Question

我正在尝试解析XML文档并使用数据来构建这种形式的（简单）json对象：

{id: '1', name: 'content-types', children: [{id: '2', name: 'requirements': children: [... and so on ...]]}

我的XML具有如下所示的节点（我仅包括其中一个-它们可以任意嵌套）：

<?xml version="1.0" encoding="utf-8"?>
<html xmlns="http://www.w3.org/1999/xhtml">
  <head/>
  <body class="taxonomies">
    <div class="taxonomy">
      <span class="id">3484069771</span>
      <span class="name">Content Types</span>
      <span class="locale">en</span>
      <div class="concepts">
        <div class="concept">
          <span class="id">3484058507</span>
          <span class="name">Promotional Publications</span>
          <div class="concepts">
            <div class="concept">
              <span class="id">3551765771</span>
              <span class="name">Datasheets</span>
            </div>
          </div>
        </div>
      </div>
    </div>
  </body>
</html>

我使用以下代码从XML构建JSON树：

buildConceptTree: function(xml){
    const doc = new dom().parseFromString(xml)
    var tree = []
    var selector = "//*[@class='taxonomies']"
    var count = 0  // this should keep track of the depth of the node being used
    function recurse(s, odd){
        var nodes
        console.log(count)
        console.log(s)
        var arr = []

        nodes = xpath.select(s, doc)
        nodes.forEach(node => {
            try {
                var children = node.childNodes
                var keys = Object.keys(children).filter(x => {return Number(x)})
                keys.forEach(key => {
                    var child = children[key]
                    console.log('child is: ')
                    console.log(child)
                    var obj = {}
                    var grandchildren = child.childNodes
                    var grandkeys = Object.keys(grandchildren).filter(x => {return Number(x)})

                    grandkeys.forEach(gk => {
                        var gc = grandchildren[gk]
                        try {
                            var nodevalue = gc['attributes'][0]['nodeValue']
                            switch(nodevalue){
                            case 'id':
                                obj['id'] = gc['textContent']
                            case 'name':
                                obj['name'] = gc['textContent']
                            case 'concepts':
                                count++
                                var rx = /taxonomy/
                                    if(!rx.test(s)){
                                        s = s+"/*[@class='taxonomy']"
                                    }
                                else{
                                    s = s
                                }
                                if (!odd){
                                    s += "/*[@class='concepts']"
                                }
                                else {
                                    s += "/*[@class='concept']"
                                }
                                odd = !odd
                                obj['children'] = recurse(s, odd)
                            }
                        }
                        catch(e){
                        }
                    })
                    arr.push(obj)
                })
            }
            catch(e){
            }

        })
        return arr


    }

    var tree = recurse(selector, false)
    return tree

},

就目前而言，此函数产生的内容类似于我提到的JSON形式，但是缺少许多节点。

此外，似乎我的递归函数并没有在最简单的情况下终止，因为它沿着xml树的更深的分支递归。我在控制台中记录了以下内容（例如），但是没有深度为191度的节点：

    191
     parser.js?d3c4:83 //*[@class='taxonomies']/*[@class='taxonomy']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']/*[@class='concept']/*[@class='concepts']
     parser.js?d3c4:92 child is:

有人可以帮我弄清楚如何更改此功能以使其获取我想要的数据吗？

Answer 1

我可能已经错过了一些要求，但是一旦停止循环浏览所有元素并开始查询所需的确切元素，问题就会变得不那么复杂了

// Parse the xml string to a document
const parser = new DOMParser();
const xmlDoc = parser.parseFromString(
  getXML(),
  "text/xml"
);

// The main logic to go from an xml element to an object
const parseTaxonomy = (taxonomy, id = 1) => ({
  id,
  name: taxonomy.querySelector(".name")
    .innerText
    .toLowerCase()
    .replace(/\s/g, "-"),
  children: Array.from(
    (taxonomy.querySelector(".concepts") || { children: [] })
      .children
  ).map(t => parseTaxonomy(t, ++id)) // Note the ++
});

// Run on the first taxonomy
// If the top level contains multiple elements, use .map
console.log(
  parseTaxonomy(
    xmlDoc.querySelector(".taxonomy")
  )
);

// The data
function getXML() {
  return `<?xml version="1.0" encoding="utf-8"?>
<html xmlns="http://www.w3.org/1999/xhtml">
  <head/>
  <body class="taxonomies">
    <div class="taxonomy">
      <span class="id">3484069771</span>
      <span class="name">Content Types</span>
      <span class="locale">en</span>
      <div class="concepts">
        <div class="concept">
          <span class="id">3484058507</span>
          <span class="name">Promotional Publications</span>
          <div class="concepts">
            <div class="concept">
              <span class="id">3551765771</span>
              <span class="name">Datasheets</span>
            </div>
          </div>
        </div>
      </div>
    </div>
  </body>
</html>`;
};

注意：我更改了您放置评论的部分，因为该评论尚未关闭，并且我希望它可以在子分类法周围使用另一个包装。

递归xml解析功能未按预期工作

1 个答案: