Elastic search name testing: standard, asis, phonetic, nickname

es-name-test.sh

#!/bin/sh
# Delete old version to make sure new settings are applied
curl -XDELETE "localhost:9200/names-test/"
echo
# See https://github.com/elasticsearch/elasticsearch-analysis-phonetic
curl -XPUT "localhost:9200/names-test" -d '{
    "settings": {
        "analysis": {
            "analyzer": {
                "lowkeyword-analyzer": {
                    "type": "custom",
                    "tokenizer": "keyword",
                    "filter" : ["lowercase"]
                },
                "phonetic-analyzer": {
                    "type": "custom",
                    "tokenizer": "whitespace",
                    "filter" : ["standard", "lowercase", "dmeta-filter"]
                },
                "nickname-analyzer": {
                    "type": "custom",
                    "tokenizer": "whitespace",
                    "filter" : ["onetoken-filter", "lowercase", "nickname-filter"]
                }
            },
            "filter" : {
                "dmeta-filter" : {
                    "type" : "phonetic",
                    "encoder" : "doublemetaphone",
                    "replace" : true
                },
                "onetoken-filter" : {
                    "type" : "limit",
                    "max_token_count" : 1
                },
                "nickname-filter" : {
                    "type" : "synonym",
                    "synonyms_path" : "../config/analysis/nick2basename.txt"
                }
            }
        }
    },
    "mappings": {
        "_default_": {
          "_timestamp" : {
            "enabled" : true,
            "store" : true
          }
        },
        "doc": {
            "properties": {
                "name" : {
                  "type" : "multi_field",
                  "fields" : {
                    "name" : {"type" : "string", "index" : "analyzed",
                        "index_analyzer": "standard", "search_analyzer": "standard"},
                    "asis" : {"type" : "string", "index" : "analyzed",
                        "index_analyzer": "lowkeyword-analyzer", "search_analyzer": "lowkeyword-analyzer"},
                    "phonetic" : {"type" : "string", "index" : "analyzed",
                        "index_analyzer": "phonetic-analyzer", "search_analyzer": "phonetic-analyzer"},
                    "basename" : {"type" : "string", "index" : "analyzed",
                        "index_analyzer": "nickname-analyzer", "search_analyzer": "nickname-analyzer"}
                  }
                },
                "date" : {
                  "type" : "date",
                  "format" : "dateOptionalTime"
                }
            }
        }
    }
}'


# curious about path analyzer? test it:
echo testing lowkeyword analyzier
curl -XGET 'localhost:9200/names-test/_analyze?analyzer=lowkeyword-analyzer&pretty=1' -d 'ALAN JURGENSEN'
echo

echo testing phonetic analyzier
curl -XGET 'localhost:9200/names-test/_analyze?analyzer=phonetic-analyzer&pretty=1' -d 'ALEN JORGENSEN'
echo

echo testing nickname analyzier
curl -XGET 'localhost:9200/names-test/_analyze?analyzer=nickname-analyzer&pretty=1' -d 'BILL ALAN'
echo

#!/bin/sh
# first load :  schema-accts-search-exp.sh

putdata () {
# Put some new docs
curl -sXPUT "localhost:9200/names-test/doc/1001" -d '{"name": "Joe Johnson", "date": "2013-08-01"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1002" -d '{"name": "JOHN JOHNSON", "date": "2013-08-02"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1003" -d '{"name": "jeff smith", "date": "2013-08-03"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1004" -d '{"name": "Jeff Johnson", "date": "2013-08-04"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1005" -d '{"name": "john smith", "date": "2013-08-05"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1006" -d '{"name": "ALAN HANSON", "date": "2013-08-06"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1007" -d '{"name": "MIKE SMITH", "date": "2013-08-07"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1008" -d '{"name": "ALAN JURGENSEN", "date": "2013-08-08"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1009" -d '{"name": "ALLEN JORGENSON", "date": "2013-08-09"}'
echo
curl -sXPUT "localhost:9200/names-test/doc/1010" -d '{"name": "ALLEN JURGENSON", "date": "2013-08-10"}'
echo

curl -sXPOST "localhost:9200/names-test/_refresh"
echo
# check for existence via:
curl -sXHEAD 'localhost:9200/names-test/doc/1001'
echo
# get doc
curl -sXGET 'localhost:9200/names-test/doc/1008?pretty'
echo
curl -sXGET 'localhost:9200/names-test/doc/1009?pretty'
echo

# get specific fields (default field is _source)
curl -sXGET 'localhost:9200/names-test/doc/1008?fields=_timestamp,name&pretty'
echo
}

# search and match _all
curl -sXPOST "localhost:9200/names-test/_search?pretty" -d '{"fields": ["_timestamp","_source"], "query": {"match_all" : {}}}'
echo

# curious about path analyzer? test it:
echo testing lowkeyword analyzier
curl -sXGET 'localhost:9200/names-test/_analyze?analyzer=lowkeyword-analyzer&pretty' -d 'ALAN JURGENSEN'
echo

echo testing phonetic analyzier
curl -sXGET 'localhost:9200/names-test/_analyze?analyzer=phonetic-analyzer&pretty' -d 'ALAN JURGENSEN'
echo

echo testing nickname to basename analyzier
curl -sXGET 'localhost:9200/names-test/_analyze?analyzer=nickname-analyzer&pretty' -d 'BILL ALAN'
echo

echo 'Name match query: name'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty" -d '{
    "query": {
        "match": {
            "name": "ALAN"
        }
    }
}'
echo

echo 'Name match query: name.asis'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty" -d '{
    "query": {
        "match": {
            "name.asis": "ALAN"
        }
    }
}'
echo

echo 'Name match query: name.asis'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty" -d '{
    "query": {
        "match": {
            "name.asis": "alan jurgensen"
        }
    }
}'
echo

echo 'Name match query: name.phonetic'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty" -d '{
    "query": {
        "match": {
            "name.phonetic": "alan jurgensen"
        }
    }
}'
echo

echo 'Name match query: name.basename'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty" -d '{
    "query": {
        "match": {
            "name.basename": "allen"
        }
    }
}'
echo

echo 'Name multi-match query: name.name name.asis name.phonetic name.basename'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty" -d '{
    "query": {
        "multi_match": {
            "query": "alan Jurgensen",
            "fields": [ "name.name", "name.asis", "name.phonetic", "name.basename" ]
        }
    }
}'
echo

echo 'Name multi-match BOOST query: name.name name.asis name.phonetic name.basename'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty" -d '{
    "query": {
        "multi_match": {
            "query": "alan Jurgensen",
            "fields": [ "name.name^1.6", "name.asis^3", "name.phonetic^1.4", "name.basename^1.2" ]
        }
    }
}'
echo

echo 'Name multi-match BIGBOOST query: name.name name.asis name.phonetic name.basename'
curl -sXPOST "localhost:9200/names-test/doc/_search?pretty&explain" -d '{
    "query": {
            "multi_match": {
                "query": "alan Jurgensen",
                "fields": [ "name.name^4", "name.asis^6", "name.phonetic^3", "name.basename^2" ]
            }
    }
}'
echo