cul_scv_hydra 0.3.0

data/lib/cul_scv_hydra/controllers/terms.rb

@@ -0,0 +1,152 @@
+require 'cul_scv_hydra'
+require 'json'
+module Cul::Scv::Hydra::Controllers
+module Terms
+  extend ActiveSupport::Concern
+  included do
+    include Blacklight::SolrHelper
+    include MediaShelf::ActiveFedoraHelper
+    before_filter :require_solr, :only=>[:index]
+  end
+
+  def index
+    if params[:layout] == "false"
+      layout = false
+    end
+    if params[:term_id]
+      @solr_name = solr_name(params[:term_id])
+    else
+      @solr_name = params[:solr_name]
+    end
+    @terms = term_values
+    render :action=>params[:action], :layout=>layout
+  end
+
+  def solr_name(term_id)
+    if term_id.nil?
+      raise "Cannot provide suggestions without a term name"
+    end
+    field_key = field_key_from(term_id, Cul::Scv::Hydra::Om::ModsDocument.terminology)
+    get_solr_name(field_key, Cul::Scv::Hydra::Om::ModsDocument)
+  end
+  def term_values
+    if @solr_name.nil?
+      logger.debug "no solr field name, return nil for term values"
+      return nil
+    end
+    stub = params[:term_value]
+    solr_params = {}
+    solr_params['wt'] = 'json'
+    solr_params['terms'] = 'true'
+    solr_params['terms.fl'] = @solr_name
+    if stub
+      solr_params['terms.lower'] = stub
+      solr_params['terms.lower.incl'] = false
+      solr_params['terms.prefix'] = stub
+      solr_params['terms.sort'] = 'index'
+    else
+      solr_params['terms.sort'] = params.fetch(:sort,'count')
+      solr_params['terms.limit'] = -1
+    end
+    logger.debug solr_params.inspect
+    solr_response = Blacklight.solr.get 'terms', {:params => solr_params}
+    result = []
+    hash = {}
+    (1...solr_response['terms'].length).step(2) { |ix|
+      _t = solr_response['terms'][ix]
+      (0..._t.length).step(2) { |jx|
+        result << [_t[jx], _t[jx + 1]]
+      }
+    }
+    logger.debug result.inspect
+    return result
+  end
+
+## field_key_from(field_name) to reverse this method from HydraFedoraMetadataHelper
+#  def field_name_for(field_key)
+#   if field_key.kind_of?(Array)
+#     return OM::XML::Terminology.term_hierarchical_name(*field_key)
+#   else
+#     field_key.to_s
+#   end
+#  end
+
+  def field_key_from(field_name, terminology)
+    # this is a hack, but necessary until/unless OM generates names differently
+   candidates = field_name.split('__')
+   field_key = []
+   candidates.each_with_index { |candidate, ix|
+     if (candidates[ix + 1] and candidates[ix + 1].match(/\d+/))
+       field_key << {OM.destringify(candidate) => Integer(candidates.delete_at(ix + 1))}
+     else
+       field_key << OM.destringify(candidate)
+     end
+   }
+   logger.debug field_key.inspect
+   field_key
+  end
+
+  # ** largely copied from ActiveFedora::NokogiriDatastream.get_values **
+  def get_solr_name(term_pointer, dsClass)
+    term = dsClass.terminology.retrieve_term(*OM.pointers_to_flat_array(term_pointer, false))
+    names = []
+    if is_hierarchical_term_pointer?(*term_pointer)
+      bases = []
+      #add first item in term_pointer as start of bases
+      # then iterate through possible nodes that might exist
+      term_pointer.first.kind_of?(Hash) ? bases << term_pointer.first.keys.first : bases << term_pointer.first
+      for i in 1..(term_pointer.length-1)
+        #iterate in reverse so that we can modify the bases array while iterating
+        (bases.length-1).downto(0) do |j|
+          current_last = (term_pointer[i].kind_of?(Hash) ? term_pointer[i].keys.first : term_pointer[i])
+          if (term_pointer[i-1].kind_of?(Hash))
+            #just use index supplied instead of trying possibilities
+            index = term_pointer[i-1].values.first
+            solr_name_base = OM::XML::Terminology.term_hierarchical_name({bases[j]=>index},current_last)
+            solr_name = generate_solr_symbol(solr_name_base, term.data_type)
+            bases.delete_at(j)
+            #insert the new solr name base if found
+            bases.insert(j,solr_name_base) if has_solr_name?(solr_name,solr_doc)
+          else
+            #detect how many nodes exist
+            index = 0
+            current_base = bases[j]
+            bases.delete_at(j)
+            solr_name_base = OM::XML::Terminology.term_hierarchical_name({current_base=>index},current_last)
+            solr_name = generate_solr_symbol(solr_name_base, term.data_type)
+            bases.insert(j,solr_name_base)
+          end
+        end
+      end
+      bases.each do |base|
+        names << generate_solr_symbol(base.to_sym, term.data_type)
+      end
+    else
+#this is not hierarchical and we can simply look for the solr name created using the terms without any indexes
+      generic_field_name_base = OM::XML::Terminology.term_generic_name(*term_pointer)
+      names << generate_solr_symbol(generic_field_name_base, term.data_type)
+    end
+    names
+  end
+
+  # ** copied from ActiveFedora::NokogiriDatastream **
+  #@return true if the term_pointer contains an index
+  # ====Example:
+  #     [:image, {:title_set=>1}, :title] return true
+  #     [:image, :title_set, :title]      return false
+  def is_hierarchical_term_pointer?(*term_pointer)
+    if term_pointer.length>1
+      term_pointer.each do |pointer|
+        if pointer.kind_of?(Hash)
+          return true
+        end
+      end
+    end
+    return false
+  end
+  # ** copied from ActiveFedora::NokogiriDatastream **
+  def generate_solr_symbol(base, data_type)
+    Solrizer::XML::TerminologyBasedSolrizer.default_field_mapper.solr_name(base.to_sym, data_type)
+  end
+end
+end

