Search Applications search API and templates

edit

Search Applications search API and templates

edit

Your search applications use search templates to perform searches. Templates help reduce complexity by exposing only template parameters, while using the full power of Elasticsearch’s query DSL to formulate queries. Templates may be set when creating or updating a search application, and can be customized. This template can be edited or updated at any time using the Put Search Application API API call.

In a nutshell, you create search templates with parameters instead of specific hardcoded search values. At search time, you pass in the actual values for these parameters, enabling customized searches without rewriting the entire query structure. Search Application templates:

  • Simplify query requests
  • Reduce request size
  • Ensure security and performance, as the query is predefined and can’t be changed arbitrarily

This document provides information and sample templates to get you started using search applications for additional use cases. These templates are designed to be easily modified to meet your needs. Once you’ve created a search application with a template, you can search your search application using this template.

Search templates use the Mustache templating language. Mustache variables are typically enclosed in double curly brackets like this: {{my-var}}.

Learn more by reading about search templates.

Default template example

edit

If no template is stored with a search application, a minimal default search template will be applied at search time. The default template implements a simple search use case.

To create a search application with the default template, issue a create or update Search Application request without specifying a template:

PUT _application/search_application/my_search_application
{
  "indices": ["index1", "index2"]
}

You can then use the get search application API call to view your newly created search application, which will also include the default template that was created for you:

GET _application/search_application/my_search_application

In this case, the response would be:

{
  "name": "my_search_application",
  "indices": [
    "index1",
    "index2"
  ],
  "updated_at_millis": 1715802354482,
  "template": {
    "script": {
      "source": """{
  "query": {
    "query_string": {
        "query": "{{query_string}}",
        "default_field": "{{default_field}}"
        }
    }
}
""",
      "lang": "mustache",
      "params": {
        "default_field": "*",
        "query_string": "*"
      }
    }
  }
}

The default template is very minimal:

{
  "template": {
    "script": {
      "source": {
        "query": {
          "query_string": {
            "query": "{{query_string}}",
            "default_field": "{{default_field}}"
          }
        }
      },
      "params": {
        "query_string": "*",
        "default_field": "*"
      }
    }
  }
}

This may be useful for initial exploration of search templates, but you’ll likely want to update this.

This template does not support additional parameters, including from, size or boost. If you need to use these, you can customize the template associated with your search application accordingly to include them as parameters.

You can see the parameters and their default values by viewing the template, but it also may be valuable to view the query that will be generated if you search your search application with various parameters.

You can use the render search application query to view the query this template would generate, including with default parameters. For example, searching the search application with no parameters:

POST _application/search_application/my_search_application/_render_query

will return:

{
  "query": {
    "query_string": {
      "query": "*",
      "default_field": "*",
      "fields": []
    }
  }
}

This uses the default parameters that were defined with the template. You can also specify one or mre parameters to the render call, for example:

POST _application/search_application/my_search_application/_render_query
{
  "params": {
    "query_string": "rock climbing"
  }
}

will return:

{
  "query": {
    "query_string": {
      "query": "rock climbing",
      "default_field": "*",
      "fields": []
    }
  }
}

In this case, the {{query_string}} parameter has been replaced with the value rock climbing, and the {{default_field}} parameter was not specified so used the default value of *.

When you actually perform a search with no parameters, it will execute the underlying query that the render call returned. In this case, a search with no parameters will return all results, in a similar manner to a parameterless call to /_search.

POST _application/search_application/my_search_application/_search

Searching with the query_string and/or default_field parameters will perform a query_string query.

The default template is subject to change in future versions of the Search Applications feature.

Try some of the other examples in this document to experiment with specific use cases, or try creating your own!

Searching a search application

edit
Template search
edit

The simplest way to interact with a search application is to use the search template that’s created and stored with it. Each search application has a single template associated with it, which defines search criteria, parameters and defaults.

You send search requests to a search application using the Search Application Search API.

With the default template, a search looks like this:

POST _application/search_application/my_search_application/_search
{
  "params": {
    "query_string": "kayaking"
  }
}

In this example, we’ve overridden the query_string parameter’s default value of *. Since we didn’t specify default_field the value of this parameter will still be *.

Alias search
edit

If you don’t want to set up a search template for your search application, an alias will be created with the same name as your search application. This may be helpful when experimenting with specific search queries that you want to use when building your search application’s search template.

If your search application’s name is my_search_application, your alias will be my_search_application. You can search this using the search API.

You should use the Search Applications management APIs to update your application and not directly use Elasticsearch APIs such as the alias API. For example, use PUT Search Application with the indices parameter. This will automatically keep the associated alias up to date and ensure that indices are added to the search application correctly.

Search template examples

edit

We have created a number of examples to explore specific use cases. Use these as a starting point for creating your own search templates.

