Built Chrome extension that records all user actions in plain-English text and can replay such actions at a later time. Need your feedback!
Written in js + Chrome extension APIs
Most interesting challenges: 1. Deriving a plain-English description of each target element, including from text nearby in the DOM 2. Generating concise, minimal CSS selectors for each target element 3. Emulating events during playback (sandbox) 4. If during playback an element doesn't exist, finding the element that most closely matches the text, selector, and nearby text (what the user originally interacted with). Works great for elements with dynamic ids.
Please let me know what you all think, or if you have any questions!
Submitted July 14, 2017 at 11:22PM by vptes1
via reddit http://ift.tt/2urxoaq