searchkick 4.0.2 → 4.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: cb0777c063adddb2c339761f92c7ffaf3646daf869b7360dafe662ce0b38056b
-  data.tar.gz: '062070815d5b78a2cb3b7c4f0dc9408730e8facd72ce223d25a6bafe9bd46d12'
+  metadata.gz: 011efe75e8dd9c13d54a7278a8e3eb8e9d4e8ef3bca601842c4f022799c3d0f4
+  data.tar.gz: bc18d8b082717fe5ebbd8891c9eda584bd4f288af3898c19495ccdc9db9f04f7
 SHA512:
-  metadata.gz: b3c08562e9a52dc950ff74ebb532beac936367e75c3a2e794079217caed899c566df50265e602c75d13e44eafd03c09821cfb10a88398d0897e5b0a85c76cf7d
-  data.tar.gz: c211a672924cec7627d0bf9b1e700c8ef6d037a745a7867e39b3cf8302f8f84d163d6911a893b0f8adf4073bf3f624cb517224e96522d0b5e209b081300a12f7
+  metadata.gz: 03dc4717ee01f14c16725e98125865c1d3331fea5cb4086af979e4b06bdb6576e427ee06a1a478a5792b857e4c087c66e43db0d0799691bccc5782ce4a3100ca
+  data.tar.gz: b5ba0becde65ea65ca5e8ad7c4122119fbc9580ee445575ad90bae03ba868b09199eeae405a7203d91bda495289d37435dd9ec145672ec7e6141720e721123cf

data/CHANGELOG.md

@@ -1,3 +1,10 @@
+## 4.1.0
+
+- Added `like` operator
+- Added `exists` operator
+- Added warnings for certain regular expressions
+- Fixed anchored regular expressions
+
 ## 4.0.2
 
 - Added block form of `scroll`

data/README.md

@@ -10,7 +10,7 @@ Searchkick handles:
 - special characters - `jalapeno` matches `jalapeño`
 - extra whitespace - `dishwasher` matches `dish washer`
 - misspellings - `zuchini` matches `zucchini`
-- custom synonyms - `qtip` matches `cotton swab`
+- custom synonyms - `pop` matches `soda`
 
 Plus:
 
@@ -99,14 +99,19 @@ Where
 
 ```ruby
 where: {
-  expires_at: {gt: Time.now}, # lt, gte, lte also available
-  orders_count: 1..10,        # equivalent to {gte: 1, lte: 10}
-  aisle_id: [25, 30],         # in
-  store_id: {not: 2},         # not
-  aisle_id: {not: [25, 30]},  # not in
-  user_ids: {all: [1, 3]},    # all elements in array
-  category: /frozen .+/,      # regexp
-  _or: [{in_stock: true}, {backordered: true}]
+  expires_at: {gt: Time.now},   # lt, gte, lte also available
+  orders_count: 1..10,          # equivalent to {gte: 1, lte: 10}
+  aisle_id: [25, 30],           # in
+  store_id: {not: 2},           # not
+  aisle_id: {not: [25, 30]},    # not in
+  user_ids: {all: [1, 3]},      # all elements in array
+  category: {like: "%frozen%"}, # like
+  category: /frozen .+/,        # regexp
+  category: {prefix: "frozen"}, # prefix
+  store_id: {exists: true},     # exists
+  _or: [{in_stock: true}, {backordered: true}],
+  _and: [{in_stock: true}, {backordered: true}],
+  _not: {store_id: 1}           # negate a condition
 }
 ```
 
@@ -313,7 +318,7 @@ A few languages require plugins:
 
 ```ruby
 class Product < ApplicationRecord
-  searchkick synonyms: [["burger", "hamburger"], ["sneakers", "shoes"]]
+  searchkick synonyms: [["pop", "soda"], ["burger", "hamburger"]]
 end
 ```
 
@@ -631,7 +636,7 @@ Autocomplete predicts what a user will type, making the search experience faster
 
 ![Autocomplete](https://gist.github.com/ankane/b6988db2802aca68a589b31e41b44195/raw/40febe948427e5bc53ec4e5dc248822855fef76f/autocomplete.png)
 
-**Note:** To autocomplete on general categories (like `cereal` rather than product names), check out [Autosuggest](https://github.com/ankane/autosuggest).
+**Note:** To autocomplete on search terms rather than results, check out [Autosuggest](https://github.com/ankane/autosuggest).
 
 **Note 2:** If you only have a few thousand records, don’t use Searchkick for autocomplete. It’s *much* faster to load all records into JavaScript and autocomplete there (eliminates network requests).
 
