Remove duplicates token filter

Remove duplicates token filter

Removes duplicate tokens in the same position.

The remove_duplicates filter uses Lucene’s RemoveDuplicatesTokenFilter.

Example

To see how the remove_duplicates filter works, you first need to produce a token stream containing duplicate tokens in the same position.

The following analyze API request uses the keyword_repeat and stemmer filters to create stemmed and unstemmed tokens for jumping dog.

resp = client.indices.analyze(
    tokenizer="whitespace",
    filter=[
        "keyword_repeat",
        "stemmer"
    ],
    text="jumping dog",
)
print(resp)

response = client.indices.analyze(
  body: {
    tokenizer: 'whitespace',
    filter: [
      'keyword_repeat',
      'stemmer'
    ],
    text: 'jumping dog'
  }
)
puts response

const response = await client.indices.analyze({
  tokenizer: "whitespace",
  filter: ["keyword_repeat", "stemmer"],
  text: "jumping dog",
});
console.log(response);

GET _analyze
{
  "tokenizer": "whitespace",
  "filter": [
    "keyword_repeat",
    "stemmer"
  ],
  "text": "jumping dog"
}

The API returns the following response. Note that the dog token in position 1 is duplicated.

{
  "tokens": [
    {
      "token": "jumping",
      "start_offset": 0,
      "end_offset": 7,
      "type": "word",
      "position": 0
    },
    {
      "token": "jump",
      "start_offset": 0,
      "end_offset": 7,
      "type": "word",
      "position": 0
    },
    {
      "token": "dog",
      "start_offset": 8,
      "end_offset": 11,
      "type": "word",
      "position": 1
    },
    {
      "token": "dog",
      "start_offset": 8,
      "end_offset": 11,
      "type": "word",
      "position": 1
    }
  ]
}

To remove one of the duplicate dog tokens, add the remove_duplicates filter to the previous analyze API request.

resp = client.indices.analyze(
    tokenizer="whitespace",
    filter=[
        "keyword_repeat",
        "stemmer",
        "remove_duplicates"
    ],
    text="jumping dog",
)
print(resp)

response = client.indices.analyze(
  body: {
    tokenizer: 'whitespace',
    filter: [
      'keyword_repeat',
      'stemmer',
      'remove_duplicates'
    ],
    text: 'jumping dog'
  }
)
puts response

const response = await client.indices.analyze({
  tokenizer: "whitespace",
  filter: ["keyword_repeat", "stemmer", "remove_duplicates"],
  text: "jumping dog",
});
console.log(response);

GET _analyze
{
  "tokenizer": "whitespace",
  "filter": [
    "keyword_repeat",
    "stemmer",
    "remove_duplicates"
  ],
  "text": "jumping dog"
}

The API returns the following response. There is now only one dog token in position 1.

{
  "tokens": [
    {
      "token": "jumping",
      "start_offset": 0,
      "end_offset": 7,
      "type": "word",
      "position": 0
    },
    {
      "token": "jump",
      "start_offset": 0,
      "end_offset": 7,
      "type": "word",
      "position": 0
    },
    {
      "token": "dog",
      "start_offset": 8,
      "end_offset": 11,
      "type": "word",
      "position": 1
    }
  ]
}

Add to an analyzer

The following create index API request uses the remove_duplicates filter to configure a new custom analyzer.

This custom analyzer uses the keyword_repeat and stemmer filters to create a stemmed and unstemmed version of each token in a stream. The remove_duplicates filter then removes any duplicate tokens in the same position.

resp = client.indices.create(
    index="my-index-000001",
    settings={
        "analysis": {
            "analyzer": {
                "my_custom_analyzer": {
                    "tokenizer": "standard",
                    "filter": [
                        "keyword_repeat",
                        "stemmer",
                        "remove_duplicates"
                    ]
                }
            }
        }
    },
)
print(resp)

response = client.indices.create(
  index: 'my-index-000001',
  body: {
    settings: {
      analysis: {
        analyzer: {
          my_custom_analyzer: {
            tokenizer: 'standard',
            filter: [
              'keyword_repeat',
              'stemmer',
              'remove_duplicates'
            ]
          }
        }
      }
    }
  }
)
puts response

const response = await client.indices.create({
  index: "my-index-000001",
  settings: {
    analysis: {
      analyzer: {
        my_custom_analyzer: {
          tokenizer: "standard",
          filter: ["keyword_repeat", "stemmer", "remove_duplicates"],
        },
      },
    },
  },
});
console.log(response);

PUT my-index-000001
{
  "settings": {
    "analysis": {
      "analyzer": {
        "my_custom_analyzer": {
          "tokenizer": "standard",
          "filter": [
            "keyword_repeat",
            "stemmer",
            "remove_duplicates"
          ]
        }
      }
    }
  }
}