Math migration

[JamieBallingall]

Introduces many of the functions currently living in purescript-math. Closes #17. See also purescript-deprecated/purescript-math#31.

---

Checklist:

• Added the change to the changelog's "Unreleased" section with a reference to this PR (e.g. "- Made a change (#0000)")
• Linked any existing issues or proposals that this pull request should close
• Updated or added relevant documentation
• Added a test for the contribution (if applicable)

[hdgarrood]

I realise that the readme already had examples in, but I think examples should go into the doc-comments instead, so that they can be seen on Pursuit.

[JamieBallingall]

I agree about examples in the doc-comments. I'd be happy to completely rework the README to give a shorter and more general overview of the functions Data.Number supports but that will mean changing what was already there before this PR.

[hdgarrood]

That sounds good to me, thanks!

[JamieBallingall]

After some delay (my apologies), this should now be resolved in 626e299

[hdgarrood]

We have an opportunity to clarify that sin takes radians here; how about using pi/2 as the value?

[JamieBallingall]

I agree but I missed this in the last commit. I'll fix shortly.

[JamieBallingall]

Fixed in 73c1e25

[JamieBallingall]

@hdgarrood Over on purescript/purescript#2980 you express hesitancy on polyfills that mutate the Math object. This repo contains a polyfill for sign that reads:

exports.sign = Math.sign ? Math.sign : function(x) {
  return x === 0 || x !== x ? x : (x < 0 ? -1 : 1);
};

That doesn't look to me like it mutates the Math object but I'm certainly no expert on Javascript. If you have concerns over sign we can remove the Javascript implementation and implement sign directly in Purescript, at some (theoretical) performance cost?

[hdgarrood]

No, that looks fine! That won’t mutate Math, it’ll just use Math.sign if it exists and otherwise it’ll use the polyfill.

[JamieBallingall]

One final tweak. I've switched trunc to work that same way as sign -- using the Math object if available with a polyfill backup.

[JordanMartinez]

LGTM!

As noted in the Math issue (my comment), we need to decide whether to replicate more functions added to the Math object in ES6.

[JordanMartinez]

🏓 @thomashoneyman This should be ready to go now.

[thomashoneyman]

I haven't been involved in these discussions, so I'd appreciate if perhaps @hdgarrood could take another look. If not, I can review, but I don't necessarily have the correct background to make these decisions.

[hdgarrood]

I have a couple of minor docs suggestions but this look great 👍

[hdgarrood]

I feel like exp 1.0 might be a better example here; any function of the form \x -> pow a x for some a is going to take 0.0 to 1.0.

[JordanMartinez]

Fixed in latest commit.

[hdgarrood]

Same here - maybe log e would be more illustrative here? I realise you have that same example on the documentation for e but I think repetition here is fine, especially as people might be linked directly to the documentation for exp and log and not see e.

[JordanMartinez]

Fixed in latest commit.