Can prefix filtering also be used on string arrays?

Yup it works

It seems there is an issue when the string contains a "/"

I suspect we drop the / because it's not indexed. You have to add it to symbols_to_index

would it requires reindexing all documents after such a change?

Yes you can't do it with alter command.

Ok thanks a lot will give a try

Yeah the count is an approximate now

Ok after deeper tests it is not working as expected. ["alimentation/autres-produits/miel/miel-de-foret"] slugPath:[alimentation/autres-produits/miel/miel-de-foret*] OK slugPath:[alimentation/autres-produits/miel*] OK slugPath:[alimentation/autres-produits/m*] OK slugPath:[alimentation/autres-produits/*] KO slugPath:[alimentation/autres-produits*] KO Here is my field config

{
  "facet": true,
  "index": true,
  "infix": false,
  "locale": "",
  "name": "categories[].categories_id.slugPath",
  "optional": true,
  "sort": false,
  "stem": false,
  "stem_dictionary": "",
  "store": true,
  "symbols_to_index": [
    "/"
  ],
  "type": "string[]"
}

Can you please create a github issue with a small reproducible example in this format: https://gist.github.com/jasonbosco/7c3432713216c378472f13e72246f46b

The script in the issue wasn't working, so I've fixed some things in order for it to work (bad JSON in the collection schema, non-existent collection to index documents on). Apart from that, the document indexed is being found by the filter by clause. Could you explain which documents would not match the filter?

Thank you giving a look at my test case. After further investigation, I realized that the tests were performed against incorrectly indexed documents. I linked my final shell script testing multiple cases, all good. I have updated and closed the issue.