True == 1 weirdness

[Jussi Piitulainen]

Blake T. Garretson writes:

> I am maintaining some old code where the programmer used 1 for True
> because booleans hadn't been added to Python yet.  I'm getting some
> weird behaviour, so I created some simple tests to illustrate my
> issue.
>
>   >>> 1 in {1:1}        #test1
>   True
>   >>> 1 in {1:1} == 1   #test2
>   False
>   >>> (1 in {1:1}) == 1 #test3
>   True
>   >>> 1 in ({1:1} == 1) #test4
>   Traceback (most recent call last):
>     File "<stdin>", line 1, in <module>
>   TypeError: argument of type 'bool' is not iterable
>   >>>

Ouch. I love chained comparisons more than most people, but this took a
while to decipher. I blame you! Your parentheses mis-primed me for the
wrong reading :) But now I expect to see a long thread about whether
chained comparisons are a natural thing to have in the language.

The second test, test2, is interpreted (almost) as

  >>>> (1 in {1:1}) and ({1:1} == 1)

which is obviously False.