searchkick 4.0.0 → 5.0.0

data/lib/searchkick/bulk_reindex_job.rb

@@ -2,16 +2,20 @@ module Searchkick
   class BulkReindexJob < ActiveJob::Base
     queue_as { Searchkick.queue_name }
 
+    # TODO remove min_id and max_id in Searchkick 6
     def perform(class_name:, record_ids: nil, index_name: nil, method_name: nil, batch_id: nil, min_id: nil, max_id: nil)
-      klass = class_name.constantize
-      index = index_name ? Searchkick::Index.new(index_name, **klass.searchkick_options) : klass.searchkick_index
+      model = Searchkick.load_model(class_name)
+      index = model.searchkick_index(name: index_name)
+
+      # legacy
       record_ids ||= min_id..max_id
-      index.import_scope(
-        Searchkick.load_records(klass, record_ids),
-        method_name: method_name,
-        batch: true,
-        batch_id: batch_id
-      )
+
+      relation = Searchkick.scope(model)
+      relation = Searchkick.load_records(relation, record_ids)
+      relation = relation.search_import if relation.respond_to?(:search_import)
+
+      RecordIndexer.new(index).reindex(relation, mode: :inline, method_name: method_name, full: false)
+      RelationIndexer.new(index).batch_completed(batch_id) if batch_id
     end
   end
 end

data/lib/searchkick/controller_runtime.rb

@@ -0,0 +1,40 @@
+# based on https://gist.github.com/mnutt/566725
+module Searchkick
+  module ControllerRuntime
+    extend ActiveSupport::Concern
+
+    protected
+
+    attr_internal :searchkick_runtime
+
+    def process_action(action, *args)
+      # We also need to reset the runtime before each action
+      # because of queries in middleware or in cases we are streaming
+      # and it won't be cleaned up by the method below.
+      Searchkick::LogSubscriber.reset_runtime
+      super
+    end
+
+    def cleanup_view_runtime
+      searchkick_rt_before_render = Searchkick::LogSubscriber.reset_runtime
+      runtime = super
+      searchkick_rt_after_render = Searchkick::LogSubscriber.reset_runtime
+      self.searchkick_runtime = searchkick_rt_before_render + searchkick_rt_after_render
+      runtime - searchkick_rt_after_render
+    end
+
+    def append_info_to_payload(payload)
+      super
+      payload[:searchkick_runtime] = (searchkick_runtime || 0) + Searchkick::LogSubscriber.reset_runtime
+    end
+
+    module ClassMethods
+      def log_process_action(payload)
+        messages = super
+        runtime = payload[:searchkick_runtime]
+        messages << ("Searchkick: %.1fms" % runtime.to_f) if runtime.to_f > 0
+        messages
+      end
+    end
+  end
+end

data/lib/searchkick/index.rb

@@ -1,9 +1,5 @@
-require "searchkick/index_options"
-
 module Searchkick
   class Index
-    include IndexOptions
-
     attr_reader :name, :options
 
     def initialize(name, options = {})
@@ -12,6 +8,10 @@ module Searchkick
       @klass_document_type = {} # cache
     end
 
+    def index_options
+      IndexOptions.new(self).index_options
+    end
+
     def create(body = {})
       client.indices.create index: name, body: body
     end
@@ -38,16 +38,19 @@ module Searchkick
       client.indices.exists_alias name: name
     end
 
+    # call to_h for consistent results between elasticsearch gem 7 and 8
+    # could do for all API calls, but just do for ones where return value is focus for now
     def mapping
-      client.indices.get_mapping index: name
+      client.indices.get_mapping(index: name).to_h
     end
 
