CODEWITHSUNDEEP
elasticsearchdelimitercpu-word
Artanis Zeratul
Artanis Zeratul

Reputation: 1003

Elasticsearch and word_delimiter token filter

I have been reading this documentation from elasticsearch: https://www.elastic.co/guide/en/elasticsearch/reference/1.7/analysis-word-delimiter-tokenfilter.html

And at the same was searching the internet for examples. Unfortunately not just there were a few examples most of them doesn't work.

I would really appreciate it if someone can post or give an example on how to use word_delimeter token filter in elasticsearch.

Thanks.

Upvotes: 2

Views: 7953

Answers (2)

user14222769
user14222769

Reputation: 11

I know this questions is old, but anyways... I cant comment because i dont have reputation points, but if you try to search "everybody" there is no "-" as provided in the above answer,if you want to search "body" in "everybody" maybe a simple wildcard would help you. https://www.elastic.co/guide/en/elasticsearch/reference/6.8/query-dsl-wildcard-query.html

i hope it was usefull

Upvotes: 0

user3775217
user3775217

Reputation: 4803

Elasticversion - 5.2

Try the following mappings

PUT demo
{
    "settings": {
        "analysis": {
          "analyzer": {
            "index_analyzer_v1" : {
              "tokenizer" : "whitespace",
              "filter" : [ "word_delimeter"]
            }
          },
          "filter": {
            "ngram_filter" : {
              "type" : "nGram",
              "min_gram": 1,
          "max_gram": 10,
          "token_chars": [
            "letter",
            "digit"
          ]
            },
            "word_delimeter" : {
              "type" : "word_delimiter",
              "generate_number_parts" : true,
              "catenate_words" : true,
              "catenate_numbers": true,
              "preserve_original" : true,
              "stem_english_possessive": true

            },
            "stop_words" : {
              "type": "stop",
          "stopwords": ["and", "is", "the", "we", "in", "are", "was", "were", "of"]
            }
          }
        }
    },
    "mappings": {
            "product": {
                "dynamic": "strict",
                "properties": {
                  "name": {
                    "type": "text",
                    "analyzer": "index_analyzer_v1"
                  }
                }
            }
        }
}

Index the following document

POST demo/product
{
  "name":"SH-09"
}

Run the following query

POST demo/_search
{
  "query": {"bool": {"must": [
    {"term": {
      "name": {
        "value": "09"
      }
    }}
  ]}}
}

Furthur more if you want to see the values stored in inverted index, run the following query

GET demo/_analyze?analyzer=index_analyzer_v1&text=SH-09

Hope this helps

Upvotes: 3

Related Questions