Merge pull request #212 from robotpy/tunable-empty-seq

tunable: Allow empty default lists when type-hinted

Author: virtuald

Diff for: magicbot/magic_tunable.py

@@ -80,6 +80,7 @@ def execute(self):
         "_ntsubtable",
         "_ntwritedefault",
         # "__doc__",
+        "__orig_class__",
         "_topic_type",
         "_nt",
     )
@@ -100,13 +101,48 @@ def __init__(
         self._ntwritedefault = writeDefault
         # self.__doc__ = doc
 
-        self._topic_type = _get_topic_type_for_value(self._ntdefault)
-        if self._topic_type is None:
-            checked_type: type = type(self._ntdefault)
+        # Defer checks for empty sequences to check type hints.
+        # Report errors here when we can so the error points to the tunable line.
+        if default or not isinstance(default, collections.abc.Sequence):
+            topic_type = _get_topic_type_for_value(default)
+            if topic_type is None:
+                checked_type: type = type(default)
+                raise TypeError(
+                    f"tunable is not publishable to NetworkTables, type: {checked_type.__name__}"
+                )
+            self._topic_type = topic_type
+
+    def __set_name__(self, owner: type, name: str) -> None:
+        type_hint: Optional[type] = None
+        # __orig_class__ is set after __init__, check it here.
+        orig_class = getattr(self, "__orig_class__", None)
+        if orig_class is not None:
+            # Accept field = tunable[Sequence[int]]([])
+            type_hint = typing.get_args(orig_class)[0]
+        else:
+            type_hint = typing.get_type_hints(owner).get(name)
+            origin = typing.get_origin(type_hint)
+            if origin is typing.ClassVar:
+                # Accept field: ClassVar[tunable[Sequence[int]]] = tunable([])
+                type_hint = typing.get_args(type_hint)[0]
+                origin = typing.get_origin(type_hint)
+            if origin is tunable:
+                # Accept field: tunable[Sequence[int]] = tunable([])
+                type_hint = typing.get_args(type_hint)[0]
+
+        if type_hint is not None:
+            topic_type = _get_topic_type(type_hint)
+        else:
+            topic_type = _get_topic_type_for_value(self._ntdefault)
+
+        if topic_type is None:
+            checked_type: type = type_hint or type(self._ntdefault)
             raise TypeError(
                 f"tunable is not publishable to NetworkTables, type: {checked_type.__name__}"
             )
 
+        self._topic_type = topic_type
+
     @overload
     def __get__(self, instance: None, owner=None) -> "tunable[V]": ...
 
@@ -218,7 +254,7 @@ class MyComponent:
             navx: ...
 
             @feedback
-            def get_angle(self):
+            def get_angle(self) -> float:
                 return self.navx.getYaw()
 
         class MyRobot(magicbot.MagicRobot):
@@ -297,6 +333,8 @@ def _get_topic_type(
         if hasattr(inner_type, "WPIStruct"):
             return lambda topic: ntcore.StructArrayTopic(topic, inner_type)
 
+    return None
+
 
 def collect_feedbacks(component, cname: str, prefix: Optional[str] = "components"):
     """

Diff for: tests/test_magicbot_tunable.py

@@ -1,3 +1,5 @@
+from typing import ClassVar, List, Sequence
+
 import ntcore
 import pytest
 from wpimath import geometry
@@ -25,6 +27,7 @@ class Component:
         topic = nt.getTopic(name)
         assert topic.getTypeString() == type_str
         assert topic.genericSubscribe().get().value() == value
+        assert getattr(component, name) == value
 
     for name, value in [
         ("rotation", geometry.Rotation2d()),
@@ -33,13 +36,15 @@ class Component:
         assert nt.getTopic(name).getTypeString() == f"struct:{struct_type.__name__}"
         topic = nt.getStructTopic(name, struct_type)
         assert topic.subscribe(None).get() == value
+        assert getattr(component, name) == value
 
     for name, struct_type, value in [
         ("rotations", geometry.Rotation2d, [geometry.Rotation2d()]),
     ]:
         assert nt.getTopic(name).getTypeString() == f"struct:{struct_type.__name__}[]"
         topic = nt.getStructArrayTopic(name, struct_type)
         assert topic.subscribe([]).get() == value
+        assert getattr(component, name) == value
 
 
 def test_tunable_errors():
@@ -50,7 +55,44 @@ class Component:
 
 
 def test_tunable_errors_with_empty_sequence():
-    with pytest.raises(ValueError):
+    with pytest.raises((RuntimeError, ValueError)):
 
         class Component:
             empty = tunable([])
+
+
+def test_type_hinted_empty_sequences() -> None:
+    class Component:
+        generic_seq = tunable[Sequence[int]](())
+        class_var_seq: ClassVar[tunable[Sequence[int]]] = tunable(())
+        inst_seq: Sequence[int] = tunable(())
+
+        generic_typing_list = tunable[List[int]]([])
+        class_var_typing_list: ClassVar[tunable[List[int]]] = tunable([])
+        inst_typing_list: List[int] = tunable([])
+
+        # TODO(davo): re-enable after py3.8 is dropped
+        # generic_list = tunable[list[int]]([])
+        # class_var_list: ClassVar[tunable[list[int]]] = tunable([])
+        # inst_list: list[int] = tunable([])
+
+    component = Component()
+    setup_tunables(component, "test_type_hinted_sequences")
+    NetworkTables = ntcore.NetworkTableInstance.getDefault()
+    nt = NetworkTables.getTable("/components/test_type_hinted_sequences")
+
+    for name in [
+        "generic_seq",
+        "class_var_seq",
+        "inst_seq",
+        "generic_typing_list",
+        "class_var_typing_list",
+        "inst_typing_list",
+        # "generic_list",
+        # "class_var_list",
+        # "inst_list",
+    ]:
+        assert nt.getTopic(name).getTypeString() == "int[]"
+        entry = nt.getEntry(name)
+        assert entry.getIntegerArray(None) == []
+        assert getattr(component, name) == []