+    # call to_h for consistent results between elasticsearch gem 7 and 8
     def settings
-      client.indices.get_settings index: name
+      client.indices.get_settings(index: name).to_h
     end
 
     def refresh_interval
-      settings.values.first["settings"]["index"]["refresh_interval"]
+      index_settings["refresh_interval"]
     end
 
     def update_settings(settings)
@@ -82,7 +85,8 @@ module Searchkick
       old_indices =
         begin
           client.indices.get_alias(name: name).keys
-        rescue Elasticsearch::Transport::Transport::Errors::NotFound
+        rescue => e
+          raise e unless Searchkick.not_found_error?(e)
           {}
         end
       actions = old_indices.map { |old_name| {remove: {index: old_name, alias: name}} } + [{add: {index: new_name, alias: name}}]
@@ -94,7 +98,7 @@ module Searchkick
       record_data = RecordData.new(self, record).record_data
 
       # remove underscore
-      get_options = Hash[record_data.map { |k, v| [k.to_s.sub(/\A_/, "").to_sym, v] }]
+      get_options = record_data.to_h { |k, v| [k.to_s.sub(/\A_/, "").to_sym, v] }
 
       client.get(get_options)["_source"]
     end
@@ -103,11 +107,12 @@ module Searchkick
       indices =
         begin
           if client.indices.respond_to?(:get_alias)
-            client.indices.get_alias
+            client.indices.get_alias(index: "#{name}*")
           else
             client.indices.get_aliases
           end
-        rescue Elasticsearch::Transport::Transport::Errors::NotFound
+        rescue => e
+          raise e unless Searchkick.not_found_error?(e)
           {}
         end
       indices = indices.select { |_k, v| v.empty? || v["aliases"].empty? } if unaliased
@@ -123,32 +128,47 @@ module Searchkick
       indices
     end
 
-    # record based
-    # use helpers for notifications
-
     def store(record)
-      bulk_indexer.bulk_index([record])
+      notify(record, "Store") do
+        queue_index([record])
+      end
     end
 
     def remove(record)
-      bulk_indexer.bulk_delete([record])
+      notify(record, "Remove") do
+        queue_delete([record])
+      end
     end
 
     def update_record(record, method_name)
-      bulk_indexer.bulk_update([record], method_name)
+      notify(record, "Update") do
+        queue_update([record], method_name)
+      end
     end
 
     def bulk_delete(records)
-      bulk_indexer.bulk_delete(records)
+      return if records.empty?
+
+      notify_bulk(records, "Delete") do
+        queue_delete(records)
+      end
     end
 
     def bulk_index(records)
-      bulk_indexer.bulk_index(records)
+      return if records.empty?
+
+      notify_bulk(records, "Import") do
+        queue_index(records)
+      end
     end
     alias_method :import, :bulk_index
 
     def bulk_update(records, method_name)
-      bulk_indexer.bulk_update(records, method_name)
+      return if records.empty?
+
+      notify_bulk(records, "Update") do
+        queue_update(records, method_name)
+      end
     end
 
     def search_id(record)
@@ -160,18 +180,25 @@ module Searchkick
     end
 
     def similar_record(record, **options)
-      like_text = retrieve(record).to_hash
-        .keep_if { |k, _| !options[:fields] || options[:fields].map(&:to_s).include?(k) }
-        .values.compact.join(" ")
-
-      options[:where] ||= {}
-      options[:where][:_id] ||= {}
-      options[:where][:_id][:not] = Array(options[:where][:_id][:not]) + [record.id.to_s]
       options[:per_page] ||= 10
-      options[:similar] = true
+      options[:similar] = [RecordData.new(self, record).record_data]
+      options[:models] ||= [record.class] unless options.key?(:model)
 
-      # TODO use index class instead of record class
-      Searchkick.search(like_text, model: record.class, **options)
+      Searchkick.search("*", **options)
+    end
+
+    def reload_synonyms
+      if Searchkick.opensearch?
+        client.transport.perform_request "POST", "_plugins/_refresh_search_analyzers/#{CGI.escape(name)}"
+      else
+        raise Error, "Requires Elasticsearch 7.3+" if Searchkick.server_below?("7.3.0")
+        begin
+          client.transport.perform_request("GET", "#{CGI.escape(name)}/_reload_search_analyzers")
+        rescue => e
+          raise Error, "Requires non-OSS version of Elasticsearch" if Searchkick.not_allowed_error?(e)
+          raise e
+        end
+      end
     end
 
