Dan Robinson, Web Developer

Most modern content management systems claim they will strip away unwanted formatting when you copy-and-paste content from external sources such as Microsoft Word documents and PDFs. But do CMS’s always deliver on that promise? No, they do not.

HTML Factory is a tool I designed for myself and my colleagues at the CU School of Medicine. Maybe you’ll find it useful, too! When your CMS’s built-in “remove formatting” feature falls short—either by leaving in stuff you want taken out, or by taking out formatting you want left in—this tool comes to the rescue, giving you fine-grained control over the process of cleaning up HTML.

HTML Factory allows you to:

• remove inline styles
• remove classes
• replace non-breaking spaces with regular spaces
• remove empty paragraphs
• replace one kind of element with another (e.g., convert bullet points to paragraphs, or vice versa)
• remove selected attributes from all elements
• “unwrap” selected elements (e.g., get rid of all <span> elements inside divs or paragraphs but leave their contents intact)
• add as many custom find-and-replace operations as desired (supports regular expressions)

Try it for yourself:
http://danrobinson.xyz/web-dev-tools/html-factory.html

In 2019, the University of Colorado decided to move its entire website into a new content management system. The old CMS, SharePoint 2010, was fearfully out of date. Unfortunately, there was no tool that could automatically migrate website content to the new CMS (Sitefinity), so the entire website would have to be rebuilt page by page.

At the time, I was working as the website manager for the Department of Surgery at the university’s medical campus in Denver. Through a friend, I heard about a problem that the School of Medicine’s IT leaders were grappling with. The central website team for the School would soon be rebuilding tens of thousands of web pages in the new CMS, and they needed to charge individual departments for that service. Unfortunately, estimating the scope of the work for each department was difficult.

Incredibly, SharePoint 2010 offered no way to count the pages in a particular department’s website or even display a unified site map. The only way to get a page count was by clicking through all of a given site’s “sub-sites” and manually counting them. The site structures were often labyrinthine and involved hundreds of sub-sites.

Clearly, a better solution was needed.

I didn’t have access to the back end of our SharePoint site, but I did have the ability to build HTML pages on the same domain, and those pages could run JavaScript. That gave me an idea. I built a script that opened SharePoint’s admin interface in an iframe, simulated click events, and systematically navigated through every subsite. The end result was a page count and a site map that could be clicked through, opening and closing branches as desired.

Not coincidentally, I was hired that month to join the School of Medicine’s central website team.