Can I parse text in pdf document before sending it to elasticsearch using FSCrawler

GROK is based on regex. May be this could help you: Custom pattern - Telephone number and others