data/lib/cul_scv_hydra/engine.rb

@@ -0,0 +1,9 @@
+# -*- encoding : utf-8 -*-
+# lib/cul_scv_hydra/engine.rb
+require "cul_scv_hydra"
+require "rails"
+
+module Cul::Scv::Hydra
+  class Engine < ::Rails::Engine
+  end
+end

data/lib/cul_scv_hydra/om.rb

@@ -0,0 +1,11 @@
+module Cul
+  module Scv
+    module Hydra
+      module Om
+      end
+    end
+  end
+end
+require "cul_scv_hydra/om/dc_metadata"
+require "cul_scv_hydra/om/scv_mods_document"
+require "cul_scv_hydra/om/standard_mods"

data/lib/cul_scv_hydra/om/dc_metadata.rb

@@ -0,0 +1,70 @@
+require 'active-fedora'
+module Cul
+module Scv
+module Hydra
+module Om
+  class DCMetadata < ::ActiveFedora::NokogiriDatastream
+    include OM::XML::Document
+    after_save :action_after_save
+    set_terminology do |t|
+      t.root(:path=>"dc", :namespace_prefix=>"oai_dc",
+             "xmlns:oai_dc"=>"http://www.openarchives.org/OAI/2.0/oai_dc/",
+             "xmlns:dc"=>"http://purl.org/dc/elements/1.1/",
+             :schema=>"http://www.openarchives.org/OAI/2.0/oai_dc.xsd")
+      t.contributor(:path=>"contributor", :namespace_prefix=>"dc")
+      t.coverage(:path=>"coverage", :namespace_prefix=>"dc")
+      t.creator(:path=>"creator", :namespace_prefix=>"dc")
+      t.date(:path=>"date", :namespace_prefix=>"dc")
+      t.description(:path=>"description", :namespace_prefix=>"dc")
+      t.format(:path=>"format", :namespace_prefix=>"dc")
+      t.identifier(:path=>"identifier", :namespace_prefix=>"dc")
+      t.language(:path=>"language", :namespace_prefix=>"dc")
+      t.publisher(:path=>"publisher", :namespace_prefix=>"dc")
+      t.relation(:path=>"relation", :namespace_prefix=>"dc")
+      t.rights(:path=>"rights", :namespace_prefix=>"dc")
+      t.source(:path=>"source", :namespace_prefix=>"dc")
+      t.subject(:path=>"subject", :namespace_prefix=>"dc")
+      t.title(:path=>"title", :namespace_prefix=>"dc")
+      t.dc_type(:path=>"type", :namespace_prefix=>"dc")
+    end
+  
+    def self.xml_template
+      builder = Nokogiri::XML::Builder.new do |xml|
+        xml.dc(
+           "xmlns:oai_dc"=>"http://www.openarchives.org/OAI/2.0/oai_dc/",
+           "xmlns:dc"=>"http://purl.org/dc/elements/1.1/",
+           "xmlns:xsi"=>"http://www.w3.org/2001/XMLSchema-instance"){
+          xml.parent.namespace_definitions.each {|ns|
+            xml.parent.namespace = ns if ns.prefix == 'oai_dc'
+          }
+        }
+      end
+      builder.doc.encoding = 'UTF-8'
+      builder.doc.root["xsi:schemaLocation"] = 'http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd'
+      return builder.doc
+    end
+    def action_after_save
+      self.dirty= false
+    end
+    def method_missing method, *args
+      query = false
+      _mname = method.id2name
+      if _mname[-1,1] == '?'
+        query = true
+        _mname = _mname[0,_mname.length-1]
+      end
+      _msym = _mname.to_sym
+      has_term = self.class.terminology.has_term?(_msym)
+      return false if query and not has_term
+      _r = super(_mname.to_sym, *args)
+      if query
+        _r.length > 0
+      else
+        _r
+      end
+    end
+  end
+end
+end
+end
+end

data/lib/cul_scv_hydra/om/scv_mods_document.rb

@@ -0,0 +1,132 @@
+require 'active-fedora'
+require 'solrizer'
+require 'cul_scv_hydra/solrizer'
+module Cul
+module Scv
+module Hydra
+module Om
+  class ModsDocument < ::ActiveFedora::NokogiriDatastream
+    include OM::XML::Document
+    include Cul::Scv::Hydra::Solrizer::TerminologyBasedSolrizer
+
+    after_save :action_after_save
+
+    set_terminology do |t|
+      t.root(:path=>"mods",
+             :xmlns=>"http://www.loc.gov/mods/v3",
+             :schema=>"http://www.loc.gov/standards/mods/v3/mods-3-4.xsd")
+      
+      t.main_title_info(:path=>'titleInfo', :index_as=>[:not_searchable], :attributes=>{:type=>:none}){
+        t.main_title(:path=>"title", :index_as=>[:displayable,:searchable])
+      }
+      
+      t.search_title_info(:path=>'titleInfo', :index_as=>[:not_searchable]){
+        t.search_title(:path=>'title', :index_as=>[:searchable])
+      }
+      t.project(:path=>"relatedItem", :attributes=>{:type=>"host", :displayLabel=>"Project"}, :index_as=>[:not_searchable]){
+        t.project_title_info(:path=>'titleInfo', :index_as=>[:not_searchable]){
+          t.lib_project(:path=>'title',:index_as=>[:displayable, :searchable])
+          t.lib_project_facet(:path=>'title', :index_as=>[:facetable, :not_searchable],:variant_of=>{:field_base=>'lib_project',:map=>:project_facet})
+        }
+      }
+      t.collection(:path=>"relatedItem", :attributes=>{:type=>"host", :displayLabel=>"Collection"}, :index_as=>[:not_searchable]){
+        t.collection_title_info(:path=>'titleInfo', :index_as=>[:not_searchable]){
+          t.lib_collection(:path=>'title', :index_as=>[:facetable, :displayable])
+        }
+      }
+      t.title(:proxy=>[:mods,:main_title_info,:main_title], :index_as=>[:displayable,:searchable, :sortable])
+      t.lib_project(:proxy=>[:project,:project_title_info, :lib_project])
+      t.lib_collection(:proxy=>[:collection,:collection_title_info, :lib_collection])
+# pattern matches
+      t.identifier(:path=>"identifier", :attributes=>{:type=>"local"}, :data_type=>:symbol)
+      t.identifier_text(:ref=>:identifier, :index_as=>[:not_searchable, :textable])
+      t.clio(:path=>"identifier", :attributes=>{:type=>"CLIO"}, :data_type=>:symbol)
+      t.abstract
+      t.subject {
+        t.topic
+      }
+      t.type_of_resource(:path=>"typeOfResource", :index_as=>[:not_searchable])
+      t.physical_description(:path=>"physicalDescription", :index_as=>[:not_searchable]){
+        t.form_marc(:path=>"form", :attributes=>{:authority=>"marcform"}, :index_as=>[:not_searchable])
+        t.form_aat(:path=>"form", :attributes=>{:authority=>"aat"}, :index_as=>[:not_searchable])
+        t.form(:attributes=>{:authority=>:none}, :index_as=>[:not_searchable])
+        t.form_nomarc(:path=>"form[@authority !='marcform']", :index_as=>[:not_searchable, :displayable, :facetable])
+        t.extent(:path=>"extent", :index_as=>[:not_searchable])
+        t.reformatting_quality(:path=>"reformattingQuality", :index_as=>[:not_searchable])
+        t.internet_media_type(:path=>"internetMediaType", :index_as=>[:not_searchable])
+        t.digital_origin(:path=>"digitalOrigin", :index_as=>[:not_searchable])
+      }
+      t.lib_format(:proxy=>[:physical_description, :form_nomarc])
+      t.location(:path=>"location", :index_as=>[:not_searchable]){
+        t.repo_text(:path=>"physicalLocation",:attributes=>{:authority=>:none},  :index_as=>[:not_searchable])
+        t.repo_code(:path=>"physicalLocation",:attributes=>{:authority=>"marcorg"}, :index_as=>[:not_searchable])
+        t.map_facet(:path=>"physicalLocation",:attributes=>{:authority=>"marcorg"}, :index_as=>[:facetable], :variant_of=>{:field_base=>'lib_repo',:map=>:marc_to_facet})
