How can I highlight elements that match a regex in CKeditor?

Question

I want to be able to highlight (i.e. wrap in a span with a color, or some other way) all text that matches a regex in CKEditor. I'd probably add a button to do this, and a button to remove highlighting. My specific use case is to highlight all mustache variables in my HTML templates (make it really easy to see where there are mustache variables).

I've implemented a version where I replace a regex matching mustaches with a span and then the capture group. This appears to break on some templates when I test.

To remove the highlighting, I use editor.removeStyle, which doesn't seem to work in all cases.

Here is an example of what I've implemented:

editor.addCommand( 'highlightMustache', {
            exec: function( editor ) {
                editor.focus();
                editor.document.$.execCommand( 'SelectAll', false, null );
                var mustacheRegex = /{{\s?([^}]*)\s?}}/g;
                var data = editor.getData().replace(mustacheRegex, '<span style="background-color: #FFFF00">{{ $1 }}</span>');
                editor.setData( data );
            }
        });

        // command to unhighlight mustache parameters
        editor.addCommand( 'unhighlightMustache', {
            exec: function( editor ) {
                editor.focus();
                editor.document.$.execCommand( 'SelectAll', false, null );
                var style = new CKEDITOR.style( { element:'span', styles: { 'background-color': '#FFFF00' },type:CKEDITOR.STYLE_INLINE,alwaysRemoveElement:1 } );
                editor.removeStyle( style );
                editor.getSelection().removeAllRanges();
            }
        });

Thanks!

Answer 1

The following approach worked for me in the past for a similar task:

1. Walk the DOM tree of the CKEditor document and combine all text nodes into a single string (let's call it S). Use CKEDITOR.dom.walker for that, this answer should help here. While walking the tree, build a collection of data structures (let's call it C) to store each text node object and the position of where its text starts within S.

2. Run your regex against S.

3. If the match is not found, stop.

4. Otherwise, using C collection, locate the start text node (let's call it SN), and offset within it, corresponding to the start character position of the match string inside S.

5. Using C collection, locate the end text node (let's call it EN), and offset within it, corresponding to the end character position of the match string inside S.

6. Create a CKEDITOR.dom.range object and position it using SN as the start and EN as the end (startContainer/startOffset/endContainer/endOffset).

7. Use CKEDITOR.dom.selection.selectRanges() to select the range from the previous step.