elastic-builder

elastic-builder

A Node.js implementation of the Elasticsearch DSL for use with the official elasticsearch javascript client with builder syntax.

Check out the API reference documentation.

elastic-builder includes TypeScript definition for superior development experience.

For a zero DSL way to build elasticsearch queries, checkout elastic-muto

Elasticsearch compatibility

elastic-builder was built for 5.x query DSL. However, the library should be usable with 2.x as well. For older versions of the DSL, you can try elastic.js or bodybuilder

elastic-builder is also compatible with elasticsearch 6.0(alpha) for the most part. However, there are some breaking changes which have been called out in the docs(ex: Script.file.

What's Included:

• Request Body Search
• Queries
• Aggregations
• Suggesters
• Search Template

Install

npm install elastic-builder --save

Usage

const esb = require('elastic-builder'); // the builder

const requestBody = esb.requestBodySearch()
  .query(esb.matchQuery('message', 'this is a test'));

// OR

const requestBody = new esb.RequestBodySearch().query(
  new esb.MatchQuery('message', 'this is a test')
);

requestBody.toJSON(); // or print to console - esb.prettyPrint(requestBody)
{
  "query": {
    "match": {
      "message": "this is a test"
    }
  }
}

For each class, MyClass, a utility function myClass has been provided which contructs the object for us without the need for new keyword.

REPL

Try it out on the command line using the node REPL:

# Start the repl
node ./node_modules/elastic-builder/repl.js
# The builder is available in the context variable esb
elastic-builder > esb.prettyPrint(
...   esb.requestBodySearch()
...     .query(esb.matchQuery('message', 'this is a test'))
... );
{
  "query": {
    "match": {
      "message": "this is a test"
    }
  }
}

Motivation

Elasticsearch only provides a low level client for making requests. elastic.js was a relatively popular library for building the request search body. However, this project is not being maintained nor is the fork. There were several changes in the 5.0 release which make the older libraries unusable.

This library is a port of elastic.js to es6 with elasticsearch 5.x compatibility.

API Reference

API reference can be accessed here - https://elastic-builder.js.org/docs. The docs include examples ported from the official elasticsearch reference.

API documentation was generated using documentation.js. It is being hosted with help from this awesome project - https://github.com/js-org/dns.js.org

Recipes

The library has a few helper recipes:

• Missing query
• Random sort query
• Filter query

const qry = esb.cookMissingQuery('user');

qry.toJSON();
{
  "bool": {
    "must_not": {
      "exists": { "field": "user" }
    }
  }
}

Check out the reference docs for more examples.

If you have any recipes, please do share or better yet, create a pull request :smile:.

Changelog

releases

Examples

Usage with official elasticsearch client:

'use strict';

const elasticsearch = require('elasticsearch');
const esb = require('elastic-builder');

const client = new elasticsearch.Client({
  host: 'localhost:9200',
  log: 'trace'
});

const requestBody = esb.requestBodySearch()
  .query(esb.matchQuery('body', 'elasticsearch'));

client.search({
    index: 'twitter',
    type: 'tweets',
    body: requestBody.toJSON()
  })
  .then(resp => {
    const hits = resp.hits.hits;
  })
  .catch(err => {
    console.trace(err.message);
  });

// Bool query
const requestBody = esb.requestBodySearch().query(
  esb.boolQuery()
    .must(esb.matchQuery('last_name', 'smith'))
    .filter(esb.rangeQuery('age').gt(30))
);
requestBody.toJSON();
{
  "query": {
    "bool": {
      "must": {
        "match": { "last_name": "smith" }
      },
      "filter": {
        "range": { "age": { "gt": 30 } }
      }
    }
  }
}

// Multi Match Query
const requestBody = esb.requestBodySearch().query(
  esb.multiMatchQuery(['title', 'body'], 'Quick brown fox')
    .type('best_fields')
    .tieBreaker(0.3)
    .minimumShouldMatch('30%')
);
requestBody.toJSON();
{
  "multi_match": {
    "query": "Quick brown fox",
    "type": "best_fields",
    "fields": ["title", "body"],
    "tie_breaker": 0.3,
    "minimum_should_match": "30%"
  }
}

