Let’s say you have events in an Algolia index. These events have several timestamp attributes (
If you configure these attributes to be a facet, is there an impact on Algolia indexation performance?
It sound natural to say “yes”, because there is an infinite set of values in such attributes. But I would like to know of there is any kind of documentation about the “nature” of attributes that should/shouldn’t be facet?
Adding dates as faceted attributes absolutely makes sense – this allows you to do things like date range filtering with your records. I would say the one caveat is to avoid making any facet with a large number of values (like a timestamp) searchable. It really wouldn’t be very useful to search within or attempt to display all of the individual values in a widget. Instead you’d want to use a date picker.
Just make sure you are saving the dates as unix timestamps .
Here’s a great example of using a date picker to filter records with unix timestamps: