Merge commit '44dec89f1fd6eb98f2d8d5c0137074990a09a99c' into feature/support_encrypted_files_20251209

* commit '44dec89f1fd6eb98f2d8d5c0137074990a09a99c': (28 commits)
  Fix: aspose-slide issue. (infiniflow#11935)
  Fix: raptor don't have attribute chat (infiniflow#11936)
  Feat: Add GPT-5.2 & pro (infiniflow#11929)
  Refa: refactor metadata filter (infiniflow#11907)
  Feat: Displaying the file option in the webhook's request body infiniflow#10427 (infiniflow#11928)
  Fix: forget-reset password (infiniflow#11927)
  Feature/docs generator (infiniflow#11858)
  Fix: correct metadata update behavior (infiniflow#11919)
  Docs: How to use restful API to update or delete metadata (infiniflow#11912)
  Feat: Add box connector (infiniflow#11845)
  Feat: Flatten the request schema of the webhook infiniflow#10427 (infiniflow#11917)
  feat: Add Single Bucket Mode for MinIO/S3 (infiniflow#11416)
  Fix: tokenizer issue. (infiniflow#11902)
  Feat: Add mineru as a model manufacturer to the system. infiniflow#10621 (infiniflow#11903)
  Refa: treat MinerU as an OCR model 2 (infiniflow#11905)
  Add DeepseekV3.2 of Tongyi-Qianwen and remove unused code (infiniflow#11898)
  Fix:async issue and sensitive logging (infiniflow#11895)
  Added semi-automatic mode to the metadata filter (infiniflow#11886)
  Fix data_sync startup crash by properly invoking async main (infiniflow#11879)
  MinerU supports for the new backend vlm-mlx-engine (infiniflow#11864)
  ...

# Conflicts:
#	docker/.env
#	pyproject.toml

Author: virgilwong

.github/workflows/tests.yml

@@ -1,4 +1,6 @@
 name: tests
+permissions:
+  contents: read
 
 on:
   push:

.gitignore

@@ -195,3 +195,6 @@ ragflow_cli.egg-info
 
 # Default backup dir
 backup
+
+
+.hypothesis

Dockerfile

@@ -52,7 +52,8 @@ RUN --mount=type=cache,id=ragflow_apt,target=/var/cache/apt,sharing=locked \
     apt install -y nginx unzip curl wget git vim less && \
     apt install -y ghostscript && \
     apt install -y pandoc && \
-    apt install -y texlive
+    apt install -y texlive && \
+    apt install -y fonts-freefont-ttf fonts-noto-cjk
 
 # Install uv
 RUN --mount=type=bind,from=infiniflow/ragflow_deps:latest,source=/,target=/deps \

admin/client/admin_client.py

@@ -351,7 +351,7 @@ def parse_command(self, command_str: str) -> dict[str, str]:
     def verify_admin(self, arguments: dict, single_command: bool):
         self.host = arguments['host']
         self.port = arguments['port']
-        print(f"Attempt to access ip: {self.host}, port: {self.port}")
+        print("Attempt to access server for admin login")
         url = f"http://{self.host}:{self.port}/api/v1/admin/login"
 
         attempt_count = 3
@@ -390,7 +390,7 @@ def verify_admin(self, arguments: dict, single_command: bool):
                     print(f"Bad response，status: {response.status_code}, password is wrong")
             except Exception as e:
                 print(str(e))
-                print(f"Can't access {self.host}, port: {self.port}")
+                print("Can't access server for admin login (connection failed)")
 
     def _format_service_detail_table(self, data):
         if isinstance(data, list):
@@ -674,7 +674,7 @@ def _handle_alter_user(self, command):
         user_name: str = user_name_tree.children[0].strip("'\"")
         password_tree: Tree = command['password']
         password: str = password_tree.children[0].strip("'\"")
-        print(f"Alter user: {user_name}, password: {password}")
+        print(f"Alter user: {user_name}, password: ******")
         url = f'http://{self.host}:{self.port}/api/v1/admin/users/{user_name}/password'
         response = self.session.put(url, json={'new_password': encrypt(password)})
         res_json = response.json()
@@ -689,7 +689,7 @@ def _handle_create_user(self, command):
         password_tree: Tree = command['password']
         password: str = password_tree.children[0].strip("'\"")
         role: str = command['role']
-        print(f"Create user: {user_name}, password: {password}, role: {role}")
+        print(f"Create user: {user_name}, password: ******, role: {role}")
         url = f'http://{self.host}:{self.port}/api/v1/admin/users'
         response = self.session.post(
             url,
@@ -951,7 +951,7 @@ def main():
 
