Skip to content

Instantly share code, notes, and snippets.

@iimos
Last active June 24, 2024 17:05
Show Gist options
  • Save iimos/e9e96f036a3c174d0bf4 to your computer and use it in GitHub Desktop.
Save iimos/e9e96f036a3c174d0bf4 to your computer and use it in GitHub Desktop.
Micro function that gives xpath by element and elements by xpath.
function xpath(el) {
if (typeof el == "string") return document.evaluate(el, document, null, 0, null)
if (!el || el.nodeType != 1) return ''
if (el.id) return "//*[@id='" + el.id + "']"
var sames = [].filter.call(el.parentNode.children, function (x) { return x.tagName == el.tagName })
return xpath(el.parentNode) + '/' + el.tagName.toLowerCase() + (sames.length > 1 ? '['+([].indexOf.call(sames, el)+1)+']' : '')
}
// Usage:
// Getting xpath for node:
var xp = xpath(elementNode)
// Executing xpath:
var iterator = xpath("//h2")
var el = iterator.iterateNext();
while (el) {
// work with element...
el = iterator.iterateNext();
}
@adelriosantiago
Copy link

adelriosantiago commented Jan 14, 2020

Why is the iterator needed? Shouldn't the function always lead to a single element?

BTW, thanks for this amazing piece of cake! It is working good so far.

@brettsprads
Copy link

Works great! Thank you so much

ps: as xpath isn't truly unique, got any thoughts on making it unique?
I'm thinking something like... combining xpath + href + css selectors to make a unique generated ID etc.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment