When attempting to convert a float("nan"), float("inf"), or float("-inf"), fractions.Fraction() raises different exceptions than int()

>>> int(float("nan"))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: cannot convert float NaN to integer
>>> fractions.Fraction(float("nan"))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/opt/local/lib/python3.2/fractions.py", line 114, in __new__
    value = Fraction.from_float(numerator)
  File "/opt/local/lib/python3.2/fractions.py", line 186, in from_float
    raise TypeError("Cannot convert %r to %s." % (f, cls.__name__))
TypeError: Cannot convert nan to Fraction.
>>> int(float("inf"))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: cannot convert float infinity to integer
>>> fractions.Fraction(float("inf"))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/opt/local/lib/python3.2/fractions.py", line 114, in __new__
    value = Fraction.from_float(numerator)
  File "/opt/local/lib/python3.2/fractions.py", line 186, in from_float
    raise TypeError("Cannot convert %r to %s." % (f, cls.__name__))
TypeError: Cannot convert inf to Fraction.
>>> int(float("-inf"))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: cannot convert float infinity to integer
>>> fractions.Fraction(float("-inf"))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/opt/local/lib/python3.2/fractions.py", line 114, in __new__
    value = Fraction.from_float(numerator)
  File "/opt/local/lib/python3.2/fractions.py", line 186, in from_float
    raise TypeError("Cannot convert %r to %s." % (f, cls.__name__))
TypeError: Cannot convert -inf to Fraction.

Should the exceptions be changed to ValueError and OverflowError?

Certainly TypeError seems inappropriate here, and using ValueError for conversions from NaN sounds good to me.

I'm not a big fan of the OverflowError for converting infinities to an integer:  nothing's actually overflowed here.  I think that should have been ValueError, too.  So either of ValueError or OverflowError works for me there.

I'd suggest fixing this for >= 3.4 only.

The from_decimal method should be changed as well.

The description of OverflowError is that is is "raised when the result of an arithmetic operation is too large to be represented", so it doesn't actually need to overflow. Still, I see that ∞ actually isn't too large to be represented (and the documentation points out that integer operations will raise MemoryError instead of OverflowError). So yes, ValueError seems more appropriate.

+1 for fixing this in 3.4, only.

Do we also want to change int(infinity) to return ValueError? I think consistent behavior between int() and Fraction() is valuable.

> I think consistent behavior between int() and Fraction() is valuable.

Agreed.  If no-one objects I'll fix the Decimal -> Fraction and float -> Fraction conversions to match what Decimal -> int and float -> int currently do (ValueError and OverflowError).

Then we can open a new issue to discuss having infinities give ValueError versus OverflowError for all four cases of {float, Decimal} -> {int, Fraction}.

Patch is attached for {float, Decimal} -> {int, Fraction} consistency (with ValueError and OverflowError). Test cases are changed as well. 

I can also change OverflowError to ValueError for all cases, but I'm not sure should I open for this separated issue or continue to work with this one.

I also attached "full_issue" patch: change OverflowError to ValueError for all cases.

The first patch looks great (modulo a stray extra line in test_int). Thank you!

Let's just apply that for now; I'll open a new issue for the suggested OverflowError / ValueError switch.  (Getting rid of the TypeError seems a clear win to me;  switching the OverflowError to ValueError may be more controversial.)

Fixed in http://hg.python.org/cpython/rev/a2b54b6d9759.  Thanks for the report and patch.

See issue #16483 for the ValueError / OverflowError switch.  To be honest, while I would have preferred ValueError to OverflowError for int(float('inf')) if we were doing this from scratch, I'm not convinced that it's actually worth switching.

Ok, I think that ValueError is more appropriate and I have already attached adopted patch for issue 16483.