Fix rerank_model bug in chat and markdown bug (#4061)

### What problem does this PR solve?

Fix rerank_model bug in chat and markdown bug
#4000
#3992
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

---------

Co-authored-by: liuhua <10215101452@stu.ecun.edu.cn>

api/apps/sdk/chat.py

@@ -82,7 +82,8 @@ def create(tenant_id):
     req["top_k"] = req.get("top_k", 1024)
     req["rerank_id"] = req.get("rerank_id", "")
     if req.get("rerank_id"):
-        if not TenantLLMService.query(tenant_id=tenant_id,llm_name=req.get("rerank_id"),model_type="rerank"):
+        value_rerank_model = ["BAAI/bge-reranker-v2-m3","maidalun1020/bce-reranker-base_v1"]
+        if req["rerank_id"] not in value_rerank_model and not TenantLLMService.query(tenant_id=tenant_id,llm_name=req.get("rerank_id"),model_type="rerank"):
             return get_error_data_result(f"`rerank_model` {req.get('rerank_id')} doesn't exist")
     if not req.get("llm_id"):
         req["llm_id"] = tenant.llm_id
@@ -161,7 +162,7 @@ def update(tenant_id,chat_id):
         req["do_refer"]=req.pop("show_quotation")
     if "dataset_ids" in req:
         if not ids:
-            return get_error_data_result("`datasets` can't be empty")
+            return get_error_data_result("`dataset_ids` can't be empty")
         if ids:
             for kb_id in ids:
                 kbs = KnowledgebaseService.accessible(kb_id=kb_id, user_id=tenant_id)
@@ -188,9 +189,6 @@ def update(tenant_id,chat_id):
     e, tenant = TenantService.get_by_id(tenant_id)
     if not e:
         return get_error_data_result(message="Tenant not found!")
-    if req.get("rerank_model"):
-        if not TenantLLMService.query(tenant_id=tenant_id,llm_name=req.get("rerank_model"),model_type="rerank"):
-            return get_error_data_result(f"`rerank_model` {req.get('rerank_model')} doesn't exist")
     # prompt
     prompt = req.get("prompt")
     key_mapping = {"parameters": "variables",
@@ -210,6 +208,10 @@ def update(tenant_id,chat_id):
         req["prompt_config"] = req.pop("prompt")
     e, res = DialogService.get_by_id(chat_id)
     res = res.to_json()
+    if req.get("rerank_id"):
+        value_rerank_model = ["BAAI/bge-reranker-v2-m3","maidalun1020/bce-reranker-base_v1"]
+        if req["rerank_id"] not in value_rerank_model and not TenantLLMService.query(tenant_id=tenant_id,llm_name=req.get("rerank_id"),model_type="rerank"):
+            return get_error_data_result(f"`rerank_model` {req.get('rerank_id')} doesn't exist")
     if "name" in req:
         if not req.get("name"):
             return get_error_data_result(message="`name` is not empty.")

docs/references/http_api_reference.md

@@ -2243,7 +2243,7 @@ Success:
                         "data": {
                             "form": {},
                             "label": "Answer",
-                            "name": "对话_0"
+                            "name": "dialog_0"
                         },
                         "dragging": false,
                         "height": 44,
@@ -2362,7 +2362,7 @@ curl --request POST \
      --data-binary '
      {
           "lang":"English"
-          "file":"明天天气如何"
+          "file":"How is the weather tomorrow?"
      }'
 ```
 

docs/references/python_api_reference.md

@@ -1435,7 +1435,7 @@ session = create_session(AGENT_ID,rag_object)
 
 ---
 
-## Converse with agent without `begin` component
+## Converse with agent
 
 ```python
 Session.ask(question: str, stream: bool = False) -> Optional[Message, iter[Message]]

rag/app/naive.py

@@ -169,11 +169,13 @@ class Markdown(MarkdownParser):
         sections = []
         tbls = []
         for sec in remainder.split("\n"):
-            if num_tokens_from_string(sec) > 10 * self.chunk_token_num:
+            if num_tokens_from_string(sec) > 3 * self.chunk_token_num:
                 sections.append((sec[:int(len(sec) / 2)], ""))
                 sections.append((sec[int(len(sec) / 2):], ""))
             else:
-                if sections and sections[-1][0].strip().find("#") == 0:
+                if sec.strip().find("#") == 0:
+                    sections.append((sec, ""))
+                elif sections and sections[-1][0].strip().find("#") == 0:
                     sec_, _ = sections.pop(-1)
                     sections.append((sec_ + "\n" + sec, ""))
                 else: