search on description

description-search.sh

# ========================================
# Testing n-gram analysis in ElasticSearch
# ========================================

# The goal is to search for terms like
# key
# keyboard
# 8
# 8 cell
# 8x DVD
# CD-RW
# CDRW
# And only find those documents with the matching words, for key, it should find keyboard, for 8, it should find Li-ION 8 cells
# CDRW and CD-RW should be treated as the same term
# 8 cell should find 8 cells , but not 6 cells


curl -X DELETE localhost:9200/test
curl -X PUT    localhost:9200/test -d '
{
  "settings" : {
    "index" : {
      "analysis" : {
        "tokenizer" :{
          "description_tokenizer" : {
            "type" : "pattern",
            "pattern" : "\\s+"
          }
        },
        "analyzer" : {
          "description_analyzer" : {
            "type" : "custom",
            "tokenizer" : "description_tokenizer",
            "filter" : [ "ngram_desc", "desc_word_filter", "lowercase"]
          },
          "description_shingle_analyzer" : {
            "type" : "custom",
            "tokenizer" : "description_tokenizer",
            "filter" : [ "desc_shingle", "lowercase"]
          }
        },
        "filter" : {
          "ngram_desc" : {
            "type" : "nGram",
            "min_gram" : 3,
            "max_gram" : 3
          },
          "desc_shingle" : {
            "type": "shingle",
            "min_shingle_size": 2,
            "max_shingle_size": 2,
            "token_separator" : ""
          },
          "desc_word_filter": {
             "type": "word_delimiter",
             "generate_word_parts" : "false",
             "generate_number_parts" : "false",
             "split_on_numerics" : "false",
             "split_on_case_change" : "false",
             "catenate_words" : "true",
             "preserve_original" : "true"
          }
        }
      }
    }
  },

  "mappings": {
    "main": {
      "properties": {
        "description": {
          "type": "multi_field",
          "fields" : {
            "description" : {
              "type" : "string",
              "analyzer": "description_analyzer"
            },
            "description_shingle" : {
              "type" : "string",
              "analyzer": "description_shingle_analyzer"
            }
          }
        }
      }
    }
  }
}
'

curl -X POST "http://localhost:9200/test/main" -d '
{
"description" : "This keyBoard for Toshiba",
"description2" : "This keyBoard for Toshiba"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description" : "This keyboard for HP",
"description2" : "This keyboard for HP"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description" : "This battery li-ion toshiba",
"description2" : "This battery li-ion toshiba"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description" : "This CD-RW/ DVD DRIVE Combo toshiba",
"description2" : "This CD-RW/ DVD DRIVE Combo toshiba"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description" : "This cd-rw DVD DRIVE Combo toshiba",
"description2" : "This cd-rw DVD DRIVE Combo toshiba"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description" : "8X DVD Drive",
"description2" : "8X DVD Drive"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description" : "6X DVD Drive",
"description2" : "6X DVD Drive"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description"  : "Li-ION 6 Cells",
"description2" : "Li-ION 6 Cells"
}'
curl -X POST "http://localhost:9200/test/main" -d '
{
"description"  : "Li-ION 8 Cells",
"description2" : "Li-ION 8 Cells"
}'
curl -X POST "http://localhost:9200/test/_refresh"

curl "http://localhost:9200/test/_analyze?text=8x+mas+menos&analyzer=description_analyzer&pretty=true"

URLS='
http://localhost:9200/test/_search?q=description:key&default_operator=AND
http://localhost:9200/test/_search?q=description:key%20toshiba&default_operator=AND
http://localhost:9200/test/_search?q=description:toshiba&default_operator=AND
http://localhost:9200/test/_search?q=description:li-ion&default_operator=AND
http://localhost:9200/test/_search?q=description:cdrw&default_operator=AND
http://localhost:9200/test/_search?q=description:cd-rw&default_operator=AND
http://localhost:9200/test/_search?q=description:8x&default_operator=AND
http://localhost:9200/test/_search?q=description2:8x&default_operator=AND
http://localhost:9200/test/_search?q=description2:6+Cells&default_operator=AND
'

#for url in ${URLS}
#This is disable on purpose
for url in "nada"
do
  echo; echo; echo ">>> ${url}&pretty=true"
  #if which open &> /dev/null; then
  #  open "${url}&pretty=true"
  #fi
    curl "${url}&pretty=true"
done


curl "http://localhost:9200/test/_analyze?text=6+cell+battery&analyzer=description_shingle_analyzer&pretty=true"

PAYLOAD='
{"query":{"bool":{"should":
[
{
  "match":
    {
      "description":{
        "query":"6 cell","operator":"or"

      }
    }
},
{
  "match":
    {
      "description_shingle":{
        "query":"6 cell","operator":"or"
      }
    }
}
],
"disable_coord" : true,
"minimum_should_match" : "70%"
}}}
'
echo;echo ">>>";echo;

curl -XGET "http://localhost:9200/test/_search?pretty=true" -d "${PAYLOAD}"