Character filters in Elasticsearch - Mini Project: Build & Apply

Create a Custom Character Filter in Elasticsearch

📖 Scenario: You are setting up an Elasticsearch index for a book store. You want to clean up the text data by replacing certain characters before analysis.

🎯 Goal: Build an Elasticsearch index with a custom character filter that replaces the ampersand (&) with the word 'and' in book titles.

📋 What You'll Learn

Create an index called bookstore

Define a custom character filter named ampersand_filter that replaces '&' with 'and'

Add an analyzer called custom_analyzer that uses the ampersand_filter

Apply the custom_analyzer to the title field in the mapping

💡 Why This Matters

🌍 Real World

Cleaning and normalizing text data in search indexes to improve search accuracy and relevance.

💼 Career

Elasticsearch engineers and developers often create custom analyzers with character filters to handle special characters and improve search results.

1

Create the bookstore index with empty settings

Create an Elasticsearch index called bookstore with empty settings and mappings objects.

Elasticsearch

{
  "settings": {},
  "mappings": {}
}
# Your code here

Need a hint?

Use the PUT /bookstore API with an empty settings and mappings object.

2

Add a custom character filter ampersand_filter

In the settings of the bookstore index, add a char_filter named ampersand_filter that replaces the character '&' with the string 'and'.

Elasticsearch

{
  "settings": {
    "analysis": {
      "char_filter": {
        "ampersand_filter": {
          "type": "mapping",
          "mappings": [
            "&=>and"
          ]
        }
      }
    }
  },
  "mappings": {}
}
# Your code here

Need a hint?

Use the mapping type for the character filter and specify the mapping as &=>and.

3

Create a custom analyzer using the ampersand_filter

In the settings.analysis, add an analyzer named custom_analyzer that uses the ampersand_filter as a char_filter and the standard tokenizer.

Elasticsearch

{
  "settings": {
    "analysis": {
      "char_filter": {
        "ampersand_filter": {
          "type": "mapping",
          "mappings": [
            "&=>and"
          ]
        }
      },
      "analyzer": {
        "custom_analyzer": {
          "type": "custom",
          "char_filter": ["ampersand_filter"],
          "tokenizer": "standard"
        }
      }
    }
  },
  "mappings": {}
}
# Your code here

Need a hint?

Define a custom analyzer with the char_filter array including ampersand_filter and use the standard tokenizer.

4

Apply the custom_analyzer to the title field mapping

In the mappings.properties, define a title field of type text that uses the custom_analyzer analyzer.

Elasticsearch

{
  "settings": {
    "analysis": {
      "char_filter": {
        "ampersand_filter": {
          "type": "mapping",
          "mappings": [
            "&=>and"
          ]
        }
      },
      "analyzer": {
        "custom_analyzer": {
          "type": "custom",
          "char_filter": ["ampersand_filter"],
          "tokenizer": "standard"
        }
      }
    }
  },
  "mappings": {
    "properties": {
      "title": {
        "type": "text",
        "analyzer": "custom_analyzer"
      }
    }
  }
}
# Your code here

Need a hint?

Under mappings.properties, define the title field as text with the custom_analyzer.