@@ -1161,7 +1166,7 @@ If you run into issues on Windows, check out [this post](https://www.rastating.c
 
 ### Searchable Fields
 
-By default, all string fields are searchable (can be used in `fields` option). Speed up indexing and reduce index size by only making some fields searchable. This disables the `_all` field unless it’s listed.
+By default, all string fields are searchable (can be used in `fields` option). Speed up indexing and reduce index size by only making some fields searchable.
 
 ```ruby
 class Product < ApplicationRecord

data/lib/searchkick.rb

@@ -49,7 +49,7 @@ module Searchkick
 
   def self.client
     @client ||= begin
-      require "typhoeus/adapters/faraday" if defined?(Typhoeus)
+      require "typhoeus/adapters/faraday" if defined?(Typhoeus) && Gem::Version.new(Faraday::VERSION) < Gem::Version.new("0.14.0")
 
       Elasticsearch::Client.new({
         url: ENV["ELASTICSEARCH_URL"],
@@ -192,6 +192,10 @@ module Searchkick
     end
   end
 
+  def self.warn(message)
+    super("[searchkick] WARNING: #{message}")
+  end
+
   # private
   def self.load_records(records, ids)
     records =

data/lib/searchkick/bulk_indexer.rb

@@ -61,7 +61,7 @@ module Searchkick
       if records.any?
         if async
           Searchkick::BulkReindexJob.perform_later(
-            class_name: records.first.class.name,
+            class_name: records.first.class.searchkick_options[:class_name],
             record_ids: records.map(&:id),
             index_name: index.name,
             method_name: method_name ? method_name.to_s : nil
@@ -140,7 +140,7 @@ module Searchkick
     def bulk_reindex_job(scope, batch_id, options)
       Searchkick.with_redis { |r| r.sadd(batches_key, batch_id) }
       Searchkick::BulkReindexJob.perform_later({
-        class_name: scope.model_name.name,
+        class_name: scope.searchkick_options[:class_name],
         index_name: index.name,
         batch_id: batch_id
       }.merge(options))

data/lib/searchkick/index.rb

@@ -259,6 +259,10 @@ module Searchkick
       @bulk_indexer ||= BulkIndexer.new(self)
     end
 
+    def import_before_promotion(index, relation, **import_options)
+      index.import_scope(relation, **import_options)
+    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)
@@ -284,8 +288,7 @@ module Searchkick
       # 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

data/lib/searchkick/index_options.rb

@@ -114,7 +114,7 @@ module Searchkick
                 type: "shingle",
                 token_separator: ""
               },
-              # lucky find http://web.archiveorange.com/archive/v/AAfXfQ17f57FcRINsof7
+              # lucky find https://web.archiveorange.com/archive/v/AAfXfQ17f57FcRINsof7
               searchkick_search_shingle: {
                 type: "shingle",
                 token_separator: "",

data/lib/searchkick/model.rb

@@ -15,6 +15,7 @@ module Searchkick
       Searchkick.models << self
 
       options[:_type] ||= -> { searchkick_index.klass_document_type(self, true) }
+      options[:class_name] = model_name.name
 
       callbacks = options.key?(:callbacks) ? options[:callbacks] : :inline
       unless [:inline, true, false, :async, :queue].include?(callbacks)
@@ -44,8 +45,8 @@ module Searchkick
           end
           alias_method Searchkick.search_method_name, :searchkick_search if Searchkick.search_method_name
 
-          def searchkick_index
-            index = class_variable_get(:@@searchkick_index)
+          def searchkick_index(name: nil)
+            index = name || class_variable_get(:@@searchkick_index)
             index = index.call if index.respond_to?(:call)
             index_cache = class_variable_get(:@@searchkick_index_cache)
             index_cache[index] ||= Searchkick::Index.new(index, searchkick_options)

data/lib/searchkick/process_batch_job.rb

@@ -2,7 +2,7 @@ module Searchkick
   class ProcessBatchJob < ActiveJob::Base
     queue_as { Searchkick.queue_name }
 
-    def perform(class_name:, record_ids:)
+    def perform(class_name:, record_ids:, index_name: nil)
       # separate routing from id
       routing = Hash[record_ids.map { |r| r.split(/(?<!\|)\|(?!\|)/, 2).map { |v| v.gsub("||", "|") } }]
       record_ids = routing.keys
@@ -26,7 +26,7 @@ module Searchkick
       end
 
       # bulk reindex
-      index = klass.searchkick_index
+      index = klass.searchkick_index(name: index_name)
       Searchkick.callbacks(:bulk) do
         index.bulk_index(records) if records.any?
         index.bulk_delete(delete_records) if delete_records.any?

data/lib/searchkick/process_queue_job.rb

@@ -2,21 +2,29 @@ module Searchkick
   class ProcessQueueJob < ActiveJob::Base
     queue_as { Searchkick.queue_name }
 
-    def perform(class_name:)
+    def perform(class_name:, index_name: nil, inline: false)
       model = class_name.constantize
+      limit = model.searchkick_options[:batch_size] || 1000
 
-      limit = model.searchkick_index.options[:batch_size] || 1000
-      record_ids = model.searchkick_index.reindex_queue.reserve(limit: limit)
-      if record_ids.any?
-        Searchkick::ProcessBatchJob.perform_later(
-          class_name: model.name,
-          record_ids: record_ids
-        )
-        # TODO when moving to reliable queuing, mark as complete
