I have a lot of poorly engineered curl | grep | tr | awk pipes written into a loop in a bash script to scrape web pages and pull out text or data. I find

I have a lot of poorly engineered curl | grep | tr | awk pipes written into a loop in a bash script to scrape web pages and pull out text or data. I find if I'm just doing a one of, I can usually identify some feature of the html to pattern match one that lets me strip out the data in a very simple script and not have to use beautifulsoup etc.