fix(frontend): resolve DocumentChunk merge conflicts with develop

Merge develop's effectiveIndexName fallback for chunk CRUD/search APIs
while keeping embedding model mismatch validation before hybrid search.

Author: Lifeng-Chen

.github/workflows/docker-deploy.yml

@@ -38,7 +38,10 @@ jobs:
       - name: Check if model is cached locally
         id: check-model
         run: |
-          if [ -f ~/model-assets/clip-vit-base-patch32/config.json ] && [ -d ~/model-assets/nltk_data ]; then
+          if [ -f ~/model-assets/clip-vit-base-patch32/config.json ] && \
+             [ -d ~/model-assets/nltk_data ] && \
+             [ -d ~/model-assets/table-transformer-structure-recognition ] && \
+             [ -d ~/model-assets/yolox ]; then
             echo "cache-hit=true" >> "$GITHUB_OUTPUT"
             cp -r ~/model-assets ./
           else
@@ -105,4 +108,4 @@ jobs:
             ./deploy.sh --mode 3 --is-mainland N --enable-terminal N --version 2 --root-dir "$HOME/nexent-production-data"
           else
             ./deploy.sh --mode 1 --is-mainland N --enable-terminal N --version 2 --root-dir "$HOME/nexent-development-data"
-          fi
+          fi

backend/agents/create_agent_info.py

@@ -1,4 +1,4 @@
-import threading
+import threading
 import logging
 from typing import List, Optional
 from urllib.parse import urljoin
@@ -498,6 +498,7 @@ async def create_tool_config_list(agent_id, tenant_id, user_id, version_no: int
             rerank = param_dict.get("rerank", False)
             rerank_model_name = param_dict.get("rerank_model_name", "")
             rerank_model = None
+            is_multimodal = bool(tool_config.params.pop("multimodal", False))
             if rerank and rerank_model_name:
                 rerank_model = get_rerank_model(
                     tenant_id=tenant_id, model_name=rerank_model_name
@@ -877,7 +878,7 @@ async def create_agent_run_info(
     # Filter MCP servers and tools, and build mcp_host with authorization
     used_mcp_urls = filter_mcp_servers_and_tools(agent_config, remote_mcp_dict)
 
-    # Build mcp_host list with authorization tokens
+    # Build mcp_host list with authorization tokens and custom headers
     mcp_host = []
     for url in used_mcp_urls:
         # Find the MCP record for this URL
@@ -892,10 +893,15 @@ async def create_agent_run_info(
                 "url": url,
                 "transport": "sse" if url.endswith("/sse") else "streamable-http"
             }
-            # Add authorization if present
+            headers = {}
             auth_token = mcp_record.get("authorization_token")
             if auth_token:
-                mcp_config["authorization"] = auth_token
+                headers["Authorization"] = auth_token
+            custom_headers = mcp_record.get("custom_headers")
+            if custom_headers and isinstance(custom_headers, dict):
+                headers.update(custom_headers)
+            if headers:
+                mcp_config["headers"] = headers
             mcp_host.append(mcp_config)
         else:
             # Fallback to string format if record not found

backend/apps/file_management_app.py

@@ -129,12 +129,13 @@ async def upload_files(
 
 @file_management_config_router.post("/process")
 async def process_files(
-        files: List[dict] = Body(
-            ..., description="List of file details to process, including path_or_url and filename"),
-        chunking_strategy: Optional[str] = Body("basic"),
-        index_name: str = Body(...),
-        destination: str = Body(...),
-        authorization: Optional[str] = Header(None)
+        files: Annotated[List[dict], Body(
+            ..., description="List of file details to process, including path_or_url and filename")],
+        index_name: Annotated[str, Body(...)],
+        destination: Annotated[str, Body(...)],
+        chunking_strategy: Annotated[Optional[str], Body(...)] = "basic",
+        model_id: Annotated[Optional[int], Body(...)] = None,
+        authorization: Annotated[Optional[str], Header()] = None
 ):
     """
     Trigger data processing for a list of uploaded files.
@@ -147,7 +148,8 @@ async def process_files(
         chunking_strategy=chunking_strategy,
         source_type=destination,
         index_name=index_name,
-        authorization=authorization
+        authorization=authorization,
+        model_id=model_id
     )
 
