Wilcard search or partial matching in Elastic search

Question

I am trying to provide the search to end user with type as they go which is is more like sqlserver. I was able to implement ES query for the given sql scenario:

Answer 1

The most efficient solution involves leveraging an ngram tokenizer in order to tokenize portions of your name field. For instance, if you have a name like peter tomson, the ngram tokenizer will tokenize and index it like this:

• pe
• pet
• pete
• peter
• peter t
• peter to
• peter tom
• peter toms
• peter tomso
• eter tomson
• ter tomson
• er tomson
• r tomson
• tomson
• tomson
• omson
• mson
• son
• on

So, when this has been indexed, searching for any of those tokens will retrieve your document with peter thomson in it.

Let's create the index:

PUT likequery
{
  "settings": {
    "analysis": {
      "analyzer": {
        "my_ngram_analyzer": {
          "tokenizer": "my_ngram_tokenizer"
        }
      },
      "tokenizer": {
        "my_ngram_tokenizer": {
          "type": "nGram",
          "min_gram": "2",
          "max_gram": "15"
        }
      }
    }
  },
  "mappings": {
    "typename": {
      "properties": {
        "name": {
          "type": "string",
          "fields": {
            "search": {
              "type": "string",
              "analyzer": "my_ngram_analyzer"
            }
          }
        },
        "type": {
          "type": "string",
          "index": "not_analyzed"
        }
      }
    }
  }
}

You'll then be able to search like this with a simple and very efficient term query:

POST likequery/_search
{
  "query": {
    "bool": {
      "should": [
        {
          "term": {
            "name.search": "peter tom"
          }
        }
      ],
      "must_not": [
        {
          "match": {
            "type": "xyz"
          }
        },
        {
          "match": {
            "type": "abc"
          }
        }
      ]
    }
  }
}

Answer 2

Well my solution is not perfect and I am not sure about performance. So you should try it on your own risk :)

This is es 5 version

PUT likequery
{
  "mappings": {
    "typename": {
      "properties": {
        "name": {
          "type": "string",
          "fields": {
            "raw": {
              "type": "keyword"
            }
          }
        },
        "type": {
          "type": "string"
        }
      }
    }
  }
}

in ES 2.1 change "type": "keyword" to "type": "string", "index": "not_analyzed"

PUT likequery/typename/1
{
  "name": "peter tomson"
}

PUT likequery/typename/2
{
  "name": "igor tkachenko"
}

PUT likequery/typename/3
{
  "name": "taras shevchenko"
}

Query is case sensetive

POST likequery/_search
{
  "query": {
    "regexp": {
      "name.raw": ".*taras shev.*"
    }
  }
}

Response

{
  "took": 5,
  "timed_out": false,
  "_shards": {
    "total": 5,
    "successful": 5,
    "failed": 0
  },
  "hits": {
    "total": 1,
    "max_score": 1,
    "hits": [
      {
        "_index": "likequery",
        "_type": "typename",
        "_id": "3",
        "_score": 1,
        "fields": {
          "raw": [
            "taras shevchenko"
          ]
        }
      }
    ]
  }
}

PS. Once again I am not sure about performance of this query since it will use scan and not index.