feat: Define workflow signal decorator and definition for registry

cadence/signal.py

@@ -0,0 +1,220 @@
+"""
+Signal definition and registration for Cadence workflows.
+
+This module provides functionality to define and register signal handlers
+for workflows, similar to ActivityDefinition but for signals.
+"""
+
+import inspect
+from dataclasses import dataclass
+from functools import update_wrapper
+from inspect import Parameter, signature
+from typing import (
+    Callable,
+    Generic,
+    ParamSpec,
+    Type,
+    TypeVar,
+    TypedDict,
+    Unpack,
+    overload,
+    get_type_hints,
+    Any,
+)
+
+P = ParamSpec("P")
+T = TypeVar("T")
+
+
+@dataclass(frozen=True)
+class SignalParameter:
+    """Parameter metadata for a signal handler."""
+
+    name: str
+    type_hint: Type | None
+    has_default: bool
+    default_value: Any
+
+
+class SignalDefinitionOptions(TypedDict, total=False):
+    """Options for defining a signal."""
+
+    name: str
+
+
+class SignalDefinition(Generic[P, T]):
+    """
+    Definition of a signal handler with metadata.
+
+    Similar to ActivityDefinition but for signal handlers.
+    Provides type safety and metadata for signal handlers.
+    """
+
+    def __init__(
+        self,
+        wrapped: Callable[P, T],
+        name: str,
+        params: list[SignalParameter],
+        is_async: bool,
+    ):
+        self._wrapped = wrapped
+        self._name = name
+        self._params = params
+        self._is_async = is_async
+        update_wrapper(self, wrapped)
+
+    def __call__(self, *args: P.args, **kwargs: P.kwargs) -> T:
+        """Call the wrapped signal handler function."""
+        return self._wrapped(*args, **kwargs)
+
+    @property
+    def name(self) -> str:
+        """Get the signal name."""
+        return self._name
+
+    @property
+    def params(self) -> list[SignalParameter]:
+        """Get the signal parameters."""
+        return self._params
+
+    @property
+    def is_async(self) -> bool:
+        """Check if the signal handler is async."""
+        return self._is_async
+
+    @property
+    def wrapped(self) -> Callable[P, T]:
+        """Get the wrapped signal handler function."""
+        return self._wrapped
+
+    @staticmethod
+    def wrap(
+        fn: Callable[P, T], opts: SignalDefinitionOptions
+    ) -> "SignalDefinition[P, T]":
+        """
+        Wrap a function as a SignalDefinition.
+
+        Args:
+            fn: The signal handler function to wrap
+            opts: Options for the signal definition
+
+        Returns:
+            A SignalDefinition instance
+
+        Raises:
+            ValueError: If name is not provided in options or return type is not None
+        """
+        name = opts.get("name") or fn.__qualname__
+        is_async = inspect.iscoroutinefunction(fn)
+        params = _get_signal_signature(fn)
+        _validate_signal_return_type(fn)
+
+        return SignalDefinition(fn, name, params, is_async)
+
+
+SignalDecorator = Callable[[Callable[P, T]], SignalDefinition[P, T]]
+
+
+@overload
+def defn(fn: Callable[P, T]) -> SignalDefinition[P, T]: ...
+
+
+@overload
+def defn(**kwargs: Unpack[SignalDefinitionOptions]) -> SignalDecorator: ...
+
+
+def defn(
+    fn: Callable[P, T] | None = None, **kwargs: Unpack[SignalDefinitionOptions]
+) -> SignalDecorator | SignalDefinition[P, T]:
+    """
+    Decorator to define a signal handler.
+
+    Can be used with or without parentheses:
+        @signal.defn(name="approval")
+        async def handle_approval(self, approved: bool):
+            ...
+
+        @signal.defn(name="approval")
+        def handle_approval(self, approved: bool):
+            ...
+
+    Args:
+        fn: The signal handler function to decorate
+        **kwargs: Options for the signal definition (name is required)
+
+    Returns:
+        The decorated function as a SignalDefinition instance
+
+    Raises:
+        ValueError: If name is not provided
+    """
+    options = SignalDefinitionOptions(**kwargs)
+
+    def decorator(inner_fn: Callable[P, T]) -> SignalDefinition[P, T]:
+        return SignalDefinition.wrap(inner_fn, options)
+
+    if fn is not None:
+        return decorator(fn)
+
+    return decorator
+
+
+def _validate_signal_return_type(fn: Callable) -> None:
+    """
+    Validate that signal handler returns None.
+
+    Args:
+        fn: The signal handler function
+
+    Raises:
+        ValueError: If return type is not None
+    """
+    try:
+        hints = get_type_hints(fn)
+        ret_type = hints.get("return", inspect.Signature.empty)
+
+        if ret_type is not None and ret_type is not inspect.Signature.empty:
+            raise ValueError(
+                f"Signal handler '{fn.__qualname__}' must return None "
+                f"(signals cannot return values), got {ret_type}"
+            )
+    except NameError:
+        pass
+
+
+def _get_signal_signature(fn: Callable[P, T]) -> list[SignalParameter]:
+    """
+    Extract parameter information from a signal handler function.
+
+    Args:
+        fn: The signal handler function
+
+    Returns:
+        List of SignalParameter objects
+
+    Raises:
+        ValueError: If parameters are not positional
+    """
+    sig = signature(fn)
+    args = sig.parameters
+    hints = get_type_hints(fn)
+    params = []
+
+    for name, param in args.items():
+        # Filter out the self parameter for instance methods
+        if param.name == "self":
+            continue
+
+        has_default = param.default != Parameter.empty
+        default = param.default if has_default else None
+
+        if param.kind in (Parameter.POSITIONAL_ONLY, Parameter.POSITIONAL_OR_KEYWORD):
+            type_hint = hints.get(name, None)
+            params.append(SignalParameter(name, type_hint, has_default, default))
+        else:
+            raise ValueError(
+                f"Signal handler '{fn.__qualname__}' parameter '{name}' must be positional, "
+                f"got {param.kind.name}"
+            )
+
+    return params

