scrivito/src/scrivito_sdk/app_support/extract_text/node_to_text.ts

Scrivito is a professional, yet easy to use SaaS Enterprise Content Management Service, built for digital agencies and medium to large businesses. It is completely maintenance-free, cost-effective, and has unprecedented performance and security.

import { childNodeListToArray } from './child_node_list_to_array';

const IGNORE_NODE = ['HEAD', 'SCRIPT'];

export function nodeToText(node: Node): string {
  const nodeName = node.nodeName;
  if (IGNORE_NODE.indexOf(nodeName) > -1) return '';
  if (nodeName === '#text') return node.textContent || '';

  return childNodeListToArray(node.childNodes)
    .map((child) =>
      INLINE_NODES.includes(child.nodeName)
        ? nodeToText(child)
        : ` ${nodeToText(child)} `
    )
    .join('');
}

/** @see htmlToTextForNode for tags we consider inline */
const INLINE_NODES = ['#text', 'A'];