HTML Cleaner

Text copied from Word or Google Docs brings its service markup with it: tags like <o:p>, MsoNormal classes, inline styles on every paragraph, conditional comments. None of this improves the layout, while the page weight grows three to four times.

Lists are a separate problem. Word does not output them as lists but as a chain of paragraphs, and it puts the bullet character right into the text. On a site, such a list loses both indentation and numbering, and a dot stays hanging before each item.

What gets removed

• Word service markup: tags with a colon, o: and v: attributes, conditional comments, bullet characters;
• inline styles, classes, identifiers, data attributes;
• empty tags, including those containing only a non-breaking space;
• obsolete tags: <font>, <center>, <strike> along with presentational attributes like align and bgcolor;
• links, images, and tables when you only need the text.

Paragraph items left over from Word are assembled back into <ul> or <ol>; the list type is determined by the bullet character.

Tags <script>, onclick handlers, and javascript: links are removed regardless of the checkboxes. People usually clean someone else's code before putting it on their own site, and the active part of it is not needed there.