Example of inline index using `esse` gem

esse_index.rb

# Usage:
# ruby esse_index.rb

require "bundler/inline"

gemfile(true, quiet: true) do
  source "https://rubygems.org"

  gem "esse"
  gem "opensearch-ruby"
  gem "pry"
end

Esse.configure do |config|
  config.cluster(:default) do |cluster|
    cluster.client = OpenSearch::Client.new(url: "http://localhost:9200")
  end
end

class ExampleIndex < Esse::Index
  self.index_name = "example"

  repository :default do
    collection do |**_context, &block|
      [
        (1..10).map { |i| { id: i, name: "name-#{i}" } },
        (11..15).map { |i| { id: i, name: "name-#{i}" } },
      ].each do |collection|
        puts "Processing collection: #{collection.map { |h| h[:id] }}"
        block.call(collection) # You may pass kwargs to the block, document will receive them in each interaction
      end
    end

    document do |record, **_collection_given_context|
      puts "--> Processing document: #{record[:id]}"
      {
        _id: record[:id],
        name: record[:name],
        indexed_at: Time.now.utc,
      }
    end
  end
end

require "pry"
binding.pry

esse_index_refactor.rb

require "bundler/inline"

gemfile(true, quiet: true) do
  source "https://rubygems.org"

  gem "esse"
  gem "opensearch-ruby"
  gem "pry"
end

Esse.configure do |config|
  config.cluster(:default) do |cluster|
    cluster.client = OpenSearch::Client.new(url: "http://localhost:9200")
  end
end

class ExampleIndexCollection < Esse::Collection
  DATA = [
    (1..10).map { |i| { id: i, name: "name-#{i}" } },
    (11..15).map { |i| { id: i, name: "name-#{i}" } },
  ]

  def each
    DATA.each do |collection|
      puts "Processing collection: #{collection.map { |h| h[:id] }}"
      yield collection
    end
  end
end

class ExampleIndexDocument < Esse::Document
  def initialize(doc, **)
    puts "--> Processing document: #{doc[:id]}"
    super
  end

  def id
    object[:id]
  end

  def source
    {
      name: object[:name],
      indexed_at: Time.now
    }
  end
end

class ExampleIndex < Esse::Index
  self.index_name = "example"

  repository :default do
    collection ExampleIndexCollection
    document ExampleIndexDocument
  end
end

require "pry"
binding.pry

xpry_1_documents_enumerator.txt

[1] pry(main)> ExampleIndex.documents
=> #<Enumerator: ...>

xpry_2_documents_first.txt

[2] pry(main)> ExampleIndex.documents.first
Processing collection: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
--> Processing document: 1
--> Processing document: 2
--> Processing document: 3
--> Processing document: 4
--> Processing document: 5
--> Processing document: 6
--> Processing document: 7
--> Processing document: 8
--> Processing document: 9
--> Processing document: 10
=> #<Esse::HashDocument id: 1, source: {:name=>"name-1", :indexed_at=>2025-01-14 21:39:26.200505929 UTC}>

xpry_3_documents_count.txt

[3] pry(main)> ExampleIndex.documents.count
Processing collection: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
--> Processing document: 1
--> Processing document: 2
--> Processing document: 3
--> Processing document: 4
--> Processing document: 5
--> Processing document: 6
--> Processing document: 7
--> Processing document: 8
--> Processing document: 9
--> Processing document: 10
Processing collection: [11, 12, 13, 14, 15]
--> Processing document: 11
--> Processing document: 12
--> Processing document: 13
--> Processing document: 14
--> Processing document: 15
=> 15

xpry_5_create_index.txt

[5] pry(main)> ExampleIndex.create_index(alias: true, suffix: 'v1')
=> {"acknowledged"=>true, "shards_acknowledged"=>true, "index"=>"example_v1"}

xpry_6_count.txt

[6] pry(main)> ExampleIndex.count
=> 0

xpry_7_import.txt

[7] pry(main)> ExampleIndex.import
Processing collection: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
--> Processing document: 1
--> Processing document: 2
--> Processing document: 3
--> Processing document: 4
--> Processing document: 5
--> Processing document: 6
--> Processing document: 7
--> Processing document: 8
--> Processing document: 9
--> Processing document: 10
Processing collection: [11, 12, 13, 14, 15]
--> Processing document: 11
--> Processing document: 12
--> Processing document: 13
--> Processing document: 14
--> Processing document: 15
=> 15
[8] pry(main)> ExampleIndex.count
=> 15
[9] pry(main)> ExampleIndex.get(id: 12)
=> {"_index"=>"example_v1", "_id"=>"12", "_version"=>1, "_seq_no"=>11, "_primary_term"=>1, "found"=>true, "_source"=>{"name"=>"name-12", "indexed_at"=>"2025-01-14 21:41:43 UTC"}}