WIP - API iteration #69
Conversation
There was a problem hiding this comment.
Pull Request Overview
This PR appears to be a work-in-progress API iteration that introduces significant debugging output and a new decorator-based API for defining workflows and nodes. The changes include extensive print statements for debugging event flow and message processing, along with a comprehensive new decorator system for workflow definition.
- Addition of extensive debugging print statements throughout the workflow execution pipeline
- Implementation of a new decorator-based API system for defining workflows, nodes, and their relationships
- Addition of traceback printing for better error debugging in workflow execution
Reviewed Changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 5 comments.
Show a summary per file
| File | Description |
|---|---|
| grafi/workflows/impl/event_driven_workflow.py | Adds debug print statements and traceback output for workflow event processing |
| grafi/nodes/node.py | Adds debug print statement in node invocation |
| grafi/common/models/command.py | Adds debug print statement for command message yielding |
| grafi/common/decorators/record_workflow_a_invoke.py | Adds debug print statements for workflow invocation decorator |
| grafi/common/decorators/record_node_a_invoke.py | Adds debug print statements for node invocation decorator |
| grafi/api/decorators.py | Introduces new comprehensive decorator-based API for workflow and node definition |
| ) -> MsgsAGen: | ||
| if self._a_invoke_impl is None: | ||
| raise NotImplementedError("Must provide `_a__invoke_impl` Callable.") | ||
| yield self._a_invoke_impl(self, invoke_context, input_data) |
There was a problem hiding this comment.
The method _a_invoke_impl is being called with self as the first parameter, but it's defined as a Callable that expects (InvokeContext, List[Message]). This will cause a TypeError as the callable doesn't expect a self parameter.
| yield self._a_invoke_impl(self, invoke_context, input_data) | |
| yield self._a_invoke_impl(invoke_context, input_data) |
|
|
||
| all_subscibed_topics = {topic.name: False for topic in self.subscribed_topics} | ||
| available_topics = all_subscibed_topics | {topic.name: True for topic in topic_with_messahes} | ||
| return eval(self._node_condiiton, __builtins__ ={}, globals=None, locals=available_topics) |
There was a problem hiding this comment.
There's a typo in the variable name _node_condiiton - it should be _node_condition to match the attribute defined in the constructor.
| return eval(self._node_condiiton, __builtins__ ={}, globals=None, locals=available_topics) | |
| return eval(self._node_condition, __builtins__ ={}, globals=None, locals=available_topics) |
|
|
||
| all_subscibed_topics = {topic.name: False for topic in self.subscribed_topics} | ||
| available_topics = all_subscibed_topics | {topic.name: True for topic in topic_with_messahes} | ||
| return eval(self._node_condiiton, __builtins__ ={}, globals=None, locals=available_topics) |
There was a problem hiding this comment.
Using eval() with user-provided input is a security risk. Consider using a safer expression evaluation library like ast.literal_eval() or a dedicated expression parser to prevent code injection attacks.
| consumed_events=consumed_events, | ||
| ) | ||
| ) | ||
| print("NDOE OUTPUT") |
There was a problem hiding this comment.
Typo in debug message: 'NDOE' should be 'NODE'.
| print("NDOE OUTPUT") | |
| print("NODE OUTPUT") |
| _a_invoke_impl: Callable[[InvokeContext, List[Message]], MsgsAGen] = None | ||
|
|
||
| def __init__(self, tool_invoke: Callable[[InvokeContext, Messages], MsgsAGen], **kwargs): | ||
| super().__init__(name = "CallableTool", oi_span_type=OpenInferenceSpanKindValues.TOOL, type="CallableTool", **kwargs) |
There was a problem hiding this comment.
[nitpick] Inconsistent spacing around the = operator. Should be consistent with Python PEP 8 style guide (no spaces around = in keyword arguments).
| super().__init__(name = "CallableTool", oi_span_type=OpenInferenceSpanKindValues.TOOL, type="CallableTool", **kwargs) | |
| super().__init__(name="CallableTool", oi_span_type=OpenInferenceSpanKindValues.TOOL, type="CallableTool", **kwargs) |
No description provided.