You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Decorum is opinionated about the total ordering it provides for floating-point values. Users cannot modify this behavior, and it is essentially hard-coded (see the canonical and constraint modules).
Maybe this should be parameterized via an additional type parameter. That parameter should have a default, and that default should probably use the "natural" ordering provided today (i.e., NaN and zero have single canonical forms). However, some users may want to use something more akin to the IEEE-754 ordering, where -NaN is less than -INF, NaN is greater than INF, and negative zero is less than zero (i.e., there is a distinction between negative and positive variants of NaN and zero).
I've seen some discussion about this regarding similar libraries, and I think alternative ordering may be useful for some applications.
The text was updated successfully, but these errors were encountered:
Removing the 0.1.0 milestone. The practical use for this isn't entirely clear to me (I'm struggling to think of a compelling example). For 0.1.0, I think it's okay to omit such a feature. I'll leave this open for future consideration.
One of Decorum's primary features is a total ordering for floating-point types. The exact ordering isn't too important (NaN and its many representations are the main problem when it comes to ordering), and allowing multiple orderings to be used would likely cause more trouble then it's worth. I'm closing this for now, but it could be worth more discussion if a convincing use case comes along.
Decorum is opinionated about the total ordering it provides for floating-point values. Users cannot modify this behavior, and it is essentially hard-coded (see the
canonical
andconstraint
modules).Maybe this should be parameterized via an additional type parameter. That parameter should have a default, and that default should probably use the "natural" ordering provided today (i.e.,
NaN
and zero have single canonical forms). However, some users may want to use something more akin to the IEEE-754 ordering, where-NaN
is less than-INF
,NaN
is greater thanINF
, and negative zero is less than zero (i.e., there is a distinction between negative and positive variants ofNaN
and zero).I've seen some discussion about this regarding similar libraries, and I think alternative ordering may be useful for some applications.
The text was updated successfully, but these errors were encountered: