Elasticsearch-dsl-py: How to use multiple analyzers in a field

[shivam05011996]

I am not able to figure out how to 2 analyzers on a field. I have the following Index Class defined -

class ConnectorApiIndex(Document):
    title = Text(analyzer=[NGRAM_TOKEN_ANALYSER, STANDARD_TOKEN_ANALYZER], search_analyzer=STANDARD_TOKEN_ANALYZER)
    description = Text(analyzer=[NGRAM_TOKEN_ANALYSER, STANDARD_TOKEN_ANALYZER], search_analyzer=STANDARD_TOKEN_ANALYZER)
    url = Text(analyzer=[NGRAM_TOKEN_ANALYSER, STANDARD_TOKEN_ANALYZER], search_analyzer=STANDARD_TOKEN_ANALYZER)
    parent_connector_id = Text()
    owner_id = Text()
    class Index:
        name = settings.SERVER_NAME + '-connector-api-index'

When I check the mapping in the elasticsearch head, in the analysis key I can see two analyzer but on the field there is only one analyzer -

{
  "index": {
    "number_of_shards": "5",
    "provided_name": "pre-dev.kwikui.com-connector-api-index",
    "creation_date": "1552045118501",
    "analysis": {
      "analyzer": {
        "standard_token_analyzer": {
          "filter": [
            "standard",
            "lowercase"
          ],
          "type": "custom",
          "tokenizer": "standard"
        },
        "ngram_token_analyser": {
          "filter": [
            "standard",
            "lowercase"
          ],
          "type": "custom",
          "tokenizer": "gram"
        }
      },
      "tokenizer": {
        "gram": {
          "type": "ngram",
          "min_gram": "3",
          "max_gram": "16"
        }
      }
    },
    "number_of_replicas": "1",
    "uuid": "YA4hjx9vTIKNq5xUS_WCJQ",
    "version": {
      "created": "6060099"
    }
  }
}
{
  "mappings": {
    "doc": {
      "properties": {
        "owner_id": {
          "type": "text"
        },
        "description": {
          "search_analyzer": "standard_token_analyzer",
          "analyzer": "ngram_token_analyser",
          "type": "text"
        },
        "title": {
          "search_analyzer": "standard_token_analyzer",
          "analyzer": "ngram_token_analyser",
          "type": "text"
        },
        "parent_connector_id": {
          "type": "text"
        },
        "url": {
          "search_analyzer": "standard_token_analyzer",
          "analyzer": "ngram_token_analyser",
          "type": "text"
        }
      }
    }
  }
}

@HonzaKral I am not able find anything regarding this in the documentation too, its kind a important for me to figure this out.

[HonzaKral]

If you want to use multiple analyzers you need to define multiple fields:

class ConnectorApiIndex(Document):
    title = Text(analyzer=NGRAM_TOKEN_ANALYSER, search_analyzer=STANDARD_TOKEN_ANALYZER,  fields={'standard': Text(analyzer=STANDARD_TOKEN_ANALYZER)})

now you can use title and title.standard in your queries.

Hope this helps!