opensearch query builder

A non-obtrusive, idiomatic and easy-to-use query and aggregation builder for the official Opensearch client for Opensearch.

Table of Contents

• Description
• Status
• Installation
• Usage
• Notes
• Features
  • Supported Queries
  • Supported Aggregations
  • Custom Queries and Aggregations
• License

Description

esquery alleviates the need to use extremely nested maps (map[string]interface{}) and serializing queries to JSON manually. It also helps eliminating common mistakes such as misspelling query types, as everything is statically typed.

Using esquery can make your code much easier to write, read and maintain, and significantly reduce the amount of code you write. Wanna know how much code you'll save? just check this project's tests.

Status

This is an early release, API may still change.

Installation

esquery is a Go module. To install, simply run this in your project's root directory:

Usage

it provides a method chaining-style API for building and executing queries and aggregations. It does not wrap the official Go client nor does it require you to change your existing code in order to integrate the library. Queries can be directly built, and executed by passing an *opensearch.Client instance (with optional search parameters). Results are returned as-is from the official client (e.g. *opensearchapi.Response objects).

Getting started is extremely simple:

package main

import (
	"log"

	"github.com/opensearch-project/opensearch-go/v2"

	oq "github.com/deliveroo/os-query"
)

func main() {
	// connect to an Opensearch instance
	client, err := opensearch.NewDefaultClient()
	if err != nil {
		log.Fatalf("Failed creating client: %s", err)
	}

	query := oq.Query(
		oq.
			Bool().
			Must(oq.Term("title", "Go and Stuff")).
			Filter(oq.Term("tag", "tech")),
	).
		Aggs(
			oq.Avg("average_score", "score"),
			oq.Max("max_score", "score"),
		).
		Size(20)

	queryString, err := query.MarshalJSON()
	if err != nil {
		log.Fatal("Failed creating query")
	}
	// run a boolean search query
	search := client.Search
	res, err := search(search.WithQuery(string(queryString)))
	if err != nil {
		log.Fatal("Failed executing query")
	}

	defer res.Body.Close()
}

Notes

• The library cannot currently generate "short queries". For example, whereas ElasticSearch can accept this:

{ "query": { "term": { "user": "Kimchy" } } }

The library will always generate this:

{ "query": { "term": { "user": { "value": "Kimchy" } } } }

This is also true for queries such as "bool", where fields like "must" can either receive one query object, or an array of query objects. esquery will generate an array even if there's only one query object.

Features

Supported Queries

The following queries are currently supported:

ElasticSearch DSL	esquery Function
"match"	Match()
"match_bool_prefix"	MatchBoolPrefix()
"match_phrase"	MatchPhrase()
"match_phrase_prefix"	MatchPhrasePrefix()
"match_all"	MatchAll()
"match_none"	MatchNone()
"multi_match"	MultiMatch()
"exists"	Exists()
"fuzzy"	Fuzzy()
"ids"	IDs()
"prefix"	Prefix()
"range"	Range()
"regexp"	Regexp()
"term"	Term()
"terms"	Terms()
"terms_set"	TermsSet()
"wildcard"	Wildcard()
"bool"	Bool()
"boosting"	Boosting()
"constant_score"	ConstantScore()
"dis_max"	DisMax()

Supported Aggregations

The following aggregations are currently supported:

ElasticSearch DSL	esquery Function
"avg"	Avg()
"weighted_avg"	WeightedAvg()
"cardinality"	Cardinality()
"max"	Max()
"min"	Min()
"sum"	Sum()
"value_count"	ValueCount()
"percentiles"	Percentiles()
"stats"	Stats()
"string_stats"	StringStats()
"top_hits"	TopHits()
"terms"	TermsAgg()

Supported Top Level Options

The following top level options are currently supported:

ElasticSearch DSL	esquery.Search Function
"highlight"	Highlight()
"explain"	Explain()
"from"	From()
"postFilter"	PostFilter()
"query"	Query()
"aggs"	Aggs()
"size"	Size()
"sort"	Sort()
"source"	SourceIncludes(), SourceExcludes()
"timeout"	Timeout()

Custom Queries and Aggregations

To execute an arbitrary query or aggregation (including those not yet supported by the library), use the CustomQuery() or CustomAgg() functions, respectively. Both accept any map[string]interface{} value.

License

This library is distributed under the terms of the Apache License 2.0.