[New-bugs-announce] [issue5759] Do not call __float__ to classec derived from str

[Alexandr Zamaraev]

New submission from Alexandr Zamaraev <shura_zam at users.sourceforge.net>:

Test case:
[code]
class S:
    def __init__(self, v):
        self.data = v

    def __int__(self):
        print("S.INT called")
        return int(str(self.data))
    def __float__(self):
        print("S.FLOAT called")
        return float(str(self.data))
        

class T(str):
    def __int__(self):
        print("T.INT called")
        return int(str(self))
    def __float__(self):
        print("T.FLOAT called")
        return float(str(self))

class U(unicode):
    def __int__(self):
        print("U.INT called")
        return int(unicode(self))
    def __float__(self):
        print("U.FLOAT called")
        return float(unicode(self))


i = S("123")
print(type(int(i)))
print(type(float(i)))

i = T("123")
print(type(int(i)))
print(type(float(i))) # <<< CALLS __float__ NOTHING

i = U("123")
print(type(int(i)))
print(type(float(i)))
[/code]
Output:
[code]
S.INT called
<type 'int'>
S.FLOAT called
<type 'float'>
T.INT called
<type 'int'>
<type 'float'>
U.INT called
<type 'int'>
U.FLOAT called
<type 'float'>
[/code]

----------
components: None
messages: 85979
nosy: shura_zam
severity: normal
status: open
title: Do not call __float__ to classec derived from str
versions: Python 2.5

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue5759>
_______________________________________