/**
* Finds current page type (categories, sections or articles)
* and corresponding id 
*
* @return {Object} Returns object with 'type' and 'id' properties
*/
function getPageInfo () {
  var _t = window.location.href.split('/hc/');
  _t = _t[1].split('-')[1].split('/');
  return {type:_t[1], id:_t[2]}
}
//USAGE:
//var page = getPageInfo();
//var pageType = page.type;
//var pageId = page.id;
//console.log(page.type);
//console.log(page.id)

• Jennifer Rowe

  • March 30, 2016 17:23

  Zendesk Documentation Team

  Thanks for sharing this, Diziana!

• Rob Levin

  • May 04, 2016 17:32

  I'd be really curious if this is the "best way" to do this. It would be preferable for zendesk to expose this from a helper. Obviously, if they change their URL format/positioning, this whole thing breaks. So a year or two from now you find yourself with code that's breaking. I mean, it's easy enough to probably debug that, but might leave your page in a crap state until someone figures out things are broken.

  All said, thanks for sharing the snippet Diziana.

  Hopefully a Zendesk employee can chime in on my question.

• Rob Levin

  • May 04, 2016 17:46
  • Edited

  Also, if it's the "home page", you won't get the "type" in that array, so I had to hack this in.

  What's even more troubling is this won't likely work on some other page type I'm not yet aware of being somewhat new to Zendesk (e.x. I already see that the "contributions" template will falsely put type of 'home' the way I've coded this).

  I've submitted a ticket and will post back if I find a better solution from the Zendesk folks.

  function getPageInfo () {
  var _t = window.location.href.split('/hc/');
  _t = _t[1].split('-')[1].split('/');
  if (_t.length < 2) {
  _t[1] = 'home'; //Hack: fudge in home page
  }
  return {type:_t[1], id:_t[2]}
  }

   

• Rob Levin

  • May 04, 2016 22:59

  Ok, ZD got back and it doesn't appear that there's an available global/advanced helper for this so that hack is probably only way for page type. That said, it's probably better to use something like article.id, section.id, if you KNOW you're on a corresponding page. In my case, I only really care about the page type so hack stays ;)

• Jennifer Rowe

  • May 04, 2016 23:02

  Zendesk Documentation Team

  Thanks for the update, Rob!

• Rob Levin

  • May 04, 2016 23:33
  • Edited

  Yeah, one last thing I got back from ZD which is at least worth considering:

  "For each Template type, you modify the code to suit that page. So, for example, on the Home page, you use:

  function getPageType() { 
  return "home"; 
  } 

   

  So you do that for category, section, etc., and then you're guaranteed it's correct without URL sniffing. But, you have more to maintain and have to include quite a bit of JS in your templates. So it's sort of "six and half dozen one or  the other"; maintainence vs. future proofing. That said, probably worth having here for folks wanting to consider options...I'm going to stick with hack #1 URL Sniff. Of course support was quick to put disclaimers on all of these hacks as is not surprising.

  End of day, my thought is that this would be a great thing for ZD to expose via a helper - jm2c :)

• Iggy Pritzker

  • June 12, 2016 21:06

  can't zendesk just add a class to page template body tag?

• Arnaud Heritier

  • August 09, 2019 20:07

  It's very sad to see how the template mechanism is poor.

  Almost everything should rely on "hacks", especially in javascript if you really want some dynamic content

• Nicole S.

  • August 09, 2019 20:39

  Zendesk Community Team

   Hey Arnaud - 

  I'd encourage you to share your feedback with our product managers by posting it in the Guide Product Feedback topic. Here's info on how to write an effective feedback post. 

• David Bjorgen

  • September 06, 2019 13:43

  Rob, If I wanted the section info, how might I go about it using this code? Specifically, I need to determine whether the article exists within a specific section.

Please sign in to leave a comment.