TinyMCE 5 not cleaning attributes from tags

Question

TinyMCE 5 just won't clean all the attributes from tags. It leaves stuff like "data-gps-track, data-controller, aria-label:"" ". It cleans stuff like "href, style, class". Any idea what's the problem here?

tinymce.init({
        selector: "#mytextarea",
        plugins: "paste, code",
        toolbar: "edit, code",
        paste_block_drop: false,
        paste_tab_spaces: 2,
        paste_as_text: cleantext,
        valid_elements: "a,code,p",
      });

You can try pasting stuff off StackOverflow, where these attributes are present.

Answer 1

I have ended up using an answer from this StackOverflow answer. I have also added a filter for clearing aria attributes to the regex.

var stripDataAttributes = function (html) {
    var tags = html.match(/(<\/?[\S][^>]*>)/gi);
    tags.forEach(function (tag) {
      html = html.replace(
        tag,
        tag.replace(
          /(data-.+?=".*?")|(data-.+?='.*?')|(data-[a-zA-Z0-9-]+)|(aria-.+?=".*?")|(aria-.+?='.*?')|(aria-[a-zA-Z0-9-]+)/g,
          ""
        )
      );
    });

    return html;
  };

tinymce.init({
    selector: "#mytextarea",
    plugins: "paste, code",
    toolbar: "edit, code",
    paste_block_drop: false,
    paste_tab_spaces: 2,
    paste_as_text: cleantext,

    valid_styles: {
      "*": "",
      p: "color",
    },
    // https://www.tiny.cloud/docs-4x/configure/content-filtering/

    // valid_elements: "*[*|-href]",
    // https://www.tiny.cloud/docs-3x/reference/Configuration3x/Configuration3x@valid_elements/
    // jen pozor na mezery (když jsem měl  valid_elements: "a, code, p" , tak mi to nešlo!)
    paste_preprocess: function (plugin, args) {
      args.content = stripDataAttributes(args.content);
      //   args.content = args.content.replace(cleanElements, "$1>");
    },
  });