Text search example
edit

The following template supports a multi_match search over specified fields and boosts:

PUT _application/search_application/my_search_application
{
  "indices": ["index1", "index2"],
  "template": {
    "script": {
      "lang": "mustache",
      "source": """
      {
        "query": {
          "multi_match": {
            "query": "{{query_string}}",
            "fields": [{{#text_fields}}"{{name}}^{{boost}}",{{/text_fields}}]
          }
        },
        "explain": "{{explain}}",
        "from": "{{from}}",
        "size": "{{size}}"
      }
      """,
      "params": {
        "query_string": "*",
        "text_fields": [
          {"name": "title", "boost": 10},
          {"name": "description", "boost": 5}
        ],
        "explain": false,
        "from": 0,
        "size": 10
      }
    }
  }
}

A search query using this template might look like this:

POST _application/search_application/my_search_application/_search
{
  "params": {
    "size": 5,
    "query_string": "mountain climbing",
    "text_fields": [
          {"name": "title", "boost": 10},
          {"name": "description", "boost": 2},
          {"name": "state", "boost": 1}
     ]
  }
}

The text_fields parameters can be overridden with new/different fields and boosts to experiment with the best configuration for your use case. This template also supports pagination and explain via parameters.

Text search + ELSER with RRF
edit

This example supports the reciprocal rank fusion (RRF)] method for combining BM25 and ELSER searches. Reciprocal Rank Fusion consistently improves the combined results of different search algorithms. It outperforms all other ranking algorithms, and often surpasses the best individual results, without calibration.

PUT _application/search_application/my-search-app
{
  "indices": [
    "index1"
  ],
  "template": {
    "script": {
      "lang": "mustache",
      "source": """
      {
        "retriever": {
          "rrf": {
            "retrievers": [
              {{#text_fields}}
              {
                "standard": {
                  "query": {
                    "match": {
                      "{{.}}": "{{query_string}}"
                    }
                  }
                }
              },
              {{/text_fields}}
              {{#elser_fields}}
              {
                "standard": {
                  "query": {
                    "text_expansion": {
                      "ml.inference.{{.}}_expanded.predicted_value": {
                        "model_text": "{{query_string}}",
                        "model_id": "<elser_model_id>"
                      }
                    }
                  }
                }
              },
              {{/elser_fields}}
            ],
            "window_size": {{rrf.window_size}},
            "rank_constant": {{rrf.rank_constant}}
          }
        }
      }
      """,
      "params": {
        "elser_fields": ["title", "meta_description"],
        "text_fields": ["title", "meta_description"],
        "query_string": "",
        "rrf": {
          "window_size": 100,
          "rank_constant": 60
        }
      }
    }
  }
}

Replace <elser_model_id> with the model ID of your ELSER deployment.

A sample query for this template will look like the following example:

POST _application/search_application/my-search-app/_search
{
  "params": {
    "query_string": "What is the most popular brand of coffee sold in the United States?",
    "elser_fields": ["title", "meta_description"],
    "text_fields": ["title", "meta_description"],
    "rrf": {
      "window_size": 50,
      "rank_constant": 25
    }
  }
}
Text search + ELSER
edit

The Elastic Learned Sparse EncodeR (ELSER) improves search relevance through text-expansion, which enables semantic search. This experimental template requires ELSER to be enabled for one or more fields. Refer to Semantic search with ELSER for more information on how to use ELSER. In this case, ELSER is enabled on the title and description fields.

This example provides a single template that you can use for various search application scenarios: text search, ELSER, or all of the above. It also provides a simple default query_string query if no parameters are specified.

PUT _application/search_application/my_search_application
{
  "indices": [
    "index1",
    "index2"
  ],
  "template": {
    "script": {
      "lang": "mustache",
      "source": """
      {
        "query": {
          "bool": {
            "should": [
              {{#text}}
              {
                "multi_match": {
                  "query": "{{query_string}}",
                  "fields": [{{#text_fields}}"{{name}}^{{boost}}",{{/text_fields}}],
                  "boost": "{{text_query_boost}}"
                }
              },
              {{/text}}
              {{#elser}}
              {{#elser_fields}}
              {
                "text_expansion": {
                  "ml.inference.{{name}}_expanded.predicted_value": {
                    "model_text": "{{query_string}}",
                    "model_id": ".elser_model_1",
                    "boost": "{{boost}}"
                  }
                }
              },
              {{/elser_fields}}
              { "bool": { "must": [] } },
              {{/elser}}
              {{^text}}
              {{^elser}}
              {
                "query_string": {
                  "query": "{{query_string}}",
                  "default_field": "{{default_field}}",
                  "default_operator": "{{default_operator}}",
                  "boost": "{{text_query_boost}}"
                }
              },
              {{/elser}}
              {{/text}}
              { "bool": { "must": [] } }
              ],
            "minimum_should_match": 1
          }
        },
        "min_score": "{{min_score}}",
        "explain": "{{explain}}",
        "from": "{{from}}",
        "size": "{{size}}"
      }
      """,
      "params": {
        "text": false,
        "elser": false,
        "elser_fields": [
          {"name": "title", "boost": 1},
          {"name": "description", "boost": 1}
        ],
        "text_fields": [
          {"name": "title", "boost": 10},
          {"name": "description", "boost": 5},
          {"name": "state", "boost": 1}
        ],
        "query_string": "*",
        "text_query_boost": 4,
        "default_field": "*",
        "default_operator": "OR",
        "explain": false,
        "from": 0,
        "size": 10,
        "min_score": 0
      }
    }
  }
}

