Revert "Fix/elasticsearch plugin ci failure (#297)"

This reverts commit 866f93b.
kmycode · Nov 20, 2023 · d152676 · d152676
1 parent 49317a7
commit d152676
Show file tree

Hide file tree

Showing 5 changed files with 216 additions and 32 deletions.
diff --git a/.env.test b/.env.test
@@ -4,4 +4,7 @@ NODE_ENV=production
 LOCAL_DOMAIN=cb6e6126.ngrok.io
 LOCAL_HTTPS=true
 # Elasticsearch
+ES_ENABLED=false
+ES_HOST=localhost
+ES_PORT=9200
 ES_PREFIX=test
diff --git a/.github/workflows/test-ruby.yml b/.github/workflows/test-ruby.yml
@@ -259,6 +259,19 @@ jobs:
         ports:
           - 6379:6379
 
+      search:
+        image: ${{ matrix.search-image }}
+        env:
+          discovery.type: single-node
+          xpack.security.enabled: false
+        options: >-
+          --health-cmd "curl http://localhost:9200/_cluster/health"
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 10
+        ports:
+          - 9200:9200
+
     env:
       DB_HOST: localhost
       DB_USER: postgres
@@ -300,32 +313,6 @@ jobs:
       - name: Set up Javascript environment
         uses: ./.github/actions/setup-javascript
 
-      - name: Configure sysctl limits
-        run: |
-          sudo swapoff -a
-          sudo sysctl -w vm.swappiness=1
-          sudo sysctl -w fs.file-max=262144
-          sudo sysctl -w vm.max_map_count=262144
-
-      - name: Install Elasticsearch
-        run: |
-          wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.17.10-amd64.deb
-          wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.17.10-amd64.deb.sha512
-          shasum -a 512 -c elasticsearch-7.17.10-amd64.deb.sha512
-          sudo dpkg -i elasticsearch-7.17.10-amd64.deb
-          sudo /usr/share/elasticsearch/bin/elasticsearch-plugin install https://github.com/WorksApplications/elasticsearch-sudachi/releases/download/v3.1.0/elasticsearch-7.17.10-analysis-sudachi-3.1.0.zip
-
-      - name: Install dictionary
-        run: |
-          wget http://sudachi.s3-website-ap-northeast-1.amazonaws.com/sudachidict/sudachi-dictionary-latest-core.zip
-          unzip sudachi-dictionary-latest-core.zip
-          sudo mkdir /etc/elasticsearch/sudachi -p
-          sudo cp sudachi-dictionary-*/system_core.dic /etc/elasticsearch/sudachi
-
-      - name: Running Elasticsearch
-        run: |
-          sudo systemctl start elasticsearch
-
       - name: Load database schema
         run: './bin/rails db:create db:schema:load db:seed'
 

diff --git a/app/chewy/accounts_index.rb b/app/chewy/accounts_index.rb
@@ -1,7 +1,72 @@
 # frozen_string_literal: true
 
 class AccountsIndex < Chewy::Index
-  settings index: index_preset(refresh_interval: '30s'), analysis: {
+  DEVELOPMENT_SETTINGS = {
+    filter: {
+      english_stop: {
+        type: 'stop',
+        stopwords: '_english_',
+      },
+
+      english_stemmer: {
+        type: 'stemmer',
+        language: 'english',
+      },
+
+      english_possessive_stemmer: {
+        type: 'stemmer',
+        language: 'possessive_english',
+      },
+    },
+
+    analyzer: {
+      natural: {
+        tokenizer: 'standard',
+        filter: %w(
+          lowercase
+          asciifolding
+          cjk_width
+          elision
+          english_possessive_stemmer
+          english_stop
+          english_stemmer
+        ),
+      },
+
+      sudachi_analyzer: {
+        tokenizer: 'standard',
+        filter: %w(
+          lowercase
+          asciifolding
+          cjk_width
+          elision
+          english_possessive_stemmer
+          english_stop
+          english_stemmer
+        ),
+      },
+
+      verbatim: {
+        tokenizer: 'standard',
+        filter: %w(lowercase asciifolding cjk_width),
+      },
+
+      edge_ngram: {
+        tokenizer: 'edge_ngram',
+        filter: %w(lowercase asciifolding cjk_width),
+      },
+    },
+
+    tokenizer: {
+      edge_ngram: {
+        type: 'edge_ngram',
+        min_gram: 1,
+        max_gram: 15,
+      },
+    },
+  }.freeze
+
+  PRODUCTION_SETTINGS = {
     filter: {
       english_stop: {
         type: 'stop',
@@ -77,7 +142,9 @@ class AccountsIndex < Chewy::Index
         discard_punctuation: 'true',
       },
     },
-  }
+  }.freeze
+
+  settings index: index_preset(refresh_interval: '30s'), analysis: Rails.env.test? ? DEVELOPMENT_SETTINGS : PRODUCTION_SETTINGS
 
   index_scope ::Account.searchable.includes(:account_stat)
 

diff --git a/app/chewy/public_statuses_index.rb b/app/chewy/public_statuses_index.rb
@@ -1,7 +1,69 @@
 # frozen_string_literal: true
 
 class PublicStatusesIndex < Chewy::Index
-  settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: {
+  DEVELOPMENT_SETTINGS = {
+    filter: {
+      english_stop: {
+        type: 'stop',
+        stopwords: '_english_',
+      },
+
+      english_stemmer: {
+        type: 'stemmer',
+        language: 'english',
+      },
+
+      english_possessive_stemmer: {
+        type: 'stemmer',
+        language: 'possessive_english',
+      },
+    },
+
+    analyzer: {
+      verbatim: {
+        tokenizer: 'uax_url_email',
+        filter: %w(lowercase),
+      },
+
+      content: {
+        tokenizer: 'standard',
+        filter: %w(
+          lowercase
+          asciifolding
+          cjk_width
+          elision
+          english_possessive_stemmer
+          english_stop
+          english_stemmer
+        ),
+      },
+
+      sudachi_analyzer: {
+        tokenizer: 'standard',
+        filter: %w(
+          lowercase
+          asciifolding
+          cjk_width
+          elision
+          english_possessive_stemmer
+          english_stop
+          english_stemmer
+        ),
+      },
+
+      hashtag: {
+        tokenizer: 'keyword',
+        filter: %w(
+          word_delimiter_graph
+          lowercase
+          asciifolding
+          cjk_width
+        ),
+      },
+    },
+  }.freeze
+
+  PRODUCTION_SETTINGS = {
     filter: {
       english_stop: {
         type: 'stop',
@@ -75,7 +137,9 @@ class PublicStatusesIndex < Chewy::Index
         discard_punctuation: 'true',
       },
     },
-  }
+  }.freeze
+
+  settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: Rails.env.test? ? DEVELOPMENT_SETTINGS : PRODUCTION_SETTINGS
 
   index_scope ::Status.unscoped
                       .kept

diff --git a/app/chewy/statuses_index.rb b/app/chewy/statuses_index.rb
@@ -1,7 +1,68 @@
 # frozen_string_literal: true
 
 class StatusesIndex < Chewy::Index
-  settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: {
+  DEVELOPMENT_SETTINGS = {
+    filter: {
+      english_stop: {
+        type: 'stop',
+        stopwords: '_english_',
+      },
+
+      english_stemmer: {
+        type: 'stemmer',
+        language: 'english',
+      },
+
+      english_possessive_stemmer: {
+        type: 'stemmer',
+        language: 'possessive_english',
+      },
+    },
+    analyzer: {
+      verbatim: {
+        tokenizer: 'uax_url_email',
+        filter: %w(lowercase),
+      },
+
+      content: {
+        tokenizer: 'standard',
+        filter: %w(
+          lowercase
+          asciifolding
+          cjk_width
+          elision
+          english_possessive_stemmer
+          english_stop
+          english_stemmer
+        ),
+      },
+
+      sudachi_analyzer: {
+        tokenizer: 'standard',
+        filter: %w(
+          lowercase
+          asciifolding
+          cjk_width
+          elision
+          english_possessive_stemmer
+          english_stop
+          english_stemmer
+        ),
+      },
+
+      hashtag: {
+        tokenizer: 'keyword',
+        filter: %w(
+          word_delimiter_graph
+          lowercase
+          asciifolding
+          cjk_width
+        ),
+      },
+    },
+  }.freeze
+
+  PRODUCTION_SETTINGS = {
     filter: {
       english_stop: {
         type: 'stop',
@@ -79,7 +140,9 @@ class StatusesIndex < Chewy::Index
         discard_punctuation: 'true',
       },
     },
-  }
+  }.freeze
+
+  settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: Rails.env.test? ? DEVELOPMENT_SETTINGS : PRODUCTION_SETTINGS
 
   index_scope ::Status.unscoped.kept.without_reblogs.includes(
     :account,