     args = cli.parse_connection_args(sys.argv)
     if 'error' in args:
-        print(f"Error: {args['error']}")
+        print("Error: Invalid connection arguments")
         return
 
     if 'command' in args:
@@ -960,7 +960,7 @@ def main():
             return
         if cli.verify_admin(args, single_command=True):
             command: str = args['command']
-            print(f"Run single command: {command}")
+            # print(f"Run single command: {command}")
             cli.run_single_command(command)
     else:
         if cli.verify_admin(args, single_command=False):

admin/server/auth.py

@@ -176,11 +176,11 @@ def decorated(*args, **kwargs):
                     "message": "Access denied",
                     "data": None
                 }), 200
-        except Exception as e:
-            error_msg = str(e)
+        except Exception:
+            logging.exception("An error occurred during admin login verification.")
             return jsonify({
                 "code": 500,
-                "message": error_msg
+                "message": "An internal server error occurred."
             }), 200
 
         return f(*args, **kwargs)

agent/component/agent_with_tools.py

@@ -271,7 +271,7 @@ async def _react_with_tools_streamly_async(self, prompt, history: list[dict], us
         last_calling = ""
         if len(hist) > 3:
             st = timer()
-            user_request = await asyncio.to_thread(full_question, messages=history, chat_mdl=self.chat_mdl)
+            user_request = await full_question(messages=history, chat_mdl=self.chat_mdl)
             self.callback("Multi-turn conversation optimization", {}, user_request, elapsed_time=timer()-st)
         else:
             user_request = history[-1]["content"]
@@ -309,7 +309,7 @@ async def complete():
             if len(hist) > 12:
                 _hist = [hist[0], hist[1], *hist[-10:]]
             entire_txt = ""
-            async for delta_ans in self._generate_streamly_async(_hist):
+            async for delta_ans in self._generate_streamly(_hist):
                 if not need2cite or cited:
                     yield delta_ans, 0
                 entire_txt += delta_ans
@@ -397,7 +397,7 @@ async def _gen_citations_async(self, text):
         retrievals = self._canvas.get_reference()
         retrievals = {"chunks": list(retrievals["chunks"].values()), "doc_aggs": list(retrievals["doc_aggs"].values())}
         formated_refer = kb_prompt(retrievals, self.chat_mdl.max_length, True)
-        async for delta_ans in self._generate_streamly_async([{"role": "system", "content": citation_plus("\n\n".join(formated_refer))},
+        async for delta_ans in self._generate_streamly([{"role": "system", "content": citation_plus("\n\n".join(formated_refer))},
                                                   {"role": "user", "content": text}
                                                   ]):
             yield delta_ans

agent/component/categorize.py

@@ -13,6 +13,7 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 #
+import asyncio
 import logging
 import os
 import re
@@ -97,7 +98,7 @@ class Categorize(LLM, ABC):
     component_name = "Categorize"
 
     @timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 10*60)))
-    def _invoke(self, **kwargs):
+    async def _invoke_async(self, **kwargs):
         if self.check_if_canceled("Categorize processing"):
             return
 
@@ -121,7 +122,7 @@ def _invoke(self, **kwargs):
         if self.check_if_canceled("Categorize processing"):
             return
 
-        ans = chat_mdl.chat(self._param.sys_prompt, [{"role": "user", "content": user_prompt}], self._param.gen_conf())
+        ans = await chat_mdl.async_chat(self._param.sys_prompt, [{"role": "user", "content": user_prompt}], self._param.gen_conf())
         logging.info(f"input: {user_prompt}, answer: {str(ans)}")
         if ERROR_PREFIX in ans:
             raise Exception(ans)
@@ -144,5 +145,9 @@ def _invoke(self, **kwargs):
         self.set_output("category_name", max_category)
         self.set_output("_next", cpn_ids)
 
+    @timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 10*60)))
+    def _invoke(self, **kwargs):
+        return asyncio.run(self._invoke_async(**kwargs))
+
     def thoughts(self) -> str:
         return "Which should it falls into {}? ...".format(",".join([f"`{c}`" for c, _ in self._param.category_description.items()]))