Map a curried function over an array using lenses

Question

I have an object and want to update each value in an array inside of the object using some other object member. Here is what I am attempting:

import * as R from 'ramda'

const obj = {
    array: [ 1, 2, 3 ],
    incrementBy: 0
};

const doIncrement = R.curry((inc,val) => {
    return val+inc;
});

const incrementLens = R.lensProp('incrementBy');
const setIncrementToOne = R.set(incrementLens,1);

const newObj = setIncrementToOne(obj);

const arrayLens = R.lensProp('array');

const incrementA = R.over(arrayLens, R.map(doIncrement(1)));

const incrementB = R.pipe(
    setIncrementToOne,
    R.chain(
        R.over(arrayLens, R.map(doIncrement)),
        R.view(incrementLens)
    )
)

console.log(JSON.stringify(incrementA(newObj))); // works: {"array":[2,3,4],"incrementBy":1}
console.log(JSON.stringify(incrementB(obj))); // does not work (TypeError: Cannot read property 'fantasy-land/map' of undefined)

Version A obviously works as I access the array directly and map my curried function over it. In version B, however, I am unsure of the order of things; what I've tried now does not appear to pass the value from R.view to the curried function appropriately.

(NB. that the use case is more complex, I would just like to understand how I can work with curried functions, maps and lenses simultaneously.)

Answer 1

When R.chain is applied to functions - chain(f, g)(x) is equivalent to f(g(x))(x) - applying function g on value 'x', returning, and then applying function f on the result of g(x) and then on value x .

In your case:

• g(x) ( R.view(incrementLens) ) is 1
• f ( R.over(arrayLens, R.map(doIncrement)) ) is a function that waits for the value (not another function)

Applying f ( R.over... ) on g(x) applies a lens on a non object value, and even if it was an object the result would have been another value, which means that you don't have a function to apply to on the result of setIncrementToOne in the pipe.

The result of g(x) in this case needs to be a function, that can be used by R.over with the arraylens - to get the R.over(arrayLens, R.map(R.add(1))) , which is identical to incrementA .

Note: I've replaced doIncrement with R.add , which does the same thing.

 const obj = { array: [1, 2, 3], incrementBy: 0 }; const incrementLens = R.lensProp('incrementBy'); const setIncrementToOne = R.set(incrementLens, 1); const newObj = setIncrementToOne(obj); const arrayLens = R.lensProp('array'); const incrementB = R.pipe( setIncrementToOne, R.chain( R.over(arrayLens), R.pipe(R.view(incrementLens), R.add, R.map) ) ) const incrementA = R.over(arrayLens, R.map(R.add(1))); console.log(JSON.stringify(incrementA(newObj))); // works: {"array":[2,3,4],"incrementBy":1} console.log(JSON.stringify(incrementB(obj))); // works: {"array":[2,3,4],"incrementBy":1}

 <script src="https://cdnjs.cloudflare.com/ajax/libs/ramda/0.26.1/ramda.js"></script>