JavaScript 如何从选定的文本中提取链接

Question

I looking-for solution how to extract url from selected text using window.getSelection and document.selection .

Text to select look that:

<p>Lorem Ipsum is <a href="http://example.com">simply <b>dummy</b> text</a> of the printing and typesetting industry.</p>

Selected text (selected by user) to extract link:

Option 1 (include text and text between a tags):

Ipsum is simply dummy text of

Option 2 (select text and fragment of link):

Ipsum is simply

The function should be return http://example.com

Answer 1

It's hard to write up for cross-browser function. See How to bind a handler to a selection change on window? .

We should capture some events like mousedown , mouseup , touchstart , touchend . The combination of these event may be fine.

function addEvent(elem, event, func) {
    if (elem.addEventListener) {
        elem.addEventListener(event, func, false);
    } else {
        elem.attachEvent('on' + event, func);
    }
}

Next is the getSelectedHTML() using window.getSelection or document.selection .

function getSelectedHTML(event) {
    // http://www.w3.org/TR/2000/REC-DOM-Level-2-Traversal-Range-20001113/ranges.html
    var range = window.getSelection ? window.getSelection()  /* W3C */
    : document.getSelection ? document.getSelection() /* redundant? */
    : document.selection ? document.selection.createRange()   /* IE */
    : null;

    if (range) {
        if (range.getRangeAt) {
            range = range.getRangeAt(0);
        } else if (range.setStart) { // Safari 1.3
            // Create range by itself
            range.setStart(range.anchorNode, range.anchorOffset);
            range.setEnd(range.focusNode, range.focusOffset);
        } else {
            // IE is already reange
        }

        var html = null, link = null;
        if (range.cloneContents) {
            var dummy = document.createElement('div');
            dummy.appendChild(range.cloneContents());
            html = dummy.innerHTML;
        } else {
            html = range.htmlText; /* IE */
        }

        if (html) {
            link = html.match(/<a.*?href\s*?=['"](.*?)['"]/);
            return link ? link[1] : null;
        }
    }

    return null;
}

This code should be checked especially with old browsers.

Here is the sample fiddle: http://jsfiddle.net/tokkonoPapa/CQ63a/

Answer 2

I'm sorry if I misunderstood the question, but here is my suggestion.

Have a function and pass it the selected HTML, and have it handle it like this.

function FindLink(str){
    // str='<p>Lorem Ipsum is <a href="http://example.com">simply <b>dummy</b> text</a> of the printing and typesetting industry.</p>';

    // we get the index of the beginning of the <a> tag
    var n = str.indexOf("<a") ;

    // we find the href property of the <a> element, starting from the <a we found
    var n2 = str.indexOf("href=\"", n) + 6;

    // we find the end of the href property
    n3 = str.indexOf("\">",n2);

    // we find the substring starting at the beginning of the link and going for however many characters the link is
    link = str.substr(n2,n3-n2);

    return link;
}

I hope that helped!