     # queue
@@ -182,22 +209,35 @@ module Searchkick
 
     # reindex
 
-    def reindex(relation, method_name, scoped:, full: false, scope: nil, **options)
+    # note: this is designed to be used internally
+    # so it does not check object matches index class
+    def reindex(object, method_name: nil, full: false, **options)
+      if object.is_a?(Array)
+        # note: purposefully skip full
+        return reindex_records(object, method_name: method_name, **options)
+      end
+
+      if !object.respond_to?(:searchkick_klass)
+        raise Error, "Cannot reindex object"
+      end
+
+      scoped = Searchkick.relation?(object)
+      # call searchkick_klass for inheritance
+      relation = scoped ? object.all : Searchkick.scope(object.searchkick_klass).all
+
       refresh = options.fetch(:refresh, !scoped)
+      options.delete(:refresh)
 
-      if method_name
-        # update
-        import_scope(relation, method_name: method_name, scope: scope)
-        self.refresh if refresh
-        true
-      elsif scoped && !full
-        # reindex association
-        import_scope(relation, scope: scope)
+      if method_name || (scoped && !full)
+        mode = options.delete(:mode) || :inline
+        raise ArgumentError, "unsupported keywords: #{options.keys.map(&:inspect).join(", ")}" if options.any?
+
+        # import only
+        import_scope(relation, method_name: method_name, mode: mode)
         self.refresh if refresh
         true
       else
-        # full reindex
-        reindex_scope(relation, scope: scope, **options)
+        full_reindex(relation, **options)
       end
     end
 
@@ -209,15 +249,14 @@ module Searchkick
     end
 
     def import_scope(relation, **options)
-      bulk_indexer.import_scope(relation, **options)
+      relation_indexer.reindex(relation, **options)
     end
 
     def batches_left
-      bulk_indexer.batches_left
+      relation_indexer.batches_left
     end
 
-    # other
-
+    # private
     def klass_document_type(klass, ignore_type = false)
       @klass_document_type[[klass, ignore_type]] ||= begin
         if !ignore_type && klass.searchkick_klass.searchkick_options[:_type]
@@ -230,7 +269,7 @@ module Searchkick
       end
     end
 
-    # should not be public
+    # private
     def conversions_fields
       @conversions_fields ||= begin
         conversions = Array(options[:conversions])
@@ -238,10 +277,12 @@ module Searchkick
       end
     end
 
+    # private
     def suggest_fields
       @suggest_fields ||= Array(options[:suggest]).map(&:to_s)
     end
 
+    # private
     def locations_fields
       @locations_fields ||= begin
         locations = Array(options[:locations])
@@ -249,19 +290,54 @@ module Searchkick
       end
     end
 
+    # private
+    def uuid
+      index_settings["uuid"]
+    end
+
     protected
 
     def client
       Searchkick.client
     end
 
-    def bulk_indexer
-      @bulk_indexer ||= BulkIndexer.new(self)
+    def queue_index(records)
+      Searchkick.indexer.queue(records.map { |r| RecordData.new(self, r).index_data })
+    end
+
+    def queue_delete(records)
+      Searchkick.indexer.queue(records.reject { |r| r.id.blank? }.map { |r| RecordData.new(self, r).delete_data })
+    end
+
+    def queue_update(records, method_name)
+      Searchkick.indexer.queue(records.map { |r| RecordData.new(self, r).update_data(method_name) })
+    end
+
+    def relation_indexer
+      @relation_indexer ||= RelationIndexer.new(self)
+    end
+
+    def index_settings
+      settings.values.first["settings"]["index"]
+    end
+
+    def import_before_promotion(index, relation, **import_options)
+      index.import_scope(relation, **import_options)
+    end
+
+    def reindex_records(object, mode: nil, refresh: false, **options)
+      mode ||= Searchkick.callbacks_value || @options[:callbacks] || true
+      mode = :inline if mode == :bulk
+
+      result = RecordIndexer.new(self).reindex(object, mode: mode, full: false, **options)
+      self.refresh if refresh
+      result
     end
 
