feat(fuzzysort/algorithm): algorithm function return a shallow copy of prepare object

[pablolmiranda]

Problem

When algorithm function returns a reference to a cached object. In this case, if the upstream code holds a reference to it, any consequent invocation of the function with the same target will mutate the object and change the upstream result.

Example to reproduce the issue

const fuzzysort = require('./fuzzysort')

const term = 'Inventory Items Average Cos'
const leftText = 'Inventory Items'
const rightText = 'Average Cost'

const tokens = term.split(' ')
const tokensLen = tokens.length

const results = []
for(const tokenIndex in tokens) {
    const left = tokens.slice(0, tokenIndex);
    const right = tokens.slice(0 + tokenIndex,tokensLen);

    const leftMatch = fuzzysort.single(left.join(' '), leftText)
    const rightMatch = fuzzysort.single(right.join(' '), rightText)

    if (leftMatch && rightMatch) {
        console.log('Saving to results:')
        console.log(leftMatch)
        console.log(rightMatch)

        results.push({
            left: leftMatch,
            right: rightMatch
        })
    }
}

console.log('')
console.log('Saved results: ')
console.log(results)

This example generate the following output:

Saving to results:
{
  target: 'Inventory Items',
  _targetLowerCodes: [
    105, 110, 118, 101, 110,
    116, 111, 114, 121,  32,
    105, 116, 101, 109, 115
  ],
  _nextBeginningIndexes: [
     9,  9,  9,  9,  9,  9,
     9,  9,  9, 10, 15, 15,
    15, 15, 15
  ],
  score: 0,
  indexes: [
     0,  1,  2, 3,  4,  5,
     6,  7,  8, 9, 10, 11,
    12, 13, 14
  ],
  obj: null
}
{
  target: 'Average Cost',
  _targetLowerCodes: [
     97, 118, 101, 114,  97,
    103, 101,  32,  99, 111,
    115, 116
  ],
  _nextBeginningIndexes: [
     7,  7, 7,  7,  7,
     7,  7, 8, 12, 12,
    12, 12
  ],
  score: -1,
  indexes: [
     0, 1, 2, 3, 4,
     5, 6, 7, 8, 9,
    10
  ],
  obj: null
}

Saved results:
[
  {
    left: {
      target: 'Inventory Items',
      _targetLowerCodes: [Array],
      _nextBeginningIndexes: [Array],
      score: 0,
      indexes: [Array],
      obj: null
    },
    right: {
      target: 'Average Cost',
      _targetLowerCodes: [Array],
      _nextBeginningIndexes: [Array],
      score: -17,
      indexes: [Array],
      obj: null
    }
  }
]

Proposed solution

Make a shallow copy of the prepared object at the return moment of the algorithm function. This will not impact on the performance of the search function.