address review

Author: picnixz

Doc/library/math.rst

@@ -434,7 +434,7 @@ Floating point manipulation functions
 
 .. function:: signbit(x)
 
-   Return :const:`True` if *x* is negative and :const:`False` otherwise.
+   Return ``True`` if the sign of *x* is negative and ``False`` otherwise.
 
    This is useful to detect the sign bit of zeroes, infinities and NaNs.
 

Lib/test/test_math.py

@@ -18,12 +18,8 @@
 eps = 1E-05
 NAN = float('nan')
 NNAN = float('-nan')
-DNAN = decimal.Decimal("nan")
-DNNAN = decimal.Decimal("-nan")
 INF = float('inf')
 NINF = float('-inf')
-DINF = decimal.Decimal("inf")
-DNINF = decimal.Decimal("-inf")
 FLOAT_MAX = sys.float_info.max
 FLOAT_MIN = sys.float_info.min
 
@@ -481,10 +477,13 @@ def testCopysign(self):
         self.assertEqual(abs(math.copysign(2., NAN)), 2.)
 
     def test_signbit(self):
-        for arg in [0, 0., 1, 1., INF, NAN, DINF, DNAN]:
+        self.assertRaises(TypeError, math.signbit)
+        self.assertRaises(TypeError, math.signbit, '1.0')
+
+        for arg in [0, 0., 1, 1., INF, NAN]:
             with self.subTest('positive', arg=arg):
                 self.assertFalse(math.signbit(arg))
-        for arg in [-0., -1, -1., NINF, NNAN, DNINF, DNNAN]:
+        for arg in [-0., -1, -1., NINF, NNAN]:
             with self.subTest('negative', arg=arg):
                 self.assertTrue(math.signbit(arg))
 

Misc/NEWS.d/next/Library/2025-06-24-10-23-37.gh-issue-135853.6xDNOG.rst

@@ -1,2 +1,2 @@
 :mod:`math`: expose C99 :func:`~math.signbit` function to determine whether
-a floating-point value is negative. Patch by Bénédikt Tran.
+the sign bit of a floating-point value is set. Patch by Bénédikt Tran.

Modules/mathmodule.c

@@ -1240,7 +1240,7 @@ math.signbit
     x: double
     /
 
-Return True if 'x' is negative and False otherwise.
+Return True if the sign of 'x' is negative and False otherwise.
 
 This is useful to detect the sign bit of zeroes, infinities and NaNs.
 [clinic start generated code]*/