will scout support mapping?

[mehrdad-shokri]

What I want to do is to use language analyzers (https://www.elastic.co/guide/en/elasticsearch/guide/current/using-language-analyzers.html), but seems like scout doesn't support custom mapping between models and documents.
Is there any way I can accomplish this?

[Melcus]

I'm interested in this as well.

[toby1991]

as well too

[brino]

There is no reason you couldn't overwrite the mapping with a custom one, or index settings for that matter. You just post them with a REST client to the elastic cluster directly. Then, index your data. Granted you're still somewhat limited, but the point is you can quickly accomplish this outside of scout without too much trouble. You could also make a console command to do this.

[mehrdad-shokri]

Actually, you can create you custom mapping with a curl command in your cluster, the import your data from scout via scout:import laravel scout will respect your mapping, but you need to override toSearchableArray to match with your custom mapping.
An example would be:

curl -XPUT 'localhost:9200/yourApp' -d '
{
  "settings":{
    "analysis":{
      "char_filter":{
        "zero_width_spaces":{
          "type":"mapping",
          "mappings":[
            "\\u200C=> "
          ]
        }
      },
      "filter":{
        "persian_stop":{
          "type":"stop",
          "stopwords":"_persian_"
        },
        "autocomplete_filter":{
          "type":"edge_ngram",
          "min_gram":1,
          "max_gram":20
        }
      },
      "analyzer":{
        "autocomple":{
          "type":"custom",
          "tokenizer":"standard",
          "char_filter":[
            "zero_width_spaces"
          ],
          "filter":[
            "autocomplete_filter",
            "lowercase",
            "arabic_normalization",
            "persian_normalization",
            "persian_stop"
          ]
        },
        "persian":{
          "type":"custom",
          "tokenizer":"standard",
          "char_filter":[
            "zero_width_spaces"
          ],
          "filter":[
            "lowercase",
            "arabic_normalization",
            "persian_normalization",
            "persian_stop"
          ]
        }
      }
    }
  },
  "mappings":{
    "places":{
      "properties":{
        "name":{
          "type":"string",
          "analyzer":"persian"
        },
        "name_suggest":{
          "type":"string",
          "analyzer":"autocomple"
        }
      }
    }
  }
}'

And your toSearchableArray would be like:

public function toSearchableArray()
    {
        $array = array("name" => $this->name, "name_suggest" => $this->name, $this->getKeyName() => $this->getKey());
        return $array;
    }

If you get any more problems, just mention me.