Search Templates in Elasticsearch

Introduction

Search templates in Elasticsearch allow you to define search queries with placeholders that can be replaced with actual values at runtime. This feature is useful for reusing common search patterns while still allowing dynamic input. In this tutorial, we will cover the basics of creating, storing, and using search templates in Elasticsearch.

Creating a Search Template

To create a search template, you need to define a template with placeholders. Placeholders are specified using the {{variable_name}} syntax. Below is an example of a simple search template:

{
  "template": {
    "query": {
      "match": {
        "title": "{{query_string}}"
      }
    }
  }
}
                

In this template, {{query_string}} is a placeholder that will be replaced with the actual search term at runtime.

Storing a Search Template

Elasticsearch provides a dedicated endpoint to store search templates. You can use the following command to store the above template:

PUT _scripts/my_search_template
{
  "script": {
    "lang": "mustache",
    "source": {
      "query": {
        "match": {
          "title": "{{query_string}}"
        }
      }
    }
  }
}
                

This command stores the template with the ID my_search_template.

Using a Stored Search Template

To use a stored search template, you can reference it in your search request. Here is an example of how to execute a search using the template we stored earlier:

POST _search/template
{
  "id": "my_search_template",
  "params": {
    "query_string": "Elasticsearch"
  }
}
                

In this request, we specify the template ID and provide the required parameters. The query_string parameter is replaced with the value "Elasticsearch" when the search is executed.

Inline Search Templates

Instead of storing a search template, you can also define it inline within your search request. Here is an example:

POST _search/template
{
  "source": {
    "query": {
      "match": {
        "title": "{{query_string}}"
      }
    }
  },
  "params": {
    "query_string": "Elasticsearch"
  }
}
                

This approach is useful for ad-hoc queries or when you do not want to store the template on the server.

Advanced Usage

Search templates can also include more complex logic and multiple placeholders. Here is an example of a more advanced template:

PUT _scripts/advanced_search_template
{
  "script": {
    "lang": "mustache",
    "source": {
      "query": {
        "bool": {
          "must": [
            { "match": { "title": "{{query_string}}" }},
            { "range": { "publish_date": { "gte": "{{start_date}}", "lte": "{{end_date}}" }}}
          ]
        }
      }
    }
  }
}
                

This template includes placeholders for a query string and a date range. You can use it in a search request as follows:

POST _search/template
{
  "id": "advanced_search_template",
  "params": {
    "query_string": "Elasticsearch",
    "start_date": "2021-01-01",
    "end_date": "2021-12-31"
  }
}
                

Conclusion

Search templates in Elasticsearch are a powerful feature for creating reusable and dynamic search queries. By using placeholders, you can define flexible search patterns that can be easily adapted to different scenarios. Whether you store templates on the server or define them inline, they can help streamline your search operations and improve maintainability.