如何根据起始值和结束值将数组分组到第一次迭代，并减少

Question

在下面的句子中，在Asteroid A at 0和d at 7一词中，space is counted as 8之后的位置。为了一张清晰的照片，我下面有很多空间。

 // 0-7,     9-10,  12,  14-18, 20-26,   28-31,  33-37,  39-41,  43-46
 // Asteroid is      a   rocky  objects  that    orbit   the     Sun

现在，我在12 - 19这里有一个带有word 12 - 19的对象，19th是空白。

   {
      "start_offset": 12,
      "end": 19,
      "text": "a rocky",
      "entity_type": "adjective",
    },

现在，我应该找到这个start and end之间的所有单词，并将其推入上面的splits键下的对象中，如下所示。

           {
              "start_offset": 12,
              "end": 19,
              "text": "a rocky",
              "entity_type": "adjective",
              "splits": [
                {
                  "start_offset": 14,
                  "end": 19,
                  "text": "rocky",
                  "entity_type": "adjective",
                },
              ]
            }, 

这个迭代，我需要做n次，最后应该分组所有的元素给出的输出。

现在，我已经尝试了像下面这样的接近的结果，但仍然需要很多改进。有人能指点我吗？

const res = arr.reduce((pv, cv) => {
    const [{ start_offset, end }] = arr
      .filter((s) => (s.start_offset <= cv.start_offset) && (s.end >= cv.end))
      .sort((s1, s2) => (s2.end - s2.start_offset) - (s1.end - s1.start_offset));
      
    const hash = `${start_offset}-${end}`;
    pv[hash] = pv[hash]
      ? { ...pv[hash], splits: [...pv[hash].splits, cv] }
      : { start_offset, end, splits: [cv] };
    
    return pv;
}, {});
const result = Object.values(res);
console.log(result)

给定输入：

let arr = [
    {
      "start_offset": 0,
      "end": 38,
      "text": "Asteroid is a rocky objects that orbit",
      "entity_type": "adjective",
    },
    {
      "start_offset": 12,
      "end": 19,
      "text": "a rocky",
      "entity_type": "adjective",
    },
    {
      "start_offset": 14,
      "end": 27,
      "text": "rocky objects",
      "entity_type": "adjective",
    },
    {
      "start_offset": 20,
      "end": 32,
      "text": "objects that",
      "entity_type": "adjective",
    },
    {
      "start_offset": 14,
      "end": 19,
      "text": "rocky",
      "entity_type": "adjective",
    },
    {
      "start_offset": 20,
      "end": 27,
      "text": "objects",
      "entity_type": "adjective",
    },
    {
      "start_offset": 33,
      "end": 47,
      "text": "orbit the Sun",
      "entity_type": "adjective",
    },
    {
      "start_offset": 43,
      "end": 47,
      "text": "Sun",
      "entity_type": "adjective",
    }
  ]

预期输出：

  let output = [
    {
      "start_offset": 0,
      "end": 38,
      "text": "Asteroid is a rocky objects that orbit",
      "entity_type": "adjective",
      "splits": [
            {
              "start_offset": 12,
              "end": 19,
              "text": "a rocky",
              "entity_type": "adjective",
              "splits": [
                {
                  "start_offset": 14,
                  "end": 19,
                  "text": "rocky",
                  "entity_type": "adjective",
                },
              ]
            },
            {
              "start_offset": 14,
              "end": 27,
              "text": "rocky objects",
              "entity_type": "adjective",
              "splits": [
                {
                  "start_offset": 20,
                  "end": 27,
                  "text": "objects",
                  "entity_type": "adjective",
                },
              ]
            },
            {
              "start_offset": 20,
              "end": 32,
              "text": "objects that",
              "entity_type": "adjective",
            },
      ]
    },
    {
      "start_offset": 33,
      "end": 47,
      "text": "orbit the Sun",
      "entity_type": "adjective",
    },
    {
      "start_offset": 43,
      "end": 47,
      "text": "Sun",
      "entity_type": "adjective",
    }
 ]

Answer 1

您可以预先排序数组，并通过查找较小级别的开始范围和结束范围来减少数组，直到找到子级别为止。

​

const
    data = [{ start_offset: 0, end: 38, text: "Asteroid is a rocky objects that orbit", entity_type: "adjective" }, { start_offset: 12, end: 19, text: "a rocky", entity_type: "adjective" }, { start_offset: 14, end: 27, text: "rocky objects", entity_type: "adjective" }, { start_offset: 20, end: 32, text: "objects that", entity_type: "adjective" }, { start_offset: 14, end: 19, text: "rocky", entity_type: "adjective" }, { start_offset: 20, end: 27, text: "objects", entity_type: "adjective" }, { start_offset: 33, end: 47, text: "orbit the Sun", entity_type: "adjective" }, { start_offset: 43, end: 47, text: "Sun", entity_type: "adjective" }],
    result = data
        .sort((a, b) => a.start_offset - b.start_offset || b.end - a.end)
        .reduce((r, { ...o }) => {
            let temp,
                child = { split: r };
            do {
                temp = child;
                temp.split = temp.split || [];
                child = temp.split.find(q => q.start_offset <= o.start_offset && q.end >= o.end);
            } while (child)
            temp.split.push(o);            
            return r;
        }, []);

console.log(result);

.as-console-wrapper { max-height: 100% !important; top: 0; }

​