Does not work with HTML

[hakre]

Does not work with HTML

[DaveRandom]

That's going to be difficult to sort out in a cross-browser manner, it's reasonably easy to look for adjacent text nodes but you'd probably need to ignore <div>the</div> cloud (block-level elements, in other words) and innerText/textContent are a) a pain in my butt (heh) and b) will screw up the HTML semantics if you modify them

[hakre]

Naturally the HTML semantics must be preserved, but it is not possible with split words like:

<i>th</i>e cloud

true. However just thinking that running an xpath query that returns all text-nodes (maybe even that contain (case-insensitively) the search words) might improve the code. This would spare to do the tree traversal with a recursive js function.

[DaveRandom]

XPath has more restricted browser support, but that is still probably a feasible solution. I agreed the complete document tree traversal is dangerous, and could affect the UX.

[panicsteve]

I'm not personally going to go to that much effort for a butt joke, but I encourage any and all forks.

[DaveRandom]

@panicsteve You clearly underestimate just how far we will go to avoid doing real work.

[moeffju]

I fixed this in my fork at https://github.com/moeffju/cloud-to-butt, which I’m currently extending for multi-language support.