// Aggregation
const requestBody = esb.requestBodySearch()
  .size(0)
  .agg(esb.termsAggregation('popular_colors', 'color'));
requestBody.toJSON();
{
  "size": 0,
  "aggs": {
    "popular_colors": {
      "terms": { "field": "color" }
    }
  }
}

// Nested Aggregation
const requestBody = esb.requestBodySearch()
  .size(0)
  .agg(
    esb.termsAggregation('colors', 'color')
      .agg(esb.avgAggregation('avg_price', 'price'))
      .agg(esb.termsAggregation('make', 'make'))
  );
requestBody.toJSON();
{
  "size": 0,
  "aggs": {
    "colors": {
      "terms": { "field": "color" },
      "aggs": {
        "avg_price": {
          "avg": { "field": "price" }
        },
        "make": {
          "terms": { "field": "make" }
        }
      }
    }
  }
}

// If you prefer using the `new` keyword
const agg = new esb.TermsAggregation('countries', 'artist.country')
  .order('rock>playback_stats.avg', 'desc')
  .agg(
    new esb.FilterAggregation('rock', new esb.TermQuery('genre', 'rock')).agg(
      new esb.StatsAggregation('playback_stats', 'play_count')
    )
  );
agg.toJSON();
{
  "countries": {
    "terms": {
      "field": "artist.country",
      "order": { "rock>playback_stats.avg": "desc" }
    },
    "aggs": {
      "rock": {
        "filter": {
          "term": { "genre": "rock" }
        },
        "aggs": {
          "playback_stats": {
            "stats": { "field": "play_count" }
          }
        }
      }
    }
  }
}

// Sort
const requestBody = esb.requestBodySearch()
  .query(esb.boolQuery().filter(esb.termQuery('message', 'test')))
  .sort(esb.sort('timestamp', 'desc'))
  .sorts([
    esb.sort('channel', 'desc'),
    esb.sort('categories', 'desc'),
    // The order defaults to desc when sorting on the _score,
    // and defaults to asc when sorting on anything else.
    esb.sort('content'),
    esb.sort('price').order('desc').mode('avg')
  ]);
requestBody.toJSON();
{
  "query": {
    "bool": {
      "filter": {
        "term": { "message": "test" }
      }
    }
  },
  "sort": [
    { "timestamp": { "order": "desc" } },
    { "channel": { "order": "desc" } },
    { "categories": { "order": "desc" } },
    "content",
    { "price": { "order": "desc", "mode": "avg" } }
  ]
}

// From / size
const requestBody = esb.requestBodySearch()
  .query(esb.matchAllQuery())
  .size(5)
  .from(10);
requestBody.toJSON();
{
  "query": { "match_all": {} },
  "size": 5,
  "from": 10
}

For more examples, check out the reference docs.

Validation

elastic-builder provides lightweight validation where ever possible:

$ node ./node_modules/elastic-builder/repl.js
elastic-builder > esb.multiMatchQuery().field('title').field('body').query('Quick brown fox').type('bwst_fields')
See https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-multi-match-query.html
Got 'type' - bwst_fields
Error: The 'type' parameter should belong to Set {
  'best_fields',
  'most_fields',
  'cross_fields',
  'phrase',
  'phrase_prefix' }
    at MultiMatchQuery.type (E:\Projects\repos\elastic-builder\lib\queries\full-text-queries\multi-match-query.js:134:23)
    at repl:1:77
    at ContextifyScript.Script.runInContext (vm.js:35:29)
    at REPLServer.defaultEval (repl.js:342:29)
    at bound (domain.js:280:14)
    at REPLServer.runBound [as eval] (domain.js:293:12)
    at REPLServer.<anonymous> (repl.js:538:10)
    at emitOne (events.js:96:13)
    at REPLServer.emit (events.js:188:7)
    at REPLServer.Interface._onLine (readline.js:239:10)

Tests

Run unit tests:

npm test

Credits

elastic-builder is heavily inspired by elastic.js and the fork by Erwan Pigneul.

bodybuilder for documentation style, build setup, demo page.

Related

elastic-muto - Easy expressive search queries for Elasticsearch

License

MIT © Suhas Karanth