cadence/workflow.py

@@ -20,6 +20,7 @@
 
 from cadence.api.v1.history_pb2 import HistoryEvent
 from cadence.data_converter import DataConverter
+from cadence.signal import SignalDefinition, SignalDefinitionOptions
 
 ResultType = TypeVar("ResultType")
 
@@ -60,10 +61,22 @@ class WorkflowDefinition:
     Provides type safety and metadata for workflow classes.
     """
 
-    def __init__(self, cls: Type, name: str, run_method_name: str):
+    def __init__(
+        self,
+        cls: Type,
+        name: str,
+        run_method_name: str,
+        signals: dict[str, SignalDefinition[..., Any]],
+    ):
         self._cls = cls
         self._name = name
         self._run_method_name = run_method_name
+        self._signals = signals
+
+    @property
+    def signals(self) -> dict[str, SignalDefinition[..., Any]]:
+        """Get the signal definitions."""
+        return self._signals
 
     @property
     def name(self) -> str:
@@ -99,6 +112,11 @@ def wrap(cls: Type, opts: WorkflowDefinitionOptions) -> "WorkflowDefinition":
             name = opts["name"]
 
         # Validate that the class has exactly one run method and find it
+        # Also validate that class does not have multiple signal methods with the same name
+        signals: dict[str, SignalDefinition[..., Any]] = {}
+        signal_names: dict[
+            str, str
+        ] = {}  # Map signal name to method name for duplicate detection
         run_method_name = None
         for attr_name in dir(cls):
             if attr_name.startswith("_"):
@@ -116,10 +134,24 @@ def wrap(cls: Type, opts: WorkflowDefinitionOptions) -> "WorkflowDefinition":
                     )
                 run_method_name = attr_name
 
+            if hasattr(attr, "_workflow_signal"):
+                signal_name = getattr(attr, "_workflow_signal")
+                if signal_name in signal_names:
+                    raise ValueError(
+                        f"Multiple @workflow.signal methods found in class {cls.__name__} "
+                        f"with signal name '{signal_name}': '{attr_name}' and '{signal_names[signal_name]}'"
+                    )
+                # Create SignalDefinition from the decorated method
+                signal_def = SignalDefinition.wrap(
+                    attr, SignalDefinitionOptions(name=signal_name)
+                )
+                signals[signal_name] = signal_def
+                signal_names[signal_name] = attr_name
+
         if run_method_name is None:
             raise ValueError(f"No @workflow.run method found in class {cls.__name__}")
 
-        return WorkflowDefinition(cls, name, run_method_name)
+        return WorkflowDefinition(cls, name, run_method_name, signals)
 
 
 def run(func: Optional[T] = None) -> Union[T, Callable[[T], T]]:
@@ -163,6 +195,36 @@ def decorator(f: T) -> T:
         return decorator(func)
 
 
+def signal(name: str | None = None) -> Callable[[T], T]:
+    """
+    Decorator to mark a method as a workflow signal handler.
+
+    Example:
+        @workflow.signal(name="approval_channel")
+        async def approve(self, approved: bool):
+            self.approved = approved
+
+    Args:
+        name: The name of the signal
+
+    Returns:
+        The decorated method with workflow signal metadata
+
+    Raises:
+        ValueError: If name is not provided
+
+    """
+    if name is None:
+        raise ValueError("name is required")
+
+    def decorator(f: T) -> T:
+        f._workflow_signal = name  # type: ignore
+        return f
+
+    # Only allow @workflow.signal(name), require name to be explicitly provided
+    return decorator
+
+
 @dataclass
 class WorkflowInfo:
     workflow_type: str

tests/cadence/_internal/workflow/test_workflow_engine_integration.py

@@ -4,7 +4,7 @@
 """
 
 import pytest
-from unittest.mock import Mock, AsyncMock, patch
+from unittest.mock import Mock, patch
 from cadence.api.v1.service_worker_pb2 import PollForDecisionTaskResponse
 from cadence.api.v1.common_pb2 import Payload, WorkflowExecution, WorkflowType
 from cadence.api.v1.history_pb2 import (
@@ -250,7 +250,7 @@ async def test_extract_workflow_input_deserialization_error(
         """Test workflow input extraction with deserialization error."""
 
         # Mock data converter to raise an exception
-        mock_client.data_converter.from_data = AsyncMock(
+        mock_client.data_converter.from_data = Mock(
             side_effect=Exception("Deserialization error")
         )