Merge pull request #127 from MITLibraries/bulk-detections

Adds BulkChecker module for Detectors

Gemfile

@@ -86,6 +86,8 @@ group :production do
 end
 
 group :development, :test do
+   gem 'awesome_print'
+
   # See https://guides.rubyonrails.org/debugging_rails_applications.html#debugging-with-the-debug-gem
   gem 'debug', platforms: %i[mri windows]
 

Gemfile.lock

@@ -100,6 +100,7 @@ GEM
       rake (>= 10.4, < 14.0)
     ast (2.4.2)
     attr_required (1.0.2)
+    awesome_print (1.9.2)
     barnes (0.0.9)
       multi_json (~> 1)
       statsd-ruby (~> 1.1)
@@ -475,6 +476,7 @@ PLATFORMS
 DEPENDENCIES
   administrate (~> 0.20.1)
   annotate
+  awesome_print
   barnes
   bootsnap
   cancancan

app/models/detector/bulk_checker.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+class Detector
+  # BulkTermChecker is expected to be added to Detectors via `extend Detector::BulkTermChecker` to allow the
+  # singleton class to access it
+  # See also: `PatternChecker` for shared instance methods
+  module BulkChecker
+    # This method is intended to be used for inspecting detections during development.
+    # Assumptions include
+    #   - the Class including this module implements a `detections` method (either via `attr_reader` or as a method)
+    #     that is only populated for Terms in which it has made a detection
+    #   - the initialize method accepts a `phrase` as a string
+    # @param output [boolean] optional. Defaults to false as that is the more likely scenario useful in development as
+    #   the logger output is often what is desired.
+    def check_all_matches(output: false)
+      count = 0
+      matches = []
+      Term.find_each do |t|
+        d = new(t.phrase)
+        next if d.detections.blank?
+
+        count += 1
+
+        matches.push [t.phrase, d.detections]
+      end
+
+      if Rails.env.development?
+        Rails.logger.ap matches
+
+        Rails.logger.ap "Total Terms: #{Term.count}"
+        Rails.logger.ap "Total Matches: #{count}"
+      end
+
+      matches if output
+    end
+  end
+end

app/models/detector/citation.rb

@@ -12,6 +12,9 @@ class Detector
   class Citation
     attr_reader :score, :subpatterns, :summary
 
+    # shared singleton methods
+    extend Detector::BulkChecker
+
     # Citation patterns are regular expressions which attempt to identify structures that are part of many citations.
     # This object is used as part of the pattern_checker method. Some of these patterns may get promoted to the Detector
     # model if they prove useful beyond a Citation context.
@@ -29,6 +32,8 @@ class Citation
 
     # The required score value is the threshold needed for a phrase to be officially recorded with a Detection via it's
     # associated Term.
+    # Hint: set this to 0 in development environments if you want to temporarily see all output
+    # of `.check_all_matches` rather than just the matches that met this threshold.
     REQUIRED_SCORE = 6
 
     # Summary thresholds are used by the calculate_score method. This class counts the number of occurrences of specific
@@ -69,6 +74,12 @@ def initialize(phrase)
       @score = calculate_score
     end
 
+    def detections
+      return unless detection?
+
+      [@summary, @subpatterns, @score]
+    end
+
     # The record method first runs all of the parsers by running the initialize method. If the resulting score is higher
     # than the REQUIRED_SCORE value, then a Detection is registered.
     # @param term [Term]

app/models/detector/lcsh.rb

@@ -9,6 +9,9 @@ class Lcsh
     # shared instance methods
     include Detector::PatternChecker
 
+    # shared singleton methods
+    extend Detector::BulkChecker
+
     # For now the initialize method just needs to run the pattern checker. A space for future development would be to
     # write additional methods to look up the detected LCSH for more information, and to confirm that the phrase is
     # actually an LCSH.

app/models/detector/pattern_checker.rb

@@ -3,6 +3,7 @@
 class Detector
   # PatternChecker is intended to be added to Detectors via `include Detector::PatternChecker` to make
   # these methods available to instances of the class
+  # See also: `BulkTermChecker` for shared singleton methods
   module PatternChecker
     # pattern_checker iterates over all patterns defined in the calling object's `pattern` method.
     #

app/models/detector/standard_identifiers.rb

@@ -13,6 +13,9 @@ def self.table_name_prefix
     # shared instance methods
     include Detector::PatternChecker
 
+    # shared singleton methods
+    extend Detector::BulkChecker
+
     # Initialization process will run pattern checkers and strip invalid ISSN detections.
     #   @param phrase String. Often a `Term.phrase`.
     #   @return Nothing intentional. Data is written to Hash `@detections` during processing.

test/models/detector/bulk_checker_test.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+require 'test_helper'
+
+class Detector
+  class CitationTest < ActiveSupport::TestCase
+    test 'citation_bulk_checker' do
+      bulk = Detector::Citation.check_all_matches(output: true)
+
+      assert_equal(1, bulk.count)
+    end
+
+    test 'journal_bulk_checker' do
+      skip 'Detector::Journal does not yet support bulk_checker'
+    end
+
+    test 'lcsh_bulk_checker' do
+      bulk = Detector::Lcsh.check_all_matches(output: true)
+
+      assert_equal(1, bulk.count)
+    end
+
+    test 'standard_identifier_bulk_checker' do
+      bulk = Detector::StandardIdentifiers.check_all_matches(output: true)
+
+      assert_equal(5, bulk.count)
+    end
+
+    test 'suggested_resources_bulk_checker' do
+      skip 'Detector::SuggestedResources does not yet support bulk_checker'
+    end
+  end
+end

test/models/detector/citation_test.rb

@@ -245,5 +245,20 @@ class CitationTest < ActiveSupport::TestCase
         assert_equal detection_count + 1, Detection.count
       end
     end
+
+    test 'detections returns nil when score is lower than configured' do
+      result = Detector::Citation.new('nothing here')
+
+      assert_equal 0, result.score
+      assert_nil result.detections
+    end
+
+    test 'detections returns expected array when score is higher than configured' do
+      result = Detector::Citation.new(terms('citation').phrase)
+
+      assert_equal result.summary, result.detections[0]
+      assert_equal result.subpatterns, result.detections[1]
+      assert_equal result.score, result.detections[2]
+    end
   end
 end