solrizer 1.2.2 → 2.0.0.rc1

data/.gitignore

@@ -1,5 +1,6 @@
 .DS_Store
 nohup.out
+Gemfile.lock
 
 *.sqlite3
 *.log
@@ -20,4 +21,4 @@ rerun.txt
 
 /.bundle
 /.yardoc
-/doc
+/doc

data/History.txt

@@ -1,3 +1,6 @@
+h2. 2.0.0
+HYDRA-827 DO NOT index terms by default
+
 h2. 1.2.2
 BUG: RSolr::Client doesn't have a delete method.  Changed to delete_by_id. (only affects solrizer shell script)  Thanks to mkorcy.
 

data/lib/solrizer/version.rb

@@ -1,3 +1,3 @@
 module Solrizer
-  VERSION = "1.2.2"
+  VERSION = "2.0.0.rc1"
 end

data/lib/solrizer/xml/terminology_based_solrizer.rb

@@ -48,14 +48,17 @@ module Solrizer::XML::TerminologyBasedSolrizer
   
   # Populate a solr document with solr fields corresponding to the given xml node
   # Field names are generated using settings from the term in the +doc+'s terminology corresponding to +term_pointer+
+  # If the supplied term does not have an index_as attribute, no indexing will be performed.
   # @param [Nokogiri::XML::Node] node to solrize
   # @param [OM::XML::Document] doc document the node came from
   # @param [Array] term_pointer Array pointing to the term that should be used for solrization settings
+  # @param [Term] term the term to be solrized
   # @param [Hash] (optional) solr_doc (values hash) to populate
+  # @return [Hash] the solr doc
   def self.solrize_node(node, doc, term_pointer, term, solr_doc = Hash.new, field_mapper = nil, opts = {})
+    return solr_doc unless term.index_as
     field_mapper ||= self.default_field_mapper
     terminology = doc.class.terminology
-    # term = terminology.retrieve_term(*term_pointer)
     
     if term.path.kind_of?(Hash) && term.path.has_key?(:attribute)
       node_value = node.value

data/spec/fixtures/mods_article.rb

@@ -0,0 +1,88 @@
+module Samples
+  class ModsArticle
+    
+    include OM::XML::Document
+    
+    set_terminology do |t|
+      t.root(:path=>"mods", :xmlns=>"http://www.loc.gov/mods/v3", :schema=>"http://www.loc.gov/standards/mods/v3/mods-3-2.xsd", "xmlns:foo"=>"http://my.custom.namespace")
+
+
+      t.title_info(:path=>"titleInfo") {
+        t.main_title(:index_as=>[:facetable],:path=>"title", :label=>"title") {
+          t.main_title_lang(:path=>{:attribute=> "xml:lang"})
+        }
+        t.french_title(:ref=>[:title_info,:main_title], :attributes=>{"xml:lang"=>"fre"})
+        
+        t.language(:index_as=>[:facetable],:path=>{:attribute=>"lang"})
+      } 
+      t.language{
+        t.lang_code(:index_as=>[:facetable], :path=>"languageTerm", :attributes=>{:type=>"code"})
+      }
+      t.abstract(:index_as=>[])
+      t.subject {
+        t.topic(:index_as=>[:facetable])
+      }      
+      t.topic_tag(:proxy=>[:subject, :topic])    
+      # t.topic_tag(:index_as=>[:facetable],:path=>"subject", :default_content_path=>"topic")
+      # This is a mods:name.  The underscore is purely to avoid namespace conflicts.
+      t.name_ {
+        # this is a namepart
+        t.namePart(:type=>:string, :label=>"generic name")
+        # affiliations are great
+        t.affiliation
+        t.institution(:path=>"affiliation", :index_as=>[:facetable], :label=>"organization")
+        t.displayForm
+        t.role(:ref=>[:role])
+        t.description(:index_as=>[:facetable])
+        t.date(:path=>"namePart", :attributes=>{:type=>"date"})
+        t.last_name(:path=>"namePart", :attributes=>{:type=>"family"}, :index_as=>[])
+        t.first_name(:path=>"namePart", :attributes=>{:type=>"given"}, :label=>"first name")
+        t.terms_of_address(:path=>"namePart", :attributes=>{:type=>"termsOfAddress"})
+        t.computing_id
+        t.name_content(:path=>"text()")
+      }
+      # lookup :person, :first_name        
+      t.person(:ref=>:name, :attributes=>{:type=>"personal"}, :index_as=>[:facetable])
+      t.department(:proxy=>[:person,:description],:index_as=>[:facetable])
+      t.organization(:ref=>:name, :attributes=>{:type=>"corporate"}, :index_as=>[:facetable])
+      t.conference(:ref=>:name, :attributes=>{:type=>"conference"}, :index_as=>[:facetable])
+      t.role {
+        t.text(:path=>"roleTerm",:attributes=>{:type=>"text"}, :index_as=>[])
+        t.code(:path=>"roleTerm",:attributes=>{:type=>"code"})
+      }
+      t.journal(:path=>'relatedItem', :attributes=>{:type=>"host"}) {
+        t.title_info(:index_as=>[:facetable],:ref=>[:title_info])
+        t.origin_info(:path=>"originInfo") {
+          t.publisher
+          t.date_issued(:path=>"dateIssued")
+          t.issuance(:index_as=>[:facetable])
+        }
+        t.issn(:path=>"identifier", :attributes=>{:type=>"issn"})
+        t.issue(:path=>"part") {
+          t.volume(:path=>"detail", :attributes=>{:type=>"volume"}, :default_content_path=>"number")
+          t.level(:path=>"detail", :attributes=>{:type=>"number"}, :default_content_path=>"number")
+          t.extent
+          t.pages(:path=>"extent", :attributes=>{:unit=>"pages"}) {
+            t.start
+            t.end
+          }
+          t.start_page(:proxy=>[:pages, :start])
+          t.end_page(:proxy=>[:pages, :end])
+          t.publication_date(:path=>"date", :index_as=>[])
+        }
+      }
+      t.note
+      t.location(:path=>"location") {
+        t.url(:path=>"url")
+      }
+      t.publication_url(:proxy=>[:location,:url])
+      t.title(:proxy=>[:title_info, :main_title])
+      t.journal_title(:proxy=>[:journal, :title_info, :main_title])
+    end
+    
+    # Changes from OM::Properties implementation
+    # renamed family_name => last_name
+    # start_page & end_page now accessible as [:journal, :issue, :pages, :start] (etc.)
+
+  end
+end

