Operator Precedence/Boolean Logic

[Steven D'Aprano]

On Thursday 23 June 2016 14:47, Lawrence D’Oliveiro wrote:

> On Thursday, June 23, 2016 at 3:12:52 PM UTC+12, Larry Hudson wrote:
>> On 06/22/2016 12:42 AM, Lawrence D’Oliveiro wrote:
>>> * boolean operators don’t have to operate on boolean values. The
>>>   language spec
>>>   <https://docs.python.org/3/reference/expressions.html#boolean-operations>
>>>   says:
>>>
>>>     “...the following values are interpreted as false: False, None, numeric
>>>     zero of all types, and empty strings and containers (including strings,
>>>     tuples, lists, dictionaries, sets and frozensets). All other values are
>>>     interpreted as true.”
>>>
>>> I feel that’s a needlessly complicated rule.

As I described in my earlier email, it isn't complicated, at least not the way 
builtins are modelled.


>>> It would have been simpler if
>>> boolean operators (and conditional expressions like in if-statements and
>>> while-statements) only allowed values of boolean types. But that’s one of
>>> the few warts in the design of Python...
>> 
>> Wart??  I *strongly* disagree.  I find it one of the strengths of Python,
>> it enhances Python's expressiveness.
> 
> Tightening it up would rule out a whole class of common errors,

Hardly common. The only two exceptions I've seen are:

- people surprised by midnight being false, but that is fixed in 3.5; 
- people surprised that empty or exhausted iterables don't evaluate as false

and the first of those is pretty rare.

But even if you were right, and I disagree that you are, "fixing" this would 
break backwards compatibility and cause vast amounts of working code to stop 
working. That is much worse than the "problem".


> from
> misunderstanding (or forgetting) the rule about what exactly gets interpreted
> as true and what as false <https://bugs.python.org/issue13936>
> <http://stackoverflow.com/questions/28116931/datetime-time0-0-evaluates-as-
false-in-boolean-context>.

Or... we could fix time objects, as was done.


-- 
Steve