将HTML下载到允许的标签

Question

I'd like to strip all HTML tags (but not the content), apart from ones given in a list.

I'd like to do this using Node.

The following regex can match tags <[a|br].+?>|<\\/[a]> but how can I then proceed to remove all tags except the ones matched?

Answer 1

To replace only a and br :

1. There is no | in sets [] . The set [a|br] is: a , | , b or r . Use a non-capturing group instead.
2. .+? should be .*? to replace <a> and <br> .
3. The set [a] could be just a .

Try this:

/<(?:a|br).*?>|<\/a>/g

https://regex101.com/r/KWJi01/2

To replace every tag except a and br :

Use this regexp:

/<(?:(?!\/?a|br).*?)>/g

https://regex101.com/r/KWJi01/3

Answer 2

One alternative solution to the problem, without using regular expressions, is the following approach:

// creating a function which takes two arguments:
// htmlString: String, a string of HTML to process,
// permitted:  Array, an array of HTML element tag-names.
function filterHTML(htmlString, permitted) {

  // here we iterate over the Array of permitted elements,
  // and convert any uppercase tag-names to lowercase:
  permitted = permitted.map(
    el => el.toLowerCase()
  );

  // creating a new element in which to hold the passed-in
  // string of HTML when parsed into HTML:
  let temp = document.createElement('div');

  // assigning the passed-in string of HTML as the
  // innerHTML of the temp element:
  temp.innerHTML = htmlString;

  // finding all elements held within the temp element,
  // and passing the NodeList to Array.from() in order
  // to convert the Array-like NodeList into an Array:
  let allElements = Array.from(temp.querySelectorAll('*'));

  // iterating over the array of found elements, using
  // Array.prototype.forEach():
  allElements.forEach(

    // using an Arrow function, 'element' is the current
    // element of the Array of elements over which we're
    // iterating:
    element => {

      // if the current element's tagName - converted to
      // lowercase - is not found within the Array of
      // permitted tags:
      if (permitted.indexOf(element.tagName.toLowerCase()) === -1) {

        // while the current (unpermitted) element has
        // a first-child:
        while (element.firstChild) {

          // we access the element's parent node, and
          // call node.insertBefore() to place the
          // first-child of the element before the
          // element (removing it from the element
          // which is unpermitted and placing it as
          // a previous-sibling):
          element.parentNode.insertBefore(element.firstChild, element);
        }

        // finding the element's parent node, and calling
        // node.removeChild() in order to remove the current
        // element from its parent node (and therefore from
        // the temp element):
        element.parentNode.removeChild(element);
      }
    });

  // here we return the innerHTML of the temp element, after
  // all unpermitted elements have been removed:
  return temp.innerHTML;
}

// the allowed tags, to be used in the above function,
// note that the tags do not have the '<', '>' or any
// attributes:
let allowedTags = ['br', 'div'];

// the following is just to visually display on the page
// the unprocessed and processed innerHTML, and also the
// rendered HTML following the processing:
document.querySelector('#input').textContent = document.querySelector('#test').innerHTML;

document.querySelector('#output').textContent = filterHTML(document.querySelector('#test').innerHTML, allowedTags).trim();

document.querySelector('#result').innerHTML = document.querySelector('#output').textContent;

 function filterHTML(htmlString, permitted) { let temp = document.createElement('div'); temp.innerHTML = htmlString; let allElements = Array.from(temp.querySelectorAll('*')); allElements.forEach( element => { if (permitted.indexOf(element.tagName.toLowerCase()) === -1) { while (element.firstChild) { element.parentNode.insertBefore(element.firstChild, element); } element.parentNode.removeChild(element); } }); return temp.innerHTML; } let allowedTags = ['br', 'div']; document.querySelector('#input').textContent = document.querySelector('#test').innerHTML; document.querySelector('#output').textContent = filterHTML(document.querySelector('#test').innerHTML, allowedTags).trim(); document.querySelector('#result').innerHTML = document.querySelector('#output').textContent; 

 div { border: 2px solid #000; margin: 0 0 1em 0; padding: 0.5em; box-sizing: border-box; border-radius: 1em; } div[id$=put] { white-space: pre-wrap; } #input { border-color: #f00; } #output { border-color: limegreen; } ::before { color: #666; display: block; border-bottom: 1px solid #666; margin-bottom: 0.5em; } #input::before { content: 'The original innerHTML of the "#test" element:'; } #output::before { content: 'The filtered innerHTML, with only permitted elements remaining:' } #result::before { content: 'This element shows the filtered HTML of the "test" element:' } 

 <div id="test"> <div><span> <a href="#">link text!</a> <br /> <hr /> <div>div text!</div> </span></div> </div> <div id="input"></div> <div id="output"></div> <div id="result"></div> 

JS Fiddle demo .

References:

• Array.from() .
• Array.prototype.forEach() .
• Array.prototype.indexOf() .
• Array.prototype.map() .
• Arrow functions .
• document.createElement() .
• document.querySelector() .
• document.querySelectorAll() .
• let statement .
• Node.firstChild .
• Node.insertBefore() .
• Node.parentNode .
• Node.removeChild() .
• String.prototype.toLowerCase() .
• while () {...} .