Core 2.0 - remove runtime exceptions

[laszlopandy]

Core 2.0:

• Remove runtime exceptions:
  • Changed type List.head : List a -> Maybe a
  • Deleted List.tail in favour of List.drop 1
  • Deleted List.foldl1, List.foldr1, List.scanl1
  • Added List.uncons
  • Changed List.maximum and List.minimum to take default values
• Added Graphics.Model
  • Including Debug will not include Collage or Element
  • Public types are re-exported from the same module as before.
  • Private types which were exposed before are now hidden: BasicForm, LineCap, LineJoin, ShapeStyle, FillStyle, ElementPrim, ImageStyle, Three.
  • Since no one should have been using the private types, I don't expect any breaking changes from the Graphics.Model refactor.
• Fixed compiling while doing breaking changes in core: run-test.sh will now delete the elm-package version of core and symlink it to the local git version.

[evancz]

Is it possible to do this in three batches? I think there are problems in the Graphics part, but not the others.

I believe some of the functions you hid are actually public. See this page for LineStyle, LineCap, LineStyle, and FillStyle.

[evancz]

Is it possible to do this in three batches? I think there are problems in the Graphics part, but not the others.

I believe some of the functions you hid are actually public. See this page for LineStyle, LineCap, LineStyle, and FillStyle.

[evancz]

Also, don't bump the version number. I prefer to do it with elm-package bump right before a release. This way you always know your reference point when you are changing stuff.

[evancz]

Also, don't bump the version number. I prefer to do it with elm-package bump right before a release. This way you always know your reference point when you are changing stuff.

[deadfoxygrandpa]

Basics has a possible runtime exception:

  function mod(a, b) {
        if (b === 0) {
            throw new Error("Cannot perform mod 0. Division by zero error.");
        }
        var r = a % b;
        var m = a === 0 ? 0 : (b > 0 ? (a >= 0 ? r : r+b) : -mod(-a,-b));

        return m === b ? 0 : m;
  }

Could this return NaN instead?

[deadfoxygrandpa]

Basics has a possible runtime exception:

  function mod(a, b) {
        if (b === 0) {
            throw new Error("Cannot perform mod 0. Division by zero error.");
        }
        var r = a % b;
        var m = a === 0 ? 0 : (b > 0 ? (a >= 0 ? r : r+b) : -mod(-a,-b));

        return m === b ? 0 : m;
  }

Could this return NaN instead?

[evancz]

@deadfoxygrandpa, I don't really know what is the right thing here, but an error sort of feels more correct. I'd be interested if there is a better/faster implementation in general though.

[evancz]

@deadfoxygrandpa, I don't really know what is the right thing here, but an error sort of feels more correct. I'd be interested if there is a better/faster implementation in general though.