multifield indexing?

example.sh

## Example from http://bit.ly/1m7DFKs
curl -XDELETE 'localhost:9200/my_index'
echo
curl -XPUT 'localhost:9200/my_index' -d'
{
    "settings": {
        "analysis": {
            "filter": {
                "trigrams_filter": {
                    "type":     "ngram",
                    "min_gram": 3,
                    "max_gram": 3
                }
            },
            "analyzer": {
                "trigrams": {
                    "type":      "custom",
                    "tokenizer": "standard",
                    "filter":   [
                        "lowercase",
                        "trigrams_filter"
                    ]
                }
            }
        }
    },
    "mappings": {
        "my_type": {
            "properties": {
                "text": {
                    "type":     "string",
                    "analyzer": "trigrams"
                }
            }
        }
    }
}'
echo
curl -XPOST 'localhost:9200/my_index/my_type/_bulk' -d'
{ "index": { "_id": 1 }}
{ "text": "Aussprachewörterbuch" }
{ "index": { "_id": 2 }}
{ "text": "Militärgeschichte" }
{ "index": { "_id": 3 }}
{ "text": "Weißkopfseeadler" }
{ "index": { "_id": 4 }}
{ "text": "Weltgesundheitsorganisation" }
{ "index": { "_id": 5 }}
{ "text": "Rindfleischetikettierungsüberwachungsaufgabenübertragungsgesetz" }'
echo
curl -XGET 'localhost:9200/my_index/my_type/_search' -d'
{
    "query": {
        "match": {
            "text": "Adler"
        }
    }
}'

multiindex.sh

curl -XDELETE 'localhost:9200/tc/'
echo "\n^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^Deleted "

curl -XPUT 'localhost:9200/tc/' -d'
{"settings":
    {
    "analysis": {
        "filter": {
            "trigrams_filter": {
                "type": "ngram",
                "min_gram": 3,
                "max_gram": 3
            }
        },
        "analyzer": {
            "trigrams": {
                "type": "custom",
                "tokenizer": "standard",
                "filter": [
                    "lowercase",
                    "trigrams_filter"
                ]
            }
        }
    },
    "mappings": {
        "charm": {
            "properties": {
                "name": {
                    "type": "string",
                    "analyzer": "trigrams"
                }
            }
        },
        "bundle": {
            "properties": {
                "name": {
                    "type": "string",
                    "analyzer": "trigrams"
                }
            }
        }
    }
}}'

echo "\n^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^Put mapping"

curl -XPUT 'localhost:9200/tc/charm/1' -d '{
    "name": "sugarcrm"
}'
echo
curl -XPUT 'localhost:9200/tc/charm/2' -d '{
    "name": "couchbase"
}'
echo
curl -XPUT 'localhost:9200/tc/charm/3' -d '{
    "name": "mysql"
}'
echo
curl -XPUT 'localhost:9200/tc/charm/4' -d '{
    "name": "cf-mysql"
}'
echo
curl -XPUT 'localhost:9200/tc/charm/5' -d '{
    "name": "PostgreSQL"
}'
echo
curl 'http://localhost:9200/tc/_search?q=name:crm'
echo
curl -XPOST 'localhost:9200/tc/_search' -d'
{
    "query" : {
        "match":{
         "name": "crm"
        }
    }
}'
echo

working-sense-example

# Delete the `charms` index
DELETE /charms

# Map the `text` field to be create ngrams of length 3
PUT /charms
{
  "settings": {
    "analysis": {
      "filter": {
        "3-20grams_filter": {
          "type": "ngram",
          "min_gram": 3,
          "max_gram": 20
        }
      },
      "analyzer": {
        "3-20grams": {
          "type": "custom",
          "tokenizer": "standard",
          "filter": [
            "lowercase",
            "3-20grams_filter"
          ]
        }
      }
    }
  },
  "mappings": {
    "charm": {
      "properties": {
        "name": {
          "type": "multi_field",
          "fields":{
            "ngrams": {
                "type": "string",
                "analyzer": "3-20grams"
                },
            "name":{
                "type": "string",
                "index": "not_analyzed"
            }
        }
      }
    }
  }
}
}

# Test the analyzer
GET /charms/_analyze?analyzer=n3_20grams&text=sugarcrm

# Index some example docs
POST /charms/charm/_bulk
{"index":{"_id":1}}
{"name":"sugarcrm"}
{"index":{"_id":2}}
{"name":"couchbase"}
{"index":{"_id":3}}
{"name":"mysql"}
{"index":{"_id":4}}
{"name":"cf-mysql"}
{"index":{"_id":5}}
{"name":"PostgreSQL"}


# Hit
GET /charms/charm/_search?q=ngrams:crm

# 3 Hits
GET /charms/charm/_search?q=ngrams:mysql

# Hit
GET /charms/charm/_search?q=name:sugarcrm

# Miss
GET /charms/charm/_search?q=name:crm

# 3 Hits
GET /charms/charm/_search?q=ngrams:sql

# Hit
GET /charms/charm/_search?q=ngrams:ouch

# Hit: Simple search - matches trigram
POST /charms/charm/_search
{
  "query": {
    "match": {
      "ngrams": "crm"
    }
  }
}

# Miss: Simple exact search - misses trigram
POST /charms/charm/_search
{
  "query": {
    "match": {
      "name": "crm"
    }
  }
}


# Hit: Simple search - matches 3
POST /charms/charm/_search
{
  "query": {
    "match": {
      "ngrams": {
        "query": "sql"
      }
    }
  }
}

# Miss: Correct Miss
POST /charms/charm/_search
{
  "query": {
    "match": {
      "name": {
        "query": "sql"
      }
    }
  }
}

# Have figured explain yet.
POST /charms/charm/_validate/query?explain
{
  "query": {
    "match": {
      "name": {
        "name": "crm"
      }
    }
  }
}

# Think we need longer names to demo trimming the long tail with minimum_should_match
POST /charms/charm/_search
{
  "query": {
    "match": {
      "ngrams": {
        "query": "ouc",
        "minimum_should_match": "80%"
      }
    }
  }
}