data/spec/units/xml_terminology_based_solrizer_spec.rb

@@ -1,15 +1,15 @@
 require 'spec_helper'
-require 'solrizer/xml'
+require 'fixtures/mods_article'
 
 describe Solrizer::XML::TerminologyBasedSolrizer do
   
   before(:all) do
-    OM::Samples::ModsArticle.send(:include, Solrizer::XML::TerminologyBasedSolrizer)
+    Samples::ModsArticle.send(:include, Solrizer::XML::TerminologyBasedSolrizer)
   end
   
   before(:each) do
     article_xml = fixture( File.join("mods_articles", "hydrangea_article1.xml") )
-    @mods_article = OM::Samples::ModsArticle.from_xml(article_xml)
+    @mods_article = Samples::ModsArticle.from_xml(article_xml)
   end
   
   describe ".to_solr" do
@@ -34,7 +34,7 @@ describe Solrizer::XML::TerminologyBasedSolrizer do
       # ActiveFedora::NokogiriDatastream.stubs(:accessors).returns(mock_accessors)
       solr_doc = Hash.new
       @mods_article.field_mapper = Solrizer::FieldMapper::Default.new
-      OM::Samples::ModsArticle.terminology.terms.each_pair do |k,v|
+      Samples::ModsArticle.terminology.terms.each_pair do |k,v|
         @mods_article.expects(:solrize_term).with(v, solr_doc, @mods_article.field_mapper)
       end
       @mods_article.to_solr(solr_doc)
@@ -49,21 +49,16 @@ describe Solrizer::XML::TerminologyBasedSolrizer do
       solr_doc["abstract_t"].should == ["ABSTRACT"]
       solr_doc["title_info_1_language_t"].should == ["finnish"]
       solr_doc["person_1_role_0_text_t"].should == ["teacher"]
+      # No index_as on the code field.
+      solr_doc["person_1_role_0_code_t"].should be_nil 
       solr_doc["person_last_name_t"].sort.should == ["FAMILY NAME", "Gautama"]
-      # This next line will fail until om > 1.0.2 is released
-      # solr_doc["topic_tag_t"].sort.should == ["CONTROLLED TERM", "TOPIC1", "TOPIC2"]
+      solr_doc["topic_tag_t"].sort.should == ["CONTROLLED TERM", "TOPIC 1", "TOPIC 2"]
       