     process_result = await trigger_data_process(files, process_params)

backend/apps/model_managment_app.py

@@ -33,7 +33,7 @@
 from fastapi.responses import JSONResponse
 from fastapi.encoders import jsonable_encoder
 from http import HTTPStatus
-from typing import List, Optional
+from typing import Annotated, List, Optional
 from services.model_health_service import (
     check_model_connectivity,
     verify_model_config_connectivity,
@@ -298,7 +298,8 @@ async def get_llm_model_list(authorization: Optional[str] = Header(None)):
 
 @router.post("/healthcheck")
 async def check_model_health(
-        display_name: str = Query(..., description="Display name to check"),
+        display_name: Annotated[str, Query(..., description="Display name to check")],
+        model_type: Annotated[str, Query(..., description="...")],
         authorization: Optional[str] = Header(None)
 ):
     """Check and update model connectivity, returning the latest status.
@@ -309,7 +310,7 @@ async def check_model_health(
     """
     try:
         _, tenant_id = get_current_user_id(authorization)
-        result = await check_model_connectivity(display_name, tenant_id)
+        result = await check_model_connectivity(display_name, tenant_id, model_type)
         return JSONResponse(status_code=HTTPStatus.OK, content={
             "message": "Successfully checked model connectivity",
             "data": result

backend/apps/remote_mcp_app.py

@@ -122,6 +122,7 @@ async def add_mcp_service_endpoint(
             server_url=payload.server_url,
             tags=payload.tags,
             authorization_token=payload.authorization_token,
+            custom_headers=payload.custom_headers,
             container_config=payload.container_config,
             registry_json=payload.registry_json,
             enabled=payload.enabled if payload.enabled is not None else False,
@@ -239,6 +240,7 @@ async def update_mcp_service_endpoint(
             description=payload.description,
             server_url=payload.server_url,
             authorization_token=payload.authorization_token,
+            custom_headers=payload.custom_headers,
             tags=payload.tags,
         )
 
@@ -421,6 +423,7 @@ async def get_mcp_record(
                 "mcp_name": mcp_record.get("mcp_name"),
                 "mcp_server": mcp_record.get("mcp_server"),
                 "authorization_token": mcp_record.get("authorization_token"),
+                "custom_headers": mcp_record.get("custom_headers"),
                 "status": "success"
             }
         )

backend/apps/vectordatabase_app.py

@@ -84,11 +84,13 @@ def create_new_index(
         # Extract optional fields from request body
         ingroup_permission = None
         group_ids = None
-        embedding_model_name = None
+        embedding_model_name: Optional[str] = None
+        is_multimodal: Optional[bool] = None
         if request:
             ingroup_permission = request.get("ingroup_permission")
             group_ids = request.get("group_ids")
-            embedding_model_name = request.get("embedding_model_name")
+            embedding_model_name = request.get("embeddingModel")
+            is_multimodal = request.get("is_multimodal")
 
         # Treat path parameter as user-facing knowledge base name for new creations
         return ElasticSearchService.create_knowledge_base(
@@ -100,6 +102,7 @@ def create_new_index(
             ingroup_permission=ingroup_permission,
             group_ids=group_ids,
             embedding_model_name=embedding_model_name,
+            is_multimodal=is_multimodal,
         )
     except Exception as e:
         raise HTTPException(
@@ -714,6 +717,7 @@ def update_chunk(
             chunk_request=payload,
             vdb_core=vdb_core,
             user_id=user_id,
+            tenant_id=tenant_id,
         )
         return JSONResponse(status_code=HTTPStatus.OK, content=result)
     except ValueError as e:
@@ -778,8 +782,17 @@ async def hybrid_search(
     """Run a hybrid (accurate + semantic) search across indices."""
     try:
         _, tenant_id = get_current_user_id(authorization)
+        resolved_index_names: List[str] = []
+        for requested_name in payload.index_names:
+            try:
+                resolved_name = get_index_name_by_knowledge_name(
+                    requested_name, tenant_id
+                )
+            except Exception:
+                resolved_name = requested_name
+            resolved_index_names.append(resolved_name)
         result = ElasticSearchService.search_hybrid(
-            index_names=payload.index_names,
+            index_names=resolved_index_names,
             query=payload.query,
             tenant_id=tenant_id,
             top_k=payload.top_k,

backend/consts/const.py

@@ -31,6 +31,10 @@ class VectorDatabaseType(str, Enum):
 # Data Processing Service Configuration
 DATA_PROCESS_SERVICE = os.getenv("DATA_PROCESS_SERVICE")
 CLIP_MODEL_PATH = os.getenv("CLIP_MODEL_PATH")
+TABLE_TRANSFORMER_MODEL_PATH = os.getenv("TABLE_TRANSFORMER_MODEL_PATH")
+UNSTRUCTURED_DEFAULT_MODEL_INITIALIZE_PARAMS_JSON_PATH = os.getenv(
+    "UNSTRUCTURED_DEFAULT_MODEL_INITIALIZE_PARAMS_JSON_PATH"
+)
 
 
 # Upload Configuration
@@ -151,6 +155,7 @@ class VectorDatabaseType(str, Enum):
 MINIO_SECRET_KEY = os.getenv("MINIO_SECRET_KEY")
 MINIO_REGION = os.getenv("MINIO_REGION")
 MINIO_DEFAULT_BUCKET = os.getenv("MINIO_DEFAULT_BUCKET")
+S3_URL_PREFIX = "s3://"
 
 
 # Postgres Configuration

backend/consts/model.py

@@ -338,6 +338,7 @@ class ProcessParams(BaseModel):
     source_type: str
     index_name: str
     authorization: Optional[str] = None
+    model_id: Optional[int] = None
 
 
 class OpinionRequest(BaseModel):
@@ -679,6 +680,8 @@ class MCPUpdateRequest(BaseModel):
     new_mcp_url: str = Field(..., description="New MCP server URL")
     new_authorization_token: Optional[str] = Field(
         None, description="New authorization token for MCP server authentication (e.g., Bearer token)")
+    custom_headers: Optional[Dict[str, Any]] = Field(
+        None, description="Custom HTTP headers as JSON object")
 
 
 # Tenant Management Data Models
@@ -1118,6 +1121,7 @@ class AddMcpServiceRequest(BaseModel):
     source: MCPSourceType = Field(default=MCPSourceType.LOCAL, description="MCP source type")
     tags: List[str] = Field(default_factory=list, description="MCP tags")
     authorization_token: Optional[str] = Field(None, description="Authorization token for MCP server")
+    custom_headers: Optional[Dict[str, Any]] = Field(None, description="Custom HTTP headers as JSON object")
     container_config: Optional[Dict[str, Any]] = Field(None, description="Container configuration")
     registry_json: Optional[Dict[str, Any]] = Field(None, description="Registry metadata JSON")
     enabled: Optional[bool] = Field(default=False, description="Whether the MCP is enabled after creation")
@@ -1157,6 +1161,7 @@ class UpdateMcpServiceRequest(BaseModel):
     server_url: str = Field(..., min_length=1, description="New MCP server URL")
     tags: List[str] = Field(default_factory=list, description="MCP tags")
     authorization_token: Optional[str] = Field(None, description="Authorization token for MCP server")
+    custom_headers: Optional[Dict[str, Any]] = Field(None, description="Custom HTTP headers as JSON object")
 
     @field_validator("name", "server_url", "description", "authorization_token", mode="before")
     @classmethod