converting float to int issue

[Alex]

Irmen de Jong wrote:

> Alex wrote:
>> Converting a float to an int produces strange results.  Is this just a
>> bug
>> or am I missing something?  I tried to search the bug database, but was
>> unsucsessful.
>> 
>> Converting a float to a long works as I would expect.
>> 
>> Python 2.2.2 (#1, Apr 16 2003, 22:00:16)
>> [GCC 3.2.1 20021207 (Gentoo Linux 3.2.1-20021207)] on linux2
>> Type "help", "copyright", "credits" or "license" for more information.
>> 
>>>>>int(1.0)
>> 
>> 0
>> 
>>>>>int(3.0)
>> 
>> 0
>> 
>>>>>int(3.14)
>> 
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in ?
>> OverflowError: float too large to convert
> 
> Something weird is going on with your specific Python and/or system.
> I verified this and it works fine on Python 2.2.2.
> 
> Is your GCC broken? Is your libc broken? Did you compile with
> esoteric optimization options? One of these is what I'm betting on.
> 
> --Irmen.
> 
> 
> PS did you try to run Python's test set? "make test"

No esoteric build options.  A 'vanilla' gentoo box, if there is such a
thing.  Everything has been working fine for months, so no obvious problems
with gcc or libc.

I'll try "make test", as you suggest, and then try building a different
version of libc.  I guess I should troll the gentoo newsgroups too.

Thanks,
Alex