So, I have to determine page type in a proprietary CRM, using JavaScript. The only way to determine the page type (that is, the only consistent difference on the front end)
Well, you can start by collecting all of the script elements. With jQuery, that's as simple as
var scripts = $("script");
Then limit that set to the elements that have a src
attribute:
var scripts = $("script[src]");
...and further limit it to those with a src
attribute beginning with "/modules/":
var scripts = $("script[src^='/modules/']");
...which given your description should result in a set of exactly one element, from which you can now pull the src
attribute value itself:
var path = $("script[src^='/modules/']").attr('src');
Ok, that was easy - now to extract the next part of the path. There are plenty of ways to do this, but split is quick & dumb: create an array of parts using '/' as the separator, then pick off the third element (which will be the one after "modules"):
var pathPart = $("script[src^='/modules/']").attr('src').split('/')[2];
Obviously, this is all very specific to the exact format of the script path you're using as an example, but it should give you a good idea of how to begin...