在Elasticsearch中使用3的edge-ngram时如何匹配1-2个字符的完全匹配

Question

I am using elasticsearch to query an index with fuzzy matches. I am using an edge-ngram tokenizer with a min_gram length of 3.

However, this returns nothing for queries that only include 1 or 2 characters. Is it possible to match on exact matches for just those 1 or 2 characters, but use the edge-ngram for queries with three characters or more?

This is my current elasticsearch index mappings:

curl -XPUT 'http://localhost:9200/person' -d '{
"settings": {
    "number_of_shards": 1,
    "analysis": {
        "filter": {
            "autocomplete_filter": {
                "type":     "edge_ngram",
                "min_gram": 3,
                "max_gram": 20
            }
        },
        "analyzer": {
            "default": {
                "type":      "custom",
                "tokenizer": "standard",
                "filter": [
                    "lowercase",
                    "autocomplete_filter"
                ]
            }
        }
    }
}
}'

To query this index, a request like:

curl -XPOST 'localhost:9200/person/type/_search' -d '{
    "query": {
        "match": {
            "_all": "Tim”
        }
    }
}'

Yields plenty of results, however a request like

curl -XPOST 'localhost:9200/person/type/_search' -d '{
    "query": {
        "match": {
            "_all": "Ti”
        }
    }
}'

gives an empty set. Ideally, the second request would return some result if there is someone in the index named Tim.

Answer 1

I am not sure If this will satisfy all your requirements. You could check if the length of user input is less than 3 then fire the below query.

{
  "query": {
    "match_phrase_prefix": {
      "_all": "ti"
    }
  }
}