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
As I can see, the accuracy argument in all functions ends up being used like this:
returnMath.round(distance/accuracy)*accuracy;
It makes no sense to me, because this rounding is not necessary and it can be done outside the library. I thought the accuracy affects the execution time so user has to choose a compromise, but that's false.
The "problem" is it's hard to get the original distance value. This can be done by using accuracy equal Math.pow(2, -52). This is ridiculous.
One solution I see is removing the accuracy argument, but this is a breaking change. Another solution is allowing a new accuracy value like null or 0 that disables the rounding.
As I can see, the
accuracyargument in all functions ends up being used like this:It makes no sense to me, because this rounding is not necessary and it can be done outside the library. I thought the accuracy affects the execution time so user has to choose a compromise, but that's false.
The "problem" is it's hard to get the original
distancevalue. This can be done by usingaccuracyequalMath.pow(2, -52). This is ridiculous.One solution I see is removing the
accuracyargument, but this is a breaking change. Another solution is allowing a newaccuracyvalue likenullor0that disables the rounding.