On Thu, Nov 22, 2018 at 06:33:54AM +1100, Chris Angelico wrote: > On Thu, Nov 22, 2018 at 6:23 AM Python <pyt...@bladeshadow.org> wrote: > > How is the first not equivalent to either one of the second or third? > > My expectation is it should produce the same result as the second. It > > *seems* like Python is ignoring the '1 in' part and just giving the > > result for '[1,2,3] == True'... Is this just a bug? > > > > Don't jump to assume it's a bug, because that kind of bug is extremely > unlikely compared to a misunderstanding :)
Not assuming, just asking the question, FWIW... > > In Python, chained comparisons behave as if the middle term is shared. > A common and useful example is: > > 1 < x < 10 > 1 < x and x < 10 That's kind of cool, but also a bit mind-breaking, if you're generally familiar with how other languages evaluate expressions, but not aware of this feature of Python. [Hi!] It's less clear this is useful with "in" but since python considers it a comparison operator, it's at least consistent, which is good. > apart from the fact that x is only evaluated once. It's not common to > use "in" and "==" in this way, but it's perfectly legal. What you > wrote is equivalent to: > > 1 in [1, 2, 3] and [1, 2, 3] == True > > which should explain the result you got. It's only confusing because > you added a redundant "== True" to the end of an otherwise simple > comparison :) The "redundant" comparison is not the reason for the confusion... I wasn't familiar with chaining comparisons. I imagine it should be obvious that this was a simplified example constructed to illustrate the problem most simply, rather than real code. After all, a comparison between two hard-coded constant-value expressions need never be evaluated, since the value is itself known and constant. Thanks for the explanation. -- https://mail.python.org/mailman/listinfo/python-list
