inline exception handling in python

wheres pythonmonks wherespythonmonks at gmail.com
Thu Aug 12 20:19:13 CEST 2010


On Thu, Aug 12, 2010 at 2:08 PM, Thomas Jollans <thomas at jollybox.de> wrote:
> On Thursday 12 August 2010, it occurred to wheres pythonmonks to exclaim:
>> try:
>>    f = n / d
>> except:
>>    f = float("nan")
>
> A catch-all except clause. Never a good idea. It's not as bad in this case, as
> there is only one expression, but there are still a couple of other exceptions
> that have a chance of occurring here: KeyboardInterrupt and SystemExit.
> So:
>
> try:
>    f = n / d
> except ZeroDivisionError:
>    f = float('nan')
>
>
>> f = n / d except float("nan");
>
> So this syntax really isn't adequate for real use: catch-all except clauses
> are frowned upon, and rightfully so.
>
> Besides, more often than not, you want to have a finally clause around when
> you're dealing with exceptions.
>
>
>> (Obviously, I am thinking about more complicated functions than "n/d"
>> -- but this works as an example.)
>
> The more complex the function is, the more likely it is to raise an exception
> you can't handle that easily.
> --
> http://mail.python.org/mailman/listinfo/python-list
>

With a bit imagination the syntax can handle specific exceptions:

f = n /d except except(ZeroDivisionError) float("nan")

f = n /d except except(ZeroDivisionError) float("nan")
except(ValueError) float("nan")

But then we cannot bind to useful variable you say...

I think the problem in my case is best solved by look before you leap,
or a wrapper function.  [I just hate function call overhead for this.
]

Thanks,

W



More information about the Python-list mailing list