Copy the contents of semanticText.js and paste into a browser console. Run the code.

Finds all text nodes that are not purely white space and evaluates them against their parent node's node name. In the case where a parent node is a span, the span's parent node is evaluated. If that node is a span, its parent node is evaluated.

• A first list provides text content that is poorly described by its HTML parent node.
• A second list provides text content that is well described by an inline element but poorly described there after.
• A third list provides all text content in the order in which it occurs in the DOM so that text can read according to its order in the code.

This tool uses and includes getNodesByType to walk the DOM

MIT License