[Tutor] Dividing a float derived from a string

Peter Otten __peter__ at web.de
Fri Nov 21 15:15:21 CET 2014

Alan Gauld wrote:

> But that's considered bad practice, it's better to put the
> valid errors only in the except line like this:
> try:
>      print float(input)*12
> except TypeError, ValueError:
>      print False

Careful, you need parens around the tuple of errors, otherwise this catches 
only TypeErrors and assigns the TypeError instance to ValueError:

>>> input = None
>>> try: print float(input) * 12
... except TypeError, ValueError: print False
>>> ValueError
TypeError('float() argument must be a string or a number',)

This is a nasty gotcha best avoided by using Python 3.

More information about the Tutor mailing list