I can’t give you the code for the mobile phone answer, but the general idea should be:
1. Let’s sort out your needs first. If I understand it wrong, you don’t need to read the following:
You want to make the original one-dimensional , data without hierarchical structure is separated according to country code, province and city code, and region code. The key of the lower-level region is the key of the upper-level province and city. Through the code, you can directly go from country to province to city, level by level. Find a region.
2. First of all, your data is an object. Each attribute is the region code of the corresponding value, and the value is an array. The second element is the code of its parent node. This is easy to handle.
3. First create an empty object.
4. Loop through the original object.
5. Every time you get an attribute of the original object, first judge the second value of its value (that is, the binary array). If it is 1, it means it is a top-level region, that is, a province or a municipality. Write these to the new When creating the object, write it under the attribute 86 country code,
6. If the second value you get is not 1, it means that it is not a top-level province or city, but a certain region. Go to the new object to find out if there is such a If the province or region corresponding to the code does not exist, create it in the new object. If it exists, write it directly to the corresponding code.
7. In this way, it can be divided according to the country~province/municipality~region.
8. I see that you also specifically marked the municipality. According to the data you gave, if the name of the province and the name of the municipality are the same, it means it is Just the municipality is fine

9. When calling on a mobile phone, there will be many things that are difficult to understand. If the method is unclear or wrong, please ignore it.

A simple example is as follows (the source data format is not JSON, so eval is forced in the simple implementation):

const src = ` 110000: ['北京市', 1], 110100: ['北京市', 110000], 110101: ['东城区', 110100], 110102: ['西城区', 110100], 110105: ['朝阳区', 110100], 110106: ['丰台区', 110100], 110107: ['石景山区', 110100], 110108: ['海淀区', 110100], 110109: ['门头沟区', 110100], 110111: ['房山区', 110100], 110112: ['通州区', 110100], 110113: ['顺义区', 110100], 110114: ['昌平区', 110100], 110115: ['大兴区', 110100], 110116: ['怀柔区', 110100], 110117: ['平谷区', 110100], 110199: ['其它区', 110100], 110200: ['县', 110000], 110228: ['密云县', 110200], 110229: ['延庆县', 110200] ` function convert (src) { // hack 将结构字符串 eval 转为对象 const tmpObj = eval('(function () { return {' + src + '} })()') // 输出目标格式，硬编码一条简单数据 const result = { 86: {}, 110000: { 110100: '市辖区' }, 110100: {} } // [110000, 110100...] const tmpArr = Object.keys(tmpObj) tmpArr.forEach(key => { result[86][key] = tmpObj[key][0] }) tmpArr.forEach(key => { if (tmpObj[key][1] === 110100) { result[110100][key] = tmpObj[key][0] } }) return result } console.log(convert(src))

Execution result:

➜ Desktop node demo { '86': { '110000': '北京市', '110100': '北京市', '110101': '东城区', '110102': '西城区', '110105': '朝阳区', '110106': '丰台区', '110107': '石景山区', '110108': '海淀区', '110109': '门头沟区', '110111': '房山区', '110112': '通州区', '110113': '顺义区', '110114': '昌平区', '110115': '大兴区', '110116': '怀柔区', '110117': '平谷区', '110199': '其它区', '110200': '县', '110228': '密云县', '110229': '延庆县' }, '110000': { '110100': '市辖区' }, '110100': { '110101': '东城区', '110102': '西城区', '110105': '朝阳区', '110106': '丰台区', '110107': '石景山区', '110108': '海淀区', '110109': '门头沟区', '110111': '房山区', '110112': '通州区', '110113': '顺义区', '110114': '昌平区', '110115': '大兴区', '110116': '怀柔区', '110117': '平谷区', '110199': '其它区' } }

https://windqyoung.github.io/...

Convert at will.