+      loop do
+        record_ids = model.searchkick_index(name: index_name).reindex_queue.reserve(limit: limit)
+        if record_ids.any?
+          batch_options = {
+            class_name: class_name,
+            record_ids: record_ids,
+            index_name: index_name
+          }
 
-        if record_ids.size == limit
-          Searchkick::ProcessQueueJob.perform_later(class_name: class_name)
+          if inline
+            # use new.perform to avoid excessive logging
+            Searchkick::ProcessBatchJob.new.perform(**batch_options)
+          else
+            Searchkick::ProcessBatchJob.perform_later(**batch_options)
+          end
+
+          # TODO when moving to reliable queuing, mark as complete
         end
+        break unless record_ids.size == limit
       end
     end
   end

data/lib/searchkick/query.rb

@@ -137,6 +137,7 @@ module Searchkick
       }
 
       if options[:debug]
+        # can remove when minimum Ruby version is 2.5
         require "pp"
 
         puts "Searchkick Version: #{Searchkick::VERSION}"
@@ -432,7 +433,7 @@ module Searchkick
 
         models = Array(options[:models])
         if models.any? { |m| m != m.searchkick_klass }
-          warn "[searchkick] WARNING: Passing child models to models option throws off hits and pagination - use type option instead"
+          Searchkick.warn("Passing child models to models option throws off hits and pagination - use type option instead")
 
           # uncomment once aliases are supported with _index
           # see https://github.com/elastic/elasticsearch/issues/23306
@@ -873,6 +874,8 @@ module Searchkick
           filters << {bool: {must_not: where_filters(value)}}
         elsif field == :_and
           filters << {bool: {must: value.map { |or_statement| {bool: {filter: where_filters(or_statement)}} }}}
+        # elsif field == :_script
+        #   filters << {script: {script: {source: value, lang: "painless"}}}
         else
           # expand ranges
           if value.is_a?(Range)
@@ -933,6 +936,14 @@ module Searchkick
                     }
                   }
                 }
+              when :like
+                # based on Postgres
+                # https://www.postgresql.org/docs/current/functions-matching.html
+                # % matches zero or more characters
+                # _ matches one character
+                # \ is escape character
+                regex = Regexp.escape(op_value).gsub(/(?<!\\)%/, ".*").gsub(/(?<!\\)_/, ".").gsub("\\%", "%").gsub("\\_", "_")
+                filters << {regexp: {field => {value: regex}}}
               when :prefix
                 filters << {prefix: {field => op_value}}
               when :regexp # support for regexp queries without using a regexp ruby object
@@ -945,6 +956,8 @@ module Searchkick
                 end
               when :in
                 filters << term_filters(field, op_value)
+              when :exists
+                filters << {exists: {field: field}}
               else
                 range_query =
                   case op
@@ -985,7 +998,32 @@ module Searchkick
       elsif value.nil?
         {bool: {must_not: {exists: {field: field}}}}
       elsif value.is_a?(Regexp)
-        {regexp: {field => {value: value.source, flags: "NONE"}}}
+        if value.casefold?
+          Searchkick.warn("Case-insensitive flag does not work with Elasticsearch")
+        end
+
+        source = value.source
+        unless source.start_with?("\\A") && source.end_with?("\\z")
+          # https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-regexp-query.html
+          Searchkick.warn("Regular expressions are always anchored in Elasticsearch")
+        end
+
+        # TODO handle other anchor characters, like ^, $, \Z
+        if source.start_with?("\\A")
+          source = source[2..-1]
+        else
+          # TODO uncomment in Searchkick 5
+          # source = ".*#{source}"
+        end
+
+        if source.end_with?("\\z")
+          source = source[0..-3]
+        else
+          # TODO uncomment in Searchkick 5
+          # source = "#{source}.*"
+        end
+
+        {regexp: {field => {value: source, flags: "NONE"}}}
       else
         {term: {field => value}}
       end

data/lib/searchkick/results.rb

@@ -62,7 +62,7 @@ module Searchkick
             end
 
           if missing_ids.any?
-            warn "[searchkick] WARNING: Records in search index do not exist in database: #{missing_ids.join(", ")}"
+            Searchkick.warn("Records in search index do not exist in database: #{missing_ids.join(", ")}")
           end
 
           results

data/lib/searchkick/version.rb

@@ -1,3 +1,3 @@
 module Searchkick
-  VERSION = "4.0.2"
+  VERSION = "4.1.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: searchkick
 version: !ruby/object:Gem::Version
-  version: 4.0.2
+  version: 4.1.0
 platform: ruby
 authors:
 - Andrew Kane
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-06-05 00:00:00.000000000 Z
+date: 2019-08-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activemodel
@@ -145,7 +145,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.0.3
+rubygems_version: 3.0.4
 signing_key: 
 specification_version: 4
 summary: Intelligent search made easy with Rails and Elasticsearch