       # These are a holdover from an old verison of OM
-      # solr_doc[:finnish_title_info_language_t].should == "finnish"
-      # solr_doc[:finnish_title_info_main_title_t].should == "Artikkelin otsikko Hydrangea artiklan 1"
+      puts "DOC: #{solr_doc.length}"
+      solr_doc['journal_0_issue_0_publication_date_t'].should == ["FEB. 2007"]
 
-      # solr_doc[:mydate_date].should == "fake-date"
-      # 
-      # solr_doc[:publisher_t].should be_nil
-      # solr_doc[:coverage_t].should be_nil
-      # solr_doc[:creation_date_dt].should be_nil
-      # solr_doc.should == ""
-    
+      
     end
     
   end
@@ -72,14 +67,14 @@ describe Solrizer::XML::TerminologyBasedSolrizer do
   
     it "should add fields to a solr document for all nodes corresponding to the given term and its children" do
       solr_doc = Hash.new
-      result = @mods_article.solrize_term(OM::Samples::ModsArticle.terminology.retrieve_term(:title_info), solr_doc)
+      result = @mods_article.solrize_term(Samples::ModsArticle.terminology.retrieve_term(:title_info), solr_doc)
       result.should == solr_doc
-      # @mods_article.solrize_term(:title_info, OM::Samples::ModsArticle.terminology.retrieve_term(:title_info), :solr_doc=>solr_doc).should == ""
+      # @mods_article.solrize_term(:title_info, Samples::ModsArticle.terminology.retrieve_term(:title_info), :solr_doc=>solr_doc).should == ""
     end
 
     it "should add multiple fields based on index_as" do
       fake_solr_doc = {}
-      term = OM::Samples::ModsArticle.terminology.retrieve_term(:name)
+      term = Samples::ModsArticle.terminology.retrieve_term(:name)
       term.children[:namePart].index_as = [:displayable, :facetable]
 
       @mods_article.solrize_term(term, fake_solr_doc)

metadata

@@ -1,13 +1,15 @@
 --- !ruby/object:Gem::Specification 
 name: solrizer
 version: !ruby/object:Gem::Version 
-  hash: 27
-  prerelease: 
+  hash: -2879808512
+  prerelease: 6
   segments: 
-  - 1
-  - 2
   - 2
-  version: 1.2.2
+  - 0
+  - 0
+  - rc
+  - 1
+  version: 2.0.0.rc1
 platform: ruby
 authors: 
 - Matt Zumwalt
@@ -15,7 +17,8 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2012-07-23 00:00:00 Z
+date: 2012-10-15 00:00:00 -05:00
+default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: nokogiri
@@ -188,9 +191,7 @@ extra_rdoc_files:
 - README.textile
 files: 
 - .gitignore
-- .rvmrc
 - Gemfile
-- Gemfile.lock
 - History.txt
 - LICENSE
 - README.textile
@@ -222,6 +223,7 @@ files:
 - spec/fixtures/druid-cm234kq4672-extProperties.xml
 - spec/fixtures/druid-cm234kq4672-stories.xml
 - spec/fixtures/druid-hc513kw4806-descMetadata.xml
+- spec/fixtures/mods_article.rb
 - spec/fixtures/mods_articles/hydrangea_article1.xml
 - spec/fixtures/test_solr_mappings.yml
 - spec/spec_helper.rb
@@ -230,6 +232,7 @@ files:
 - spec/units/field_name_mapper_spec.rb
 - spec/units/xml_extractor_spec.rb
 - spec/units/xml_terminology_based_solrizer_spec.rb
+has_rdoc: true
 homepage: http://github.com/projecthydra/solrizer
 licenses: []
 
