Add analyzer_params config for milvus vectordb (#18180)

api/.env.example

 MILVUS_TOKEN=
 MILVUS_USER=root
 MILVUS_PASSWORD=Milvus
+MILVUS_ANALYZER_PARAMS=
 
 # MyScale configuration
 MYSCALE_HOST=127.0.0.1

api/configs/middleware/vdb/milvus_config.py

         "older versions",
         default=True,
     )
+
+    MILVUS_ANALYZER_PARAMS: Optional[str] = Field(
+        description='Milvus text analyzer parameters, e.g., {"type": "chinese"} for Chinese segmentation support.',
+        default=None,
+    )

api/core/rag/datasource/vdb/milvus/milvus_vector.py

     batch_size: int = 100  # Batch size for operations
     database: str = "default"  # Database name
     enable_hybrid_search: bool = False  # Flag to enable hybrid search
+    analyzer_params: Optional[str] = None  # Analyzer params
 
     @model_validator(mode="before")
     @classmethod
             "user": self.user,
             "password": self.password,
             "db_name": self.database,
+            "analyzer_params": self.analyzer_params,
         }
 
 
 
                 # Create the text field, enable_analyzer will be set True to support milvus automatically
                 # transfer text to sparse_vector, reference: https://milvus.io/docs/full-text-search.md
-                fields.append(
-                    FieldSchema(
-                        Field.CONTENT_KEY.value,
-                        DataType.VARCHAR,
-                        max_length=65_535,
-                        enable_analyzer=self._hybrid_search_enabled,
-                    )
-                )
+                content_field_kwargs: dict[str, Any] = {
+                    "max_length": 65_535,
+                    "enable_analyzer": self._hybrid_search_enabled,
+                }
+                if (
+                    self._hybrid_search_enabled
+                    and self._client_config.analyzer_params is not None
+                    and self._client_config.analyzer_params.strip()
+                ):
+                    content_field_kwargs["analyzer_params"] = self._client_config.analyzer_params
+
+                fields.append(FieldSchema(Field.CONTENT_KEY.value, DataType.VARCHAR, **content_field_kwargs))
+
                 # Create the primary key field
                 fields.append(FieldSchema(Field.PRIMARY_KEY.value, DataType.INT64, is_primary=True, auto_id=True))
                 # Create the vector field, supports binary or float vectors
                 password=dify_config.MILVUS_PASSWORD or "",
                 database=dify_config.MILVUS_DATABASE or "",
                 enable_hybrid_search=dify_config.MILVUS_ENABLE_HYBRID_SEARCH or False,
+                analyzer_params=dify_config.MILVUS_ANALYZER_PARAMS or "",
             ),
         )

docker/.env.example

 MILVUS_USER=
 MILVUS_PASSWORD=
 MILVUS_ENABLE_HYBRID_SEARCH=False
+MILVUS_ANALYZER_PARAMS=
 
 # MyScale configuration, only available when VECTOR_STORE is `myscale`
 # For multi-language support, please set MYSCALE_FTS_PARAMS with referring to:

docker/docker-compose.yaml

   MILVUS_USER: ${MILVUS_USER:-}
   MILVUS_PASSWORD: ${MILVUS_PASSWORD:-}
   MILVUS_ENABLE_HYBRID_SEARCH: ${MILVUS_ENABLE_HYBRID_SEARCH:-False}
+  MILVUS_ANALYZER_PARAMS: ${MILVUS_ANALYZER_PARAMS:-}
   MYSCALE_HOST: ${MYSCALE_HOST:-myscale}
   MYSCALE_PORT: ${MYSCALE_PORT:-8123}
   MYSCALE_USER: ${MYSCALE_USER:-default}