A text search query using this template might look like this:

POST _application/search_application/my_search_application/_search
{
  "params": {
    "text": true,
    "size": 5,
    "query_string": "mountain climbing",
    "text_fields": [
          {"name": "title", "boost": 10},
          {"name": "description", "boost": 5},
          {"name": "state", "boost": 1}
     ]
  }
}

An ELSER search query using this template will look like the following example:

POST _application/search_application/my_search_application/_search
{
  "params": {
    "elser": true,
    "query_string": "where is the best mountain climbing?",
    "elser_fields": [
      {"name": "title", "boost": 1},
      {"name": "description", "boost": 1}
    ]
  }
}

A combined text search and ELSER search query using this template will look like the following example:

POST _application/search_application/my_search_application/_search
{
  "params": {
    "elser": true,
    "text": true,
    "query_string": "where is the best mountain climbing?",
    "elser_fields": [
      {"name": "title", "boost": 1},
      {"name": "description", "boost": 1}
    ],
    "text_query_boost": 4,
    "min_score": 10
  }
}

Text search results and ELSER search results are expected to have significantly different scores in some cases, which makes ranking challenging. To find the best search result mix for your dataset, we suggest experimenting with the boost values provided in the example template:

  • text_query_boost to boost the BM25 query as a whole
  • boost fields to boost individual text search fields
  • min_score parameter to omit significantly low confidence results

The above boosts should be sufficient for many use cases, but there are cases when adding a rescore query or index boost to your template may be beneficial. Remember to update your search application to use the new template using the put search application command.

Finally, a parameterless search using this template would fall back to a default search returning all documents:

POST _application/search_application/my_search_application/_search
ELSER search
edit

This example supports a streamlined version of ELSER search.

PUT _application/search_application/my_search_application
{
  "indices": [
    "index1",
    "index2"
    ],
    "template": {
      "script": {
        "lang": "mustache",
        "source": """
        {
          "query": {
            "bool": {
              "should": [
                {{#elser_fields}}
                {
                  "text_expansion": {
                    "ml.inference.{{name}}_expanded.predicted_value": {
                      "model_text": "{{query_string}}",
                      "model_id": "<elser_model_id>"
                    }
                  }
                },
                {{/elser_fields}}
                ]
            }
          },
          "min_score": "{{min_score}}"
        }
        """,
        "params": {
          "query_string": "*",
          "min_score": "10",
          "elser_fields": [
            {
              "name": "title"
            },
            {
              "name": "description"
            }
            ]
        }
      }
    }
}

Replace <elser_model_id> with the model ID of your ELSER deployment.

A sample query for this template will look like the following example:

POST _application/search_application/my_search_application/_search
  {
    "params": {
      "query_string": "Where is the best place for mountain climbing?"
    }
  }
kNN search
edit

This example supports k-nearest neighbor (kNN) search

A template supporting exact kNN search will look like the following example:

PUT _application/search_application/my_search_application
{
  "indices": [
    "index1"
  ],
  "template": {
    "script": {
      "lang": "mustache",
      "source": """
        {
          "query": {
            "script_score": {
              "query": {
                "bool": {
                  "filter": {
                    "range": {
                      "{{field}}": {
                        "{{operator}}": {{value}}
                      }
                    }
                  }
                }
              },
              "script": {
                "source": "cosineSimilarity({{#toJson}}query_vector{{/toJson}}, '{{dense_vector_field}}') + 1.0"
              }
            }
          }
        }
        """,
      "params": {
        "field": "price",
        "operator": "gte",
        "value": 1000,
        "dense_vector_field": "product-vector",
        "query_vector": []
      }
    }
  }
}

A search query using this template will look like the following example:

POST _application/search_application/my_search_application/_search
{
  "params": {
    "field": "price",
    "operator": "gte",
    "value": 500
  }
}

A template supporting approximate kNN search will look like the following example:

PUT _application/search_application/my_search_application
{
  "indices": [
    "index1"
  ],
  "template": {
    "script": {
      "lang": "mustache",
      "source": """
      {
          "knn": {
            "field": "{{knn_field}}",
            "query_vector": {{#toJson}}query_vector{{/toJson}},
            "k": "{{k}}",
            "num_candidates": {{num_candidates}}
          },
          "fields": {{#toJson}}fields{{/toJson}}
      }
      """,
      "params": {
        "knn_field": "image-vector",
        "query_vector": [],
        "k": 10,
        "num_candidates": 100,
        "fields": ["title", "file-type"]
      }
    }
  }
}

A search query using this template will look like the following example:

POST _application/search_application/my_search_application/_search
{
  "params": {
    "knn_field": "image-vector",
        "query_vector": [-5, 9, -12],
        "k": 10,
        "num_candidates": 100,
        "fields": ["title", "file-type"]
  }
}