Merge branch 'elasticsearch-word-tokens-with-underscores' into 'master'

Include letters, numbers & underscore always as Elasticsearch token

See merge request gitlab-org/gitlab!36255

doc/development/elasticsearch.md

@@ -111,7 +111,8 @@ Patterns:
 - `'"((?:\\"|[^"]|\\")*)"'`: captures terms inside quotes, removing the quotes
 - `"'((?:\\'|[^']|\\')*)'"`: same as above, for single-quotes
 - `'\.([^.]+)(?=\.|\s|\Z)'`: separate terms with periods in-between
-- `'([\p{L}_.-]+)'` : some common chars in file names to keep the whole filename intact (eg. `my_file-ñame.txt`)
+- `'([\p{L}_.-]+)'`: some common chars in file names to keep the whole filename intact (eg. `my_file-ñame.txt`)
+- `'([\p{L}\d_]+)'`: letters, numbers and underscores are the most common tokens in programming. Always capture them greedily regardless of context.
 
 ## Gotchas
 

ee/changelogs/unreleased/elasticsearch-word-tokens-with-underscores.yml

+---
+title: Allow searching word tokens with letters, numbers and underscores in advanced global search
+merge_request: 36255
+author:
+type: changed

ee/lib/elastic/latest/config.rb

@@ -61,7 +61,8 @@ module Elastic
                   '"((?:\\"|[^"]|\\")*)"', # capture terms inside quotes, removing the quotes
                   "'((?:\\'|[^']|\\')*)'", # same as above, for single quotes
                   '\.([^.]+)(?=\.|\s|\Z)', # separate terms on periods
-                  '([\p{L}_.-]+)' # some common chars in file names to keep the whole filename intact (eg. my_file-name.txt)
+                  '([\p{L}_.-]+)', # some common chars in file names to keep the whole filename intact (eg. my_file-name.txt)
+                  '([\p{L}\d_]+)' # letters, numbers and underscores are the most common tokens in programming. Always capture them greedily regardless of context.
                 ]
               }
             },

ee/spec/lib/gitlab/elastic/search_results_spec.rb

@@ -637,6 +637,12 @@ RSpec.describe Gitlab::Elastic::SearchResults, :elastic, :sidekiq_might_not_need
           ParenthesesBetweenTokens)tokenAfterParentheses
           a.b.c=missing_token_around_equals
 
+          def self.ruby_method_name(ruby_method_arg)
+          RubyClassInvoking.ruby_method_call(with_arg)
+
+          def self.ruby_method_123(ruby_another_method_arg)
+          RubyClassInvoking.ruby_call_method_123(with_arg)
+
         FILE
       end
       let(:file_name) { 'elastic_specialchars_test.md' }
@@ -703,6 +709,22 @@ RSpec.describe Gitlab::Elastic::SearchResults, :elastic, :sidekiq_might_not_need
       it 'finds a token after = without a space' do
         expect(search_for('missing_token_around_equals')).to include(file_name)
       end
+
+      it 'finds a ruby method name even if preceeded with dot' do
+        expect(search_for('ruby_method_name')).to include(file_name)
+      end
+
+      it 'finds a ruby method name with numbers' do
+        expect(search_for('ruby_method_123')).to include(file_name)
+      end
+
+      it 'finds a ruby method call even if preceeded with dot' do
+        expect(search_for('ruby_method_call')).to include(file_name)
+      end
+
+      it 'finds a ruby method call with numbers' do
+        expect(search_for('ruby_call_method_123')).to include(file_name)
+      end
     end
   end