Merge branch 'main' into list05

Author: czgdp1807

integration_tests/test_list_04.py

@@ -1,4 +1,3 @@
-from random import randrange
 from math import sqrt
 from ltypes import i32
 

integration_tests/test_math.py

@@ -1,7 +1,7 @@
 from math import (factorial, isqrt, perm, comb, degrees, radians, exp, pow,
-                  ldexp, fabs, gcd, lcm, floor, ceil, remainder, expm1, fmod, log1p, trunc, fsum, prod)
-from ltypes import i8, i16, i32, i64, f32, f64
-from numpy import empty
+                  ldexp, fabs, gcd, lcm, floor, ceil, remainder, expm1, fmod, log1p, trunc,
+                  modf, fsum, prod)
+from ltypes import i32, i64, f32, f64
 
 eps: f64
 eps = 1e-12
@@ -191,7 +191,6 @@ def test_fsum():
     res = fsum(arr_f64)
     assert abs(res - 19.0) < eps
 
-
 def test_prod():
     res: f64
     eps: f64 = 1e-12
@@ -220,6 +219,20 @@ def test_prod():
     res = prod(arr_f64)
     assert abs(res - 80.64) < eps
 
+def test_modf():
+    i: f64
+    i = 3.14
+
+    res: tuple[f64, f64]
+    res = modf(i)
+    assert abs(res[0] - 0.14) <= 1e-6
+    assert abs(res[1] - 3.0) <= 1e-6
+
+    i = -442.3
+    res = modf(i)
+    assert abs(res[0] + 0.3) <= 1e-6
+    assert abs(res[1] + 442.0) <= 1e-6
+
 
 def check():
     test_factorial_1()
@@ -243,6 +256,7 @@ def check():
     test_trunc()
     test_fsum()
     test_prod()
+    test_modf()
 
 
 check()

src/libasr/runtime/lfortran_intrinsics.c

@@ -732,7 +732,7 @@ LFORTRAN_API int32_t _lpython_bit_length4(int32_t num)
 LFORTRAN_API int32_t _lpython_bit_length8(int64_t num)
 {
     int32_t res = 0;
-    num = abs(num);
+    num = llabs(num);
     for(; num; num >>= 1, res++);
     return res;
 }

src/runtime/math.py

@@ -7,6 +7,16 @@
 
 # TODO: Change floor used inside functions implemented here to
 # floordiv operator (//) once the multiple import issue is fixed
+
+@overload
+def modf(x: f64) -> tuple[f64, f64]:
+    """
+    Return fractional and integral parts of `x` as a pair.
+
+    Both results carry the sign of x and are floats.
+    """
+    return (x - int(x), float(int(x)))
+
 @overload
 def factorial(x: i32) -> i32:
     """