Remove 'empty' variables from url after search/filter

[PieterL75]

NetBox version

v3.3.2

Feature type

Change to existing functionality

Proposed functionality

When a search or filter is done, the resulting page has an url with all possible fields.
It would be neater if only the field that has data is added to the url variable.
Especially when there are a lot of custom fields.

ex :
https://netbox/dcim/devices/?q=test&serial=&asset_tag=&mac_address=&device_type_id=79&console_ports=&console_server_ports=&power_ports=&power_outlets=&interfaces=&pass_through_ports=&has_primary_ip=&virtual_chassis_member=&local_context_data=&cf_hostname=&cf_ltag=&cf_observium=&cf_observium_id=&cf_reference=&cf_u_bios_uuid=&cf_warrantydate=

could be reduced to

https://netbox/dcim/devices/?q=test&device_type_id=79

Use case

When URLs are shared or stored, it gives a better view in what is going to be filtered.

Database changes

No response

External dependencies

No response

[jeremystretch]

This is an artifact of how web browsers process forms with GET actions. I'm not aware of anything we can do to suppress the extraneous parameters natively; it would require extending the submission process with a Javascript function that determines which fields to append to the URL string based on their values.

[PieterL75]

Solution is to add this to the filter_list.html

<script>
function cleangeturl(form) {
  
  var form_elements = form.elements;
  
  for (var i=0; i<form_elements.length; i++ ) {
    var element = form_elements[i]
    if (element.name && element.value == '') {
      element.disabled=true;
    }
  }
</script>

and update the <form with

<form action="." method="get" onsubmit="return cleangeturl(this)">

[jeremystretch]

@PieterL75 I agree with the general approach, however our intent is to build all Javascript from Typescript compiled into central libraries, rather than embedding JS directly within each page.