@@ -250,16 +253,18 @@ required_ruby_version: !ruby/object:Gem::Requirement
 required_rubygems_version: !ruby/object:Gem::Requirement 
   none: false
   requirements: 
-  - - ">="
+  - - ">"
     - !ruby/object:Gem::Version 
-      hash: 3
+      hash: 25
       segments: 
-      - 0
-      version: "0"
+      - 1
+      - 3
+      - 1
+      version: 1.3.1
 requirements: []
 
 rubyforge_project: 
-rubygems_version: 1.8.17
+rubygems_version: 1.6.2
 signing_key: 
 specification_version: 3
 summary: A utility for building solr indexes, usually from Fedora repository content with solrizer-fedora extension gem.
@@ -269,6 +274,7 @@ test_files:
 - spec/fixtures/druid-cm234kq4672-extProperties.xml
 - spec/fixtures/druid-cm234kq4672-stories.xml
 - spec/fixtures/druid-hc513kw4806-descMetadata.xml
+- spec/fixtures/mods_article.rb
 - spec/fixtures/mods_articles/hydrangea_article1.xml
 - spec/fixtures/test_solr_mappings.yml
 - spec/spec_helper.rb

data/.rvmrc

@@ -1,35 +0,0 @@
-#!/usr/bin/env bash
-
-# This is an RVM Project .rvmrc file, used to automatically load the ruby
-# development environment upon cd'ing into the directory
-
-ruby_string="ree-1.8.7"
-gemset_name="solrizer"
-
-#
-rvm_install_on_use_flag=1
-
-# Specify our desired <ruby>[@<gemset>], the @gemset name is optional.
-environment_id="${ruby_string}@${gemset_name}"
-
-# First, attempt to load the desired environment directly from the environment
-# file. This is very fast and efficient compared to running through the entire
-# CLI and selector. If you want feedback on which environment was used then
-# insert the word 'use' after --create as this triggers verbose mode.
-#
-if [[ -d "${rvm_path:-$HOME/.rvm}/environments" \
-  && -s "${rvm_path:-$HOME/.rvm}/environments/$environment_id" ]] ; then
-  \. "${rvm_path:-$HOME/.rvm}/environments/$environment_id"
-else
-  # If the environment file has not yet been created, use the RVM CLI to select.
-  rvm --create "$environment_id"
-fi
-
-#(
-  # Ensure that Bundler is installed, install it if it is not.
-  if ! command -v bundle ; then
-    printf "The rubygem 'bundler' is not installed, installing it now.\n"
-    gem install bundler
-  fi
-#)&
-

data/Gemfile.lock

@@ -1,72 +0,0 @@
-PATH
-  remote: .
-  specs:
-    solrizer (1.2.0)
-      daemons
-      mediashelf-loggable (~> 0.4.7)
-      nokogiri
-      om (>= 1.5.0)
-      stomp
-      xml-simple
-
-GEM
-  remote: http://rubygems.org/
-  specs:
-    RedCloth (4.2.8)
-    archive-tar-minitar (0.5.2)
-    columnize (0.3.4)
-    daemons (1.1.8)
-    diff-lcs (1.1.3)
-    linecache (0.46)
-      rbx-require-relative (> 0.0.4)
-    linecache19 (0.5.12)
-      ruby_core_source (>= 0.1.4)
-    mediashelf-loggable (0.4.9)
-    metaclass (0.0.1)
-    mocha (0.10.0)
-      metaclass (~> 0.0.1)
-    nokogiri (1.5.2)
-    om (1.6.0)
-      mediashelf-loggable
-      nokogiri (>= 1.4.2)
-    rbx-require-relative (0.0.5)
-    rcov (0.9.10)
-    rspec (2.7.0)
-      rspec-core (~> 2.7.0)
-      rspec-expectations (~> 2.7.0)
-      rspec-mocks (~> 2.7.0)
-    rspec-core (2.7.1)
-    rspec-expectations (2.7.0)
-      diff-lcs (~> 1.1.2)
-    rspec-mocks (2.7.0)
-    ruby-debug (0.10.4)
-      columnize (>= 0.1)
-      ruby-debug-base (~> 0.10.4.0)
-    ruby-debug-base (0.10.4)
-      linecache (>= 0.3)
-    ruby-debug-base19 (0.11.25)
-      columnize (>= 0.3.1)
-      linecache19 (>= 0.5.11)
-      ruby_core_source (>= 0.1.4)
-    ruby-debug19 (0.11.6)
-      columnize (>= 0.3.1)
-      linecache19 (>= 0.5.11)
-      ruby-debug-base19 (>= 0.11.19)
-    ruby_core_source (0.1.5)
-      archive-tar-minitar (>= 0.5.2)
-    stomp (1.2.2)
-    xml-simple (1.1.1)
-    yard (0.7.2)
-
-PLATFORMS
-  ruby
-
-DEPENDENCIES
-  RedCloth
-  mocha
-  rcov
-  rspec (~> 2.0)
-  ruby-debug
-  ruby-debug19
-  solrizer!
-  yard