     # https://gist.github.com/jarosan/3124884
     # http://www.elasticsearch.org/blog/changing-mapping-with-zero-downtime/
-    def reindex_scope(relation, import: true, resume: false, retain: false, async: false, refresh_interval: nil, scope: nil)
+    # TODO deprecate async in favor of mode: :async, wait: true/false
+    def full_reindex(relation, import: true, resume: false, retain: false, async: false, refresh_interval: nil, scope: nil)
       if resume
         index_name = all_indices.sort.last
         raise Searchkick::Error, "No index to resume" unless index_name
@@ -275,20 +351,22 @@ module Searchkick
       end
 
       import_options = {
-        resume: resume,
-        async: async,
+        mode: (async ? :async : :inline),
         full: true,
+        resume: resume,
         scope: scope
       }
 
+      uuid = index.uuid
+
       # check if alias exists
       alias_exists = alias_exists?
       if alias_exists
-        # import before promotion
-        index.import_scope(relation, **import_options) if import
+        import_before_promotion(index, relation, **import_options) if import
 
         # get existing indices to remove
         unless async
+          check_uuid(uuid, index.uuid)
           promote(index.name, update_refresh_interval: !refresh_interval.nil?)
           clean_indices unless retain
         end
@@ -313,6 +391,7 @@ module Searchkick
           # already promoted if alias didn't exist
           if alias_exists
             puts "Jobs complete. Promoting..."
+            check_uuid(uuid, index.uuid)
             promote(index.name, update_refresh_interval: !refresh_interval.nil?)
           end
           clean_indices unless retain
@@ -324,12 +403,51 @@ module Searchkick
         index.refresh
         true
       end
-    rescue Elasticsearch::Transport::Transport::Errors::BadRequest => e
-      if e.message.include?("No handler for type [text]")
-        raise UnsupportedVersionError, "This version of Searchkick requires Elasticsearch 5 or greater"
+    rescue => e
+      if Searchkick.transport_error?(e) && e.message.include?("No handler for type [text]")
+        raise UnsupportedVersionError
       end
 
       raise e
     end
+
+    # safety check
+    # still a chance for race condition since its called before promotion
+    # ideal is for user to disable automatic index creation
+    # https://www.elastic.co/guide/en/elasticsearch/reference/current/docs-index_.html#index-creation
+    def check_uuid(old_uuid, new_uuid)
+      if old_uuid != new_uuid
+        raise Searchkick::Error, "Safety check failed - only run one Model.reindex per model at a time"
+      end
+    end
+
+    def notify(record, name)
+      if Searchkick.callbacks_value == :bulk
+        yield
+      else
+        name = "#{record.class.searchkick_klass.name} #{name}" if record && record.class.searchkick_klass
+        event = {
+          name: name,
+          id: search_id(record)
+        }
+        ActiveSupport::Notifications.instrument("request.searchkick", event) do
+          yield
+        end
+      end
+    end
+
+    def notify_bulk(records, name)
+      if Searchkick.callbacks_value == :bulk
+        yield
+      else
+        event = {
+          name: "#{records.first.class.searchkick_klass.name} #{name}",
+          count: records.size
+        }
+        ActiveSupport::Notifications.instrument("request.searchkick", event) do
+          yield
+        end
+      end
+    end
   end
 end

data/lib/searchkick/index_cache.rb

@@ -0,0 +1,30 @@
+module Searchkick
+  class IndexCache
+    def initialize(max_size: 20)
+      @data = {}
+      @mutex = Mutex.new
+      @max_size = max_size
+    end
+
+    # probably a better pattern for this
+    # but keep it simple
+    def fetch(name)
+      # thread-safe in MRI without mutex
+      # due to how context switching works
+      @mutex.synchronize do
+        if @data.key?(name)
+          @data[name]
+        else
+          @data.clear if @data.size >= @max_size
+          @data[name] = yield
+        end
+      end
+    end
+
+    def clear
+      @mutex.synchronize do
+        @data.clear
+      end
+    end
+  end
+end