Question

I have a variable containing a string (extracted from a XML feed). The string value can be of type integer, date or string. I need to convert it from string to given data type. I am doing it this way but it is a little bit ugly so I am asking if there is a better technique. If I would checking for more types, I will end with very nested try – except blocks.

def normalize_availability(self, value):
    """
    Normalize the availability date.
    """
    try:
        val = int(value)
    except ValueError:
        try:
            val = datetime.datetime.strptime(value, '%Y-%m-%d')
        except (ValueError, TypeError):
            # Here could be another try - except block if more types needed
            val = value

Thanks!

Answer 1

Use a handy helper function.

def tryconvert(value, default, *types):
    """Converts value to one of the given types.  The first type that succeeds is
       used, so the types should be specified from most-picky to least-picky (e.g.
       int before float).  The default is returned if all types fail to convert
       the value.  The types needn't actually be types (any callable that takes a
       single argument and returns a value will work)."""
    value = value.strip()
    for t in types:
        try:
            return t(value)
        except (ValueError, TypeError):
            pass
    return default

Then write a function for parsing the date/time:

def parsedatetime(value, format="%Y-%m-%d")
    return datetime.datetime.striptime(value, format)

Now put ’em together:

value = tryconvert(value, None, parsedatetime, int)