On Jun 4, 12:42 pm, Ian Kelly <ian.g.ke...@gmail.com> wrote:
> > By this manner, we can roll three common tests into one
> > method:
> > * Boolean conversion
> > * member truthiness for iterables
> > * type checking
> How exactly does this is_valid method perform the first two? Are you
> suggesting that an empty sequence should not be considered "valid"?
I'm suggesting that the rules for Python's current "implicit
conversion to Boolean" simply be moved into a "explicit function"
named "isvalid", that also does a type check. Here is some Python code
that might help you understand.
py> def isvalid(object_, type_):
... if isinstance(object_, type_) and object_:
... return True
... return False
py> isvalid([], list)
False
py> isvalid([1], list)
True
py> isvalid(0, int)
False
py> isvalid(1, int)
True
py> isvalid({}, dict)
False
py> isvalid("", str)
False
py> isvalid(" ", str)
True
Now, let's go back to my earlier example of where i was expecting a
list but got a string instead. If i use Python's current implicit
conversion to Boolean my code will do something i don't want it to do.
py> lst = " "
py> if lst:
... print("I'm a liar")
... else:
... print("I'm honest")
I'm a liar
But unlike this simple example (which failed quickly) in the real
world, it may not fail for a long time. And when it does fail, you
will be pulling your hair out tracking down the origin of the bug. If
however i use my "isvalid" function, my conditional will not lie to
me:
py> lst = " "
py> if isvalid(lst, list):
... print("I'm a liar")
... else:
... print("I'm honest")
I'm honest
Now. You're not always going to need to "isvalid" function. Sometimes
you just need to test type, sometimes you just need convert to
Boolean, and sometimes you can just fly by the seat of your pants. The
point is, remove implicitly and inject explicitly.
Furthermore: If the current "implicit conversion to Boolean" can be
optimized by Python, then there is no reason why an explicit "isvalid"
function cannot -- any talk to the contrary is just BS.
If you still feel that this idea is garbage, then, keep on writing
your sloppy code. My proposal is the best method to handle the
problems that arise with duck typed languages in a manner that is not
restrictive or laborious -- it's actually quite elegant.
*school-bell-rings*
--
http://mail.python.org/mailman/listinfo/python-list
