Improve get_func_args (#5885)

scrapy/utils/python.py

@@ -174,33 +174,33 @@ def binary_is_text(data):
 
 
 def get_func_args(func, stripself=False):
-    """Return the argument name list of a callable"""
-    if inspect.isfunction(func):
-        spec = inspect.getfullargspec(func)
-        func_args = spec.args + spec.kwonlyargs
-    elif inspect.isclass(func):
-        return get_func_args(func.__init__, True)
-    elif inspect.ismethod(func):
-        return get_func_args(func.__func__, True)
-    elif inspect.ismethoddescriptor(func):
-        return []
-    elif isinstance(func, partial):
-        return [
-            x
-            for x in get_func_args(func.func)[len(func.args) :]
-            if not (func.keywords and x in func.keywords)
-        ]
-    elif hasattr(func, "__call__"):
-        if inspect.isroutine(func):
-            return []
-        if getattr(func, "__name__", None) == "__call__":
-            return []
-        return get_func_args(func.__call__, True)
+    """Return the argument name list of a callable object"""
+    if not callable(func):
+        raise TypeError(f"func must be callable, got '{type(func).__name__}'")
+
+    args = []
+    try:
+        sig = inspect.signature(func)
+    except ValueError:
+        return args
+
+    if isinstance(func, partial):
+        partial_args = func.args
+        partial_kw = func.keywords
+
+        for name, param in sig.parameters.items():
+            if param.name in partial_args:
+                continue
+            if partial_kw and param.name in partial_kw:
+                continue
+            args.append(name)
     else:
-        raise TypeError(f"{type(func)} is not callable")
-    if stripself:
-        func_args.pop(0)
-    return func_args
+        for name in sig.parameters.keys():
+            args.append(name)
+
+    if stripself and args and args[0] == "self":
+        args = args[1:]
+    return args
 
 
 def get_spec(func):

tests/test_utils_python.py

@@ -235,20 +235,16 @@ class UtilsPythonTestCase(unittest.TestCase):
         self.assertEqual(get_func_args(partial_f3), ["c"])
         self.assertEqual(get_func_args(cal), ["a", "b", "c"])
         self.assertEqual(get_func_args(object), [])
+        self.assertEqual(get_func_args(str.split, stripself=True), ["sep", "maxsplit"])
+        self.assertEqual(get_func_args(" ".join, stripself=True), ["iterable"])
 
         if platform.python_implementation() == "CPython":
-            # TODO: how do we fix this to return the actual argument names?
-            self.assertEqual(get_func_args(str.split), [])
-            self.assertEqual(get_func_args(" ".join), [])
+            # doesn't work on CPython: https://bugs.python.org/issue42785
             self.assertEqual(get_func_args(operator.itemgetter(2)), [])
         elif platform.python_implementation() == "PyPy":
-            self.assertEqual(
-                get_func_args(str.split, stripself=True), ["sep", "maxsplit"]
-            )
             self.assertEqual(
                 get_func_args(operator.itemgetter(2), stripself=True), ["obj"]
             )
-            self.assertEqual(get_func_args(" ".join, stripself=True), ["iterable"])
 
     def test_without_none_values(self):
         self.assertEqual(without_none_values([1, None, 3, 4]), [1, 3, 4])