Add builtin pow() (fix #294)

stdlib/internal/builtin.codon

@@ -323,6 +323,29 @@ def oct(n):
 def hex(n):
     return _int_format(n.__index__(), 16, "0x")
 
+def pow(base: float, exp: float):
+    return base ** exp
+
+@overload
+def pow(base: int, exp: int, mod: Optional[int] = None):
+    if exp < 0:
+        raise ValueError("pow() negative int exponent not supported")
+
+    if mod is not None:
+        if mod == 0:
+            raise ValueError("pow() 3rd argument cannot be 0")
+        base %= mod
+
+    result = 1
+    while exp > 0:
+        if exp & 1:
+            x = result * base
+            result = x % mod if mod is not None else x
+        y = base * base
+        base = y % mod if mod is not None else y
+        exp >>= 1
+    return result % mod if mod is not None else result
+
 @extend
 class int:
     def _from_str(s: str, base: int):

test/core/bltin.codon

@@ -258,6 +258,34 @@ def test_divmod():
         assert math.isclose(result[0], exp_result[0])
         assert math.isclose(result[1], exp_result[1])
 
+@test
+def test_pow():
+    assert pow(3, 4) == 81
+    assert pow(-3, 3) == -27
+    assert pow(1, 0) == 1
+    assert pow(-1, 0) == 1
+    assert pow(0, 0) == 1
+    assert pow(12, 12, 42) == 36
+    assert pow(1234, 4321, 99) == 46
+    assert pow(9999, 9999, 2) == 1
+    assert pow(0, 0, 1) == 0
+
+    try:
+        pow(1, -1, 2)
+        assert False
+    except ValueError as e:
+        assert 'negative' in str(e)
+
+    try:
+        pow(1, 1, 0)
+        assert False
+    except ValueError as e:
+        assert 'cannot be 0' in str(e)
+
+    assert pow(1.5, 2) == 2.25
+    assert pow(9, 0.5) == 3.0
+    assert pow(2.0, -1.0) == 0.5
+
 @test
 def test_num_from_str():
     assert int('0') == 0
@@ -337,6 +365,7 @@ test_gen_builtins()
 test_int_format()
 test_reversed()
 test_divmod()
+test_pow()
 test_num_from_str()
 test_files(open)
 import gzip