On Thu, Jan 19, 2017 at 10:53 PM, Jeroen Demeyer <jdeme...@cage.ugent.be> wrote: > On 2017-01-19 17:44, William Stein wrote: >> >> On Thu, Jan 19, 2017 at 9:39 AM, Jeroen Demeyer <jdeme...@cage.ugent.be> >> wrote: >>> >>> On 2017-01-19 17:17, William Stein wrote: >>>> >>>> >>>> ... and presumably what we already decided is now completely >>>> impossible to implement and banned from Python3? >>> >>> >>> >>> I don't think we ever decided anything. For Parents, we use Robert >>> Bradshaw's implementation from 2008 but it's not documented what it does >>> or >>> why it does things that way. For example, there is a doctest >>> >>> sage: ZZ < QQ >>> True >>> >>> but I doubt that this is the right thing to do. >> >> >> Let me rewrite the above: ... and presumably what we already >> **implemented** is now completely >> impossible to implement and banned from Python3? > > > It's mostly possible. However, what's the point of trying to reverse > engineer the current code when it probably doesn't do the right thing? Since > we have to change comparison code anyway for Python 3, we should use this > opportunity to properly define how comparison should work. > > Secondly, Python 3 has different "best practices" for comparison compared to > Python 2. Example: > > Python 2.7.5: >>>> 1 < "x" > True > > Python 3.4.1: >>>> 1 < "x" > Traceback (most recent call last): > File "<stdin>", line 1, in <module> > TypeError: unorderable types: int() < str()
To be clear, while Python 3 got rid of comparisons between built-in types that were ill-defined, that doesn't prevent one from implementing comparisons between one's own types however we want: Python 3.4.5 (default, Oct 10 2016, 14:41:48) [GCC 5.4.0] on cygwin Type "help", "copyright", "credits" or "license" for more information. >>> class myint(int): ... def __lt__(self, other): ... if isinstance(other, str): ... return True ... return super().__lt__(other) ... >>> 1 < "x" Traceback (most recent call last): File "<stdin>", line 1, in <module> TypeError: unorderable types: int() < str() >>> myint(1) < "x" True -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