+        t.map_display(:path=>"physicalLocation",:attributes=>{:authority=>"marcorg"}, :index_as=>[:displayable, :not_searchable], :variant_of=>{:field_base=>'lib_repo',:map=>:marc_to_display})
+        t.shelf_locator(:path=>"shelfLocator", :index_as=>[:not_searchable, :textable])
+      }
+      t.name_personal(:path=>'name',:attributes=>{:type=>'personal'}, :index_as=>[:not_searchable]){
+        t.name_part(:path=>'namePart', :index_as=>[:facetable, :displayable, :searchable], :variant_of=>{:field_base=>:lib_name})
+      }
+      t.name_corporate(:path=>'name',:attributes=>{:type=>'corporate'}, :index_as=>[:not_searchable]){
+        t.name_part(:path=>'namePart', :index_as=>[:facetable, :displayable, :searchable], :variant_of=>{:field_base=>:lib_name})
+      }
+      t.note(:path=>"note", :index_as=>[:not_searchable, :textable])
+      t.access_condition(:path=>"accessCondition", :attributes=>{:type=>"useAndReproduction"}, :index_as => [:searchable], :data_type => :symbol)
+      t.record_info(:path=>"recordInfo", :index_as=>[:not_searchable]) {
+        t.record_creation_date(:path=>"recordCreationDate",:attributes=>{:encoding=>"w3cdtf"}, :index_as=>[:not_searchable])
+        t.record_content_source(:path=>"recordContentSource",:attributes=>{:authority=>"marcorg"}, :index_as=>[:not_searchable])
+        t.language_of_cataloging(:path=>"languageOfCataloging", :index_as=>[:not_searchable]){
+          t.language_term(:path=>"languageTerm", :index_as=>[:not_searchable], :attributes=>{:type=>:none})
+          t.language_code(:path=>"languageTerm",:attributes=>{:type=>'code',:authority=>"iso639-2b"}, :index_as=>[:not_searchable])
+        }
+        t.record_origin(:path=>"recordOrigin", :index_as=>[:not_searchable])
+      }
+      t.language_term(:proxy=>[:record_info, :language_of_cataloging, :language_term])
+      t.language_code(:proxy=>[:record_info, :language_of_cataloging, :language_code])
+
+      t.origin_info(:path=>"originInfo", :index_as=>[:not_searchable]){
+        t.date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf'}, :index_as=>[:not_searchable])
+        t.key_date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf',:keyDate=>'yes'}, :index_as=>[:not_searchable])
+        t.start_date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf',:keyDate=>'yes',:point=>'start'}, :index_as=>[:not_searchable])
+        t.end_date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf',:point=>'end'}, :index_as=>[:not_searchable])
+      }
+    end
+  
+    def self.xml_template
+      builder = Nokogiri::XML::Builder.new do |xml|
+        xml.mods(:version=>"3.4", 
+           "xmlns"=>"http://www.loc.gov/mods/v3",
+           "xmlns:xlink"=>"http://www.w3.org/1999/xlink",
+           "xmlns:xsi"=>"http://www.w3.org/2001/XMLSchema-instance"){
+        }
+      end
+      builder.doc.encoding = 'UTF-8'
+      # for some reason, this is the only way to get an equivalent nokogiri root node; the attribute can't be in the original builder call
+      builder.doc.root["xsi:schemaLocation"] = 'http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-4.xsd'
+      return builder.doc
+    end
+    def action_after_save
+      self.dirty= false
+    end
+    def method_missing method, *args
+      query = false
+      _mname = method.id2name
+      if _mname[-1,1] == '?'
+        query = true
+        _mname = _mname[0,_mname.length-1]
+      end
+      _msym = _mname.to_sym
+      has_term = self.class.terminology.has_term?(_msym)
+      return false if query and not has_term
+      _r = super(_mname.to_sym, *args)
+      if query
+        _r.length > 0
+      else
+        _r
+      end
+    end
+  end
+end
+end
+end
+end

