+    def query
+      self.raw_response[0]
+    end
+
+    def suggestions
+      self.raw_response[1]
+    end
+
+    def urls_for_suggestions
+      self.raw_response[3]
+    end
+
+    private
+
+    def build_response a = Array.new
+      self.suggestions.each_index do |i|
+        a << {"id"=>get_id_from_url(urls_for_suggestions[i]), "label"=>suggestions[i]}
+      end
+      return a
+    end
+
+    def get_id_from_url url
+      uri = URI(url)
+      return uri.path.split(/\//).last
+    end
+
+  end
+end

data/lib/qa/authorities/loc.rb

@@ -0,0 +1,200 @@
+require 'uri'
+
+module Qa::Authorities
+  class Loc < Qa::Authorities::Base
+
+    # Initialze the Loc class with a query and get the http response from LOC's server.
+    # This is set to a JSON object
+    def initialize(q, sub_authority='')
+      q += '*'
+      authority_url = sub_authorityURL(sub_authority)
+      self.query_url =  "http://id.loc.gov/search/?q=#{q}&q=#{authority_url}&format=json"
+
+      super
+    end
+
+    def sub_authorityURL(sub_authority)
+      vocab_base_url = 'cs%3Ahttp%3A%2F%2Fid.loc.gov%2Fvocabulary%2F'
+      authority_base_url = 'cs%3Ahttp%3A%2F%2Fid.loc.gov%2Fauthorities%2F'
+      datatype_base_url = 'cs%3Ahttp%3A%2F%2Fid.loc.gov%2Fdatatypes%2F'
+      vocab_preservation_base_url = 'cs%3Ahttp%3A%2F%2Fid.loc.gov%2Fvocabulary%2Fpreservation%2F'
+
+      case sub_authority
+        when 'subjects'
+          return authority_base_url + URI.escape(sub_authority)
+        when 'names'
+          return authority_base_url + URI.escape(sub_authority)
+        when 'classification'
+          return authority_base_url + URI.escape(sub_authority)
+        when 'childrensSubjects'
+          return authority_base_url + URI.escape(sub_authority)
+        when 'genreForms'
+          return authority_base_url + URI.escape(sub_authority)
+        when 'graphicMaterials'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'organizations'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'relators'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'countries'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'geographicAreas'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'languages'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'iso639-1'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'iso639-2'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'iso639-5'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'edtf'
+          return datatype_base_url + URI.escape(sub_authority)
+        when 'preservation'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'actionsGranted'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'agentType'
+          return vocab_base_url + URI.escape(sub_authority)
+        when 'contentLocationType'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'copyrightStatus'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'cryptographicHashFunctions'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'environmentCharacteristic'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'environmentPurpose'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'eventRelatedAgentRole'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'eventRelatedObjectRole'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'eventType'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'formatRegistryRole'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'hardwareType'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'inhibitorTarget'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'inhibitorType'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'objectCategory'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'preservationLevelRole'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'relationshipSubType'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'relationshipType'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'rightsBasis'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'rightsRelatedAgentRole'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'signatureEncoding'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'signatureMethod'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'softwareType'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        when 'storageMedium'
+          return vocab_preservation_base_url + URI.escape(sub_authority)
+        else
+          return ''
+      end
+    end
+
+    def self.sub_authorities
+      ['iso639-2', 'subjects', 'names', 'classification', 'childrensSubjects', 'genreForms', 'graphicMaterials', 'organizations', 'relators', 'countries', 'geographicAreas', 'languages', 'iso639-5', 'edtf', 'preservation', 'actionsGranted', 'agentType', 'contentLocationType', 'copyrightStatus', 'cryptographicHashFunctions', 'environmentCharacteristic', 'environmentPurpose', 'eventRelatedAgentRole', 'eventRelatedObjectRole', 'eventType', 'formatRegistryRole', 'hardwareType', 'inhibitorTarget', 'inhibitorType', 'objectCategory', 'preservationLevelRole', 'relationshipSubType', 'relationshipType', 'rightsBasis', 'rightsRelatedAgentRole', 'signatureEncoding', 'signatureMethod', 'softwareType', 'storageMedium']
+    end
+
+
+    def parse_authority_response
+      result = []
+      self.raw_response.each do |single_response|
+        if single_response[0] == "atom:entry"
+          id = nil
+          label = ''
+          single_response.each do |result_part|
+            if(result_part[0] == 'atom:title')
+              label = result_part[2]
+            end
+
+            if(result_part[0] == 'atom:id')
+              id = result_part[2]
+            end
+
+          end
+
+          id ||= label
+          result << {"id"=>id, "label"=>label}
+
+        end
+      end
+      self.response = result
+    end
+
+    def get_full_record(id)
+      full_record = nil
+      parsed_result = {}
+      self.raw_response.each do |single_response|
+        if single_response[0] == "atom:entry"
+
+          single_response.each do |result_part|
+            if(result_part[0] == 'atom:title')
+              if id == result_part[2]
+                full_record = single_response
+              end
+            end
+
+            if(result_part[0] == 'atom:id')
+              if id == result_part[2]
+                full_record = single_response
+              end
+            end
+
+          end
+
+        end
+      end
+
+
+      if full_record != nil
+        full_record.each do |section|
+          if section.class == Array
+            label = section[0].split(':').last.to_s
+            case label
+              when 'title'
+                parsed_result[label] = section[2]
+              when 'link'
+                if section[1]['type'] != nil
+                  parsed_result[label + "||#{section[1]['type']}"] = section[1]["href"]
+                else
+                  parsed_result[label] = section[1]["href"]
+                end
+              when 'id'
+                parsed_result[label] = section[2]
+              when 'author'
+                author_list = []
+                #FIXME: Find example with two authors to better understand this data.
+                author_list << section[2][2]
+                parsed_result[label] = author_list
+              when 'updated'
+                parsed_result[label] = section[2]
+              when 'created'
+                parsed_result[label] = section[2]
+            end
+
+          end
+        end
+      else
+        raise Exception 'Lookup without using a result search first not implemented yet'
+      end
+
+      parsed_result
+
+    end
+
+  end
+end

