List Commands Library for 50g

[DavidM]

(09-09-2017 01:28 PM)David Hayden Wrote:  I think rounding is more appropriate. Consider the case where where some calculation is used to compute the index. If the calculation results in an index of 0.999999999 or 1.000000001, the result of GET should probably be the same.

On the other hand, I can't recall a single instance in decades of programming where I've needed complex calculations to compute an index that would result in needing rounding.

I think the reason I've leaned toward using %IP># over COERCE in most of the commands is in consideration of situations where a given list's size is simply divided by some constant. Seems like IP is safer than rounding for that (to me). Regardless of that, though, my current thinking is that a precedent has already been set by the built-in commands, and following that precedent is more important. I just need to go through all the commands and find where I've IP'd an index (or count) to assess if the issue applies and what impact a change would make. It's on my task list, just haven't done it yet.