ElasticSearch Exact Name matching

elastic-search-name-matching.sh

export http_proxy=
export https_proxy=

curl -XDELETE 'http://localhost:9200/test/'

echo "Creating the mapping"
curl -XPUT 'http://localhost:9200/test/?pretty=1'  -d '
{
   "mappings" : {
      "member" : {
         "properties" : {
            "object_name" : {
               "fields" : {
                  "unmunged" : {
                     "search_analyzer" : "name_unmunged",
                     "index_analyzer" : "name_unmunged",
                     "type" : "string"
                  },
                  "standard" : {
                     "search_analyzer" : "standard",
                     "index_analyzer" : "standard",
                     "type" : "string"
                  }
               },
               "type" : "multi_field"
            },
            "object_name_other_language" : {
               "fields" : {
                  "unmunged" : {
                     "search_analyzer" : "name_unmunged",
                     "index_analyzer" : "name_unmunged",
                     "type" : "string"
                  },
                  "standard" : {
                     "search_analyzer" : "standard",
                     "index_analyzer" : "standard",
                     "type" : "string"
                  }
               },
               "type" : "multi_field"
            }
         }
      }
   },
   "settings" : {
      "analysis" : {
         "analyzer" : {
            "name_unmunged" : {
               "filter" : [
                  "lowercase"
               ],
               "type" : "custom",
               "tokenizer" : "standard"
            }
         }
      },
      "index": {
          "number_of_shards": 1
      }
   }
}
'; echo

echo "Indexing"
curl -XPOST 'http://localhost:9200/_bulk?pretty=1'  -d '
{"index" : {"_index" : "test", "_type" : "member"}}
{"object_name" : "Albino Elephant", "COMMENT_object_name_other_language" : "Albino Elephant"}
{"index" : {"_index" : "test", "_type" : "member"}}
{"object_name" : "Cute Albino Elephant", "object_name_other_language" : "Cute Albino Elephant"}
{"index" : {"_index" : "test", "_type" : "member"}}
{"object_name" : "The Cutest Albino Elephant", "object_name_other_language" : "The Cutest Albino Elephant"}
'; echo

# Wait for indexing
echo "Refreshing"
curl -XPOST 'http://localhost:9200/_refresh?pretty=yes'; echo

echo "Doing the search"
curl -XGET 'http://localhost:9200/test/member/_search?pretty=1'  -d '
{
    "explain" : "false",
    "query" : {
        "dis_max" : {
            "queries" : [
                {
                    "text" : {
                        "object_name.unmunged" : "Albino Elephant"
                    }
                },
                {
                    "text" : {
                        "object_name_other_language.unmunged" : "Albino Elephant"
                    }
                }
            ]
        }
    }
}
'; echo

# Results with the COMMENT_ in place for document #1:
  #   "hits" : [ {
  #     "_index" : "test",
  #     "_type" : "member",
  #     "_id" : "GKZDxtHpSKaksQH4reX89g",
  #     "_score" : 0.2169777, "_source" : {"object_name" : "The Cutest Albino Elephant", "object_name_other_language" : "The Cutest Albino Elephant"}
  #   }, {
  #     "_index" : "test",
  #     "_type" : "member",
  #     "_id" : "DVbb0HMeSdSkJajpwZ8PjA",
  #     "_score" : 0.2169777, "_source" : {"object_name" : "Cute Albino Elephant", "object_name_other_language" : "Cute Albino Elephant"}
  #   }, {
  #     "_index" : "test",
  #     "_type" : "member",
  #     "_id" : "oZxMt0PIQU2f5s7I7jo7fg",
  #     "_score" : 0.08322528, "_source" : {"object_name" : "Albino Elephant", "COMMENT_object_name_other_language" : "Albino Elephant"}
  #   } ]
  # }

# Results without it:
  #   "hits" : [ {
  #     "_index" : "test",
  #     "_type" : "member",
  #     "_id" : "QZwB23TgRY-N8hq7iZ43vQ",
  #     "_score" : 0.2712221, "_source" : {"object_name" : "Albino Elephant", "object_name_other_language" : "Albino Elephant"}
  #   }, {
  #     "_index" : "test",
  #     "_type" : "member",
  #     "_id" : "cMeLRXlZShyoZsYVI_BQsQ",
  #     "_score" : 0.2169777, "_source" : {"object_name" : "Cute Albino Elephant", "object_name_other_language" : "Cute Albino Elephant"}
  #   }, {
  #     "_index" : "test",
  #     "_type" : "member",
  #     "_id" : "eC1j163mTLeoltIjt7SXkg",
  #     "_score" : 0.2169777, "_source" : {"object_name" : "The Cutest Albino Elephant", "object_name_other_language" : "The Cutest Albino Elephant"}
  #   } ]
  # }