data/lib/qa/authorities/local.rb

@@ -0,0 +1,69 @@
+module Qa::Authorities
+  class Local < Qa::Authorities::Base
+    
+    attr_accessor :response
+    
+    def initialize(q, sub_authority)
+      begin
+        sub_authority_hash = YAML.load(File.read(File.join(Qa::Authorities::Local.sub_authorities_path, "#{sub_authority}.yml")))
+      rescue
+        sub_authority_hash = {}
+      end
+      @terms = normalize_terms(sub_authority_hash.fetch(:terms, []))
+      if q.blank?
+        self.response = @terms
+      else
+        sub_terms = []
+        @terms.each { |term| sub_terms << term if term[:term].downcase.start_with?(q.downcase) }
+        self.response = sub_terms
+      end
+    end
+
+    def normalize_terms(terms)
+      normalized_terms = []
+      terms.each do |term|
+        if term.is_a? String
+          normalized_terms << { :id => term, :term => term }
+        else
+          term[:id] = term[:id] || term[:term]
+          normalized_terms << term
+        end
+      end
+      normalized_terms
+    end
+
+    def parse_authority_response
+      parsed_response = []
+      self.response.each do |res|
+        parsed_response << { :id => res[:id], :label => res[:term] }
+      end
+      self.response = parsed_response
+    end
+    
+    def get_full_record(id)
+      target_term = {}
+      @terms.each do |term|
+        if term[:id] == id
+          target_term = term
+        end
+      end
+      target_term.to_json
+    end
+    
+    def self.sub_authorities_path
+      config_path = AUTHORITIES_CONFIG[:local_path]
+      if config_path.starts_with?(File::Separator)
+        config_path
+      else
+        File.join(Rails.root, config_path)
+      end
+    end
+    
+    def self.sub_authorities
+      sub_auths = []
+      Dir.foreach(Qa::Authorities::Local.sub_authorities_path) { |file| sub_auths << File.basename(file, File.extname(file)) }
+      sub_auths
+    end
+        
+  end
+end