Merge branch 'langgenius:main' into main

api/app.py

@@ -25,6 +25,10 @@ def is_db_command():
         # grpc gevent
         grpc_gevent.init_gevent()
 
+        import psycogreen.gevent  # type: ignore
+
+        psycogreen.gevent.patch_psycopg()
+
     from app_factory import create_app
 
     app = create_app()

api/core/app/apps/workflow/app_generator.py

@@ -221,6 +221,7 @@ def single_iteration_generate(
             single_iteration_run=WorkflowAppGenerateEntity.SingleIterationRunEntity(
                 node_id=node_id, inputs=args["inputs"]
             ),
+            workflow_run_id=str(uuid.uuid4()),
         )
         contexts.tenant_id.set(application_generate_entity.app_config.tenant_id)
 

api/core/app/entities/app_invoke_entities.py

@@ -195,7 +195,7 @@ class WorkflowAppGenerateEntity(AppGenerateEntity):
 
     # app config
     app_config: WorkflowUIBasedAppConfig
-    workflow_run_id: Optional[str] = None
+    workflow_run_id: str
 
     class SingleIterationRunEntity(BaseModel):
         """

api/core/app/task_pipeline/workflow_cycle_manage.py

@@ -102,7 +102,8 @@ def _handle_workflow_run_start(
         inputs = dict(WorkflowEntry.handle_special_values(inputs) or {})
 
         # init workflow run
-        workflow_run_id = str(self._workflow_system_variables.get(SystemVariableKey.WORKFLOW_RUN_ID, uuid4()))
+        # TODO: This workflow_run_id should always not be None, maybe we can use a more elegant way to handle this
+        workflow_run_id = str(self._workflow_system_variables.get(SystemVariableKey.WORKFLOW_RUN_ID) or uuid4())
 
         workflow_run = WorkflowRun()
         workflow_run.id = workflow_run_id

api/core/model_runtime/model_providers/__base/tokenizers/gpt2_tokenzier.py

@@ -1,11 +1,13 @@
 from os.path import abspath, dirname, join
 from threading import Lock
-from typing import Any
+from typing import Any, cast
 
+import gevent.threadpool  # type: ignore
 from transformers import GPT2Tokenizer as TransformerGPT2Tokenizer  # type: ignore
 
 _tokenizer: Any = None
 _lock = Lock()
+_pool = gevent.threadpool.ThreadPool(1)
 
 
 class GPT2Tokenizer:
@@ -20,7 +22,9 @@ def _get_num_tokens_by_gpt2(text: str) -> int:
 
     @staticmethod
     def get_num_tokens(text: str) -> int:
-        return GPT2Tokenizer._get_num_tokens_by_gpt2(text)
+        future = _pool.spawn(GPT2Tokenizer._get_num_tokens_by_gpt2, text)
+        result = future.get(block=True)
+        return cast(int, result)
 
     @staticmethod
     def get_encoder() -> Any:

api/core/workflow/nodes/http_request/executor.py

@@ -158,7 +158,10 @@ def _init_body(self):
                     if len(data) != 1:
                         raise RequestBodyError("json body type should have exactly one item")
                     json_string = self.variable_pool.convert_template(data[0].value).text
-                    json_object = json.loads(json_string, strict=False)
+                    try:
+                        json_object = json.loads(json_string, strict=False)
+                    except json.JSONDecodeError as e:
+                        raise RequestBodyError(f"Failed to parse JSON: {json_string}") from e
                     self.json = json_object
                     # self.json = self._parse_object_contains_variables(json_object)
                 case "binary":

api/core/workflow/workflow_entry.py

@@ -239,6 +239,10 @@ def mapping_user_inputs_to_variable_pool(
             ):
                 raise ValueError(f"Variable key {node_variable} not found in user inputs.")
 
+            # environment variable already exist in variable pool, not from user inputs
+            if variable_pool.get(variable_selector):
+                continue
+
             # fetch variable node id from variable selector
             variable_node_id = variable_selector[0]
             variable_key_list = variable_selector[1:]

api/libs/oauth_data_source.py

@@ -255,7 +255,8 @@ def notion_block_parent_page_id(self, access_token: str, block_id: str):
         response = requests.get(url=f"{self._NOTION_BLOCK_SEARCH}/{block_id}", headers=headers)
         response_json = response.json()
         if response.status_code != 200:
-            raise ValueError(f"Error fetching block parent page ID: {response_json.message}")
+            message = response_json.get("message", "unknown error")
+            raise ValueError(f"Error fetching block parent page ID: {message}")
         parent = response_json["parent"]
         parent_type = parent["type"]
         if parent_type == "block_id":

api/pyproject.toml

@@ -61,6 +61,7 @@ openai = "~1.52.0"
 openpyxl = "~3.1.5"
 pandas = { version = "~2.2.2", extras = ["performance", "excel"] }
 pandas-stubs = "~2.2.3.241009"
+psycogreen = "~1.0.2"
 psycopg2-binary = "~2.9.6"
 pycryptodome = "3.19.1"
 pydantic = "~2.9.2"