Feat elasticsearch japanese (#12194)

docker/.env.example

@@ -383,7 +383,7 @@ SUPABASE_URL=your-server-url
 # ------------------------------
 
 # The type of vector store to use.
-# Supported values are `weaviate`, `qdrant`, `milvus`, `myscale`, `relyt`, `pgvector`, `pgvecto-rs`, `chroma`, `opensearch`, `tidb_vector`, `oracle`, `tencent`, `elasticsearch`, `analyticdb`, `couchbase`, `vikingdb`, `oceanbase`.
+# Supported values are `weaviate`, `qdrant`, `milvus`, `myscale`, `relyt`, `pgvector`, `pgvecto-rs`, `chroma`, `opensearch`, `tidb_vector`, `oracle`, `tencent`, `elasticsearch`, `elasticsearch-ja`, `analyticdb`, `couchbase`, `vikingdb`, `oceanbase`.
 VECTOR_STORE=weaviate
 
 # The Weaviate endpoint URL. Only available when VECTOR_STORE is `weaviate`.
@@ -512,7 +512,7 @@ TENCENT_VECTOR_DB_SHARD=1
 TENCENT_VECTOR_DB_REPLICAS=2
 
 # ElasticSearch configuration, only available when VECTOR_STORE is `elasticsearch`
-ELASTICSEARCH_HOST=0.0.0.0
+ELASTICSEARCH_HOST=elasticsearch
 ELASTICSEARCH_PORT=9200
 ELASTICSEARCH_USERNAME=elastic
 ELASTICSEARCH_PASSWORD=elastic

docker/docker-compose.yaml

@@ -883,20 +883,28 @@ services:
     container_name: elasticsearch
     profiles:
       - elasticsearch
+      - elasticsearch-ja
     restart: always
     volumes:
+      - ./elasticsearch/docker-entrypoint.sh:/docker-entrypoint-mount.sh
       - dify_es01_data:/usr/share/elasticsearch/data
     environment:
       ELASTIC_PASSWORD: ${ELASTICSEARCH_PASSWORD:-elastic}
+      VECTOR_STORE: ${VECTOR_STORE:-}
       cluster.name: dify-es-cluster
       node.name: dify-es0
       discovery.type: single-node
-      xpack.license.self_generated.type: trial
+      xpack.license.self_generated.type: basic
       xpack.security.enabled: 'true'
       xpack.security.enrollment.enabled: 'false'
       xpack.security.http.ssl.enabled: 'false'
     ports:
       - ${ELASTICSEARCH_PORT:-9200}:9200
+    deploy:
+      resources:
+        limits:
+          memory: 2g
+    entrypoint: [ 'sh', '-c', "sh /docker-entrypoint-mount.sh" ]
     healthcheck:
       test: [ 'CMD', 'curl', '-s', 'http://localhost:9200/_cluster/health?pretty' ]
       interval: 30s

docker/elasticsearch/docker-entrypoint.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+
+set -e
+
+if [ "${VECTOR_STORE}" = "elasticsearch-ja" ]; then
+    # Check if the ICU tokenizer plugin is installed
+    if ! /usr/share/elasticsearch/bin/elasticsearch-plugin list | grep -q analysis-icu; then
+        printf '%s\n' "Installing the ICU tokenizer plugin"
+        if ! /usr/share/elasticsearch/bin/elasticsearch-plugin install analysis-icu; then
+            printf '%s\n' "Failed to install the ICU tokenizer plugin"
+            exit 1
+        fi
+    fi
+    # Check if the Japanese language analyzer plugin is installed
+    if ! /usr/share/elasticsearch/bin/elasticsearch-plugin list | grep -q analysis-kuromoji; then
+        printf '%s\n' "Installing the Japanese language analyzer plugin"
+        if ! /usr/share/elasticsearch/bin/elasticsearch-plugin install analysis-kuromoji; then
+            printf '%s\n' "Failed to install the Japanese language analyzer plugin"
+            exit 1
+        fi
+    fi
+fi
+
+# Run the original entrypoint script
+exec /bin/tini -- /usr/local/bin/docker-entrypoint.sh