data/lib/cul_scv_hydra/om/standard_mods.rb

@@ -0,0 +1,111 @@
+require 'active-fedora'
+require 'solrizer'
+require 'cul_scv_hydra/solrizer'
+module Cul
+module Scv
+module Hydra
+module Om
+  class StandardMods
+    include OM::XML::Document
+    include ::Solrizer::XML::TerminologyBasedSolrizer
+  
+    set_terminology do |t|
+      t.root(:path=>"mods",
+             :xmlns=>"http://www.loc.gov/mods/v3",
+             :schema=>"http://www.loc.gov/standards/mods/v3/mods-3-4.xsd")
+      t.identifier(:path=>"identifier", :attributes=>{:type=>"local"}, :data_type=>:symbol)
+      t.clio(:path=>"identifier", :attributes=>{:type=>"CLIO"}, :data_type=>:symbol)
+      t.title_info(:path=>"titleInfo", :index_as=>[:not_searchable]) {
+        t.main_title(:path=>"title", :index_as=>[:not_searchable])
+      }
+      t.title(:path=>'mods/oxns:titleInfo/oxns:title', :index_as=>[:searchable,:displayable, :sortable])
+      t.abstract
+      t.subject {
+        t.topic
+      }
+      t.type_of_resource(:path=>"typeOfResource", :index_as=>[:not_searchable])
+      t.physical_description(:path=>"physicalDescription", :index_as=>[:not_searchable]){
+        t.form_marc(:path=>"form", :attributes=>{:authority=>"marcform"}, :index_as=>[:not_searchable])
+        t.form_nomarc(:path=>"form[@authority !='marcform']", :index_as=>[:not_searchable, :displayable, :facetable])
+        t.extent(:path=>"extent", :index_as=>[:not_searchable])
+        t.reformatting_quality(:path=>"reformattingQuality", :index_as=>[:not_searchable])
+        t.internet_media_type(:path=>"internetMediaType", :index_as=>[:not_searchable])
+        t.digital_origin(:path=>"digitalOrigin", :index_as=>[:not_searchable])
+      }
+      t.lib_format(:proxy=>[:physical_description, :form_nomarc])
+      t.location(:path=>"location", :index_as=>[:not_searchable]){
+        t.repo_text(:path=>"physicalLocation",:attributes=>{:authority=>:none},  :index_as=>[:not_searchable])
+        t.repo_code(:path=>"physicalLocation",:attributes=>{:authority=>"marcorg"}, :index_as=>[:not_searchable])
+      }
+      t.lib_repo_text(:ref=>[:location, :repo_text], :label=>"lib_repo", :index_as=>[:searchable])
+      t.lib_repo(:ref=>[:location, :repo_code], :index_as=>[:not_searchable,:facetable, :displayable])
+      t.project_host(:path=>"relatedItem", :attributes=>{:type=>"host", :displayLabel=>"Project"}, :index_as=>[:not_searchable]){
+        t.p_title(:path=>'titleInfo',:index_as=>[:not_searchable])
+      }
+      t.lib_project(:proxy=>[:project_host, :p_title],:index_as=>[:facetable,:displayable, :not_searchable])
+      t.collection_host(:path=>"relatedItem", :attributes=>{:type=>"host", :displayLabel=>"Collection"}, :index_as=>[:not_searchable]){
+        t.c_title(:path=>'titleInfo',:index_as=>[:not_searchable])
+      }
+      t.lib_project(:path=>"mods/oxns:relatedItem[@type='host'][@displayLabel='Project']/oxns:titleInfo/oxns:title",:index_as=>[:facetable,:displayable, :not_searchable])
+      t.lib_collection(:path=>"mods/oxns:relatedItem[@type='host'][@displayLabel='Collection']/oxns:titleInfo/oxns:title",:index_as=>[:facetable,:displayable, :not_searchable])
+      t.note(:path=>"note")
+      t.access_condition(:path=>"accessCondition", :attributes=>{:type=>"useAndReproduction"}, :index_as => [:searchable], :data_type => :symbol)
+      t.record_info(:path=>"recordInfo", :index_as=>[:not_searchable]) {
+        t.record_creation_date(:path=>"recordCreationDate",:attributes=>{:encoding=>"w3cdtf"}, :index_as=>[:not_searchable])
+        t.record_content_source(:path=>"recordContentSource",:attributes=>{:authority=>"marcorg"}, :index_as=>[:not_searchable])
+        t.language_of_cataloging(:path=>"languageOfCataloging", :index_as=>[:not_searchable]){
+          t.language_term(:path=>"languageTerm", :index_as=>[:not_searchable], :attributes=>{:type=>:none})
+          t.language_code(:path=>"languageTerm",:attributes=>{:type=>'code',:authority=>"iso639-2b"}, :index_as=>[:not_searchable])
+        }
+        t.record_origin(:path=>"recordOrigin", :index_as=>[:not_searchable])
+      }
+
+      t.origin_info(:path=>"originInfo", :index_as=>[:not_searchable]){
+        t.date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf'}, :index_as=>[:not_searchable])
+        t.key_date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf',:keyDate=>'yes'}, :index_as=>[:not_searchable])
+        t.start_date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf',:keyDate=>'yes',:point=>'start'}, :index_as=>[:not_searchable])
+        t.end_date(:path=>"dateIssued", :attributes=>{:encoding=>'w3cdtf',:point=>'end'}, :index_as=>[:not_searchable])
+      }
+    end
+  
+    def self.xml_template
+      builder = Nokogiri::XML::Builder.new do |xml|
+        xml.mods(:version=>"3.4", 
+           "xmlns"=>"http://www.loc.gov/mods/v3",
+           "xmlns:xsi"=>"http://www.w3.org/2001/XMLSchema-instance"){
+        }
+      end
+      builder.doc.encoding = 'UTF-8'
+      builder.doc.root["xsi:schemaLocation"] = 'http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-4.xsd'
+      return builder.doc
+    end
+    def method_missing method, *args
+      query = false
+      _mname = method.id2name
+      if _mname[-1,1] == '?'
+        query = true
+        _mname = _mname[0,_mname.length-1]
+      end
+      _msym = _mname.to_sym
+      begin
+        has_term = self.class.terminology.has_term?(_msym)
+
+        _r = (has_term)? find_by_terms(_msym, *args) : nil
+        if query
+          return !( _r.nil? || _r.size()==0)
+        else
+          return _r
+        end
+      rescue
+        super
+      end
+    end
+    def update_values(params)
+      super
+      self.dirty = true
+    end
+  end
+end
+end
+end
+end