fluent-plugin-kafka 0.15.3 → 0.16.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 730f14a0d24b98ac7df9683bc0836f826ff80def0ee61c833bc4c1e61bf435ac
-  data.tar.gz: 0a647554fb96f88ae75c8b872399545bfdb6b26d12247f3715c56dda2e628d57
+  metadata.gz: 7fcaf3e8fbb836ab8db3fa21003a713f9048de076097e2c95a9f30e5d1b05c08
+  data.tar.gz: 8437b2c9401238d811973422a65d2c9ee34bd8afc513f32412a22f93e03204a0
 SHA512:
-  metadata.gz: 802bef06a0cb0703e0e3c1d8c9581a96007b23a6b574d2c64f3bca362f6d4c531cbc4c9c17c78162705d655fc2e864af16182772a1db0ce2150c0e853b4f55ef
-  data.tar.gz: 0d3695bc1289b75cee1071fcf8e697250de9f0177b3facd69dd62e931a7b705fd93b24f886e683586d01f9daf9888d487ff6058da791d3779d09c076af854dd8
+  metadata.gz: 6d407e6f12dbafc6f5fad59ac6cbae4b13fa5f0a212cf82dba604f871afc286899290972b3afca5d1031888ee2f141c12d5d2e3d6b7324c7246317029149c491
+  data.tar.gz: b278441842361cc53836fce087e87e6d2d800bccbe9b307c2f0b39d3c4c357aae96275d2bfb808ab2455ebce692948d1241f4f991a5036473da1d31da699832b

data/.github/workflows/linux.yml

@@ -0,0 +1,26 @@
+name: linux
+on:
+  - push
+  - pull_request
+jobs:
+  build:
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby: [ '2.4', '2.5', '2.6', '2.7', '3.0' ]
+        os:
+          - ubuntu-latest
+    name: Ruby ${{ matrix.ruby }} unit testing on ${{ matrix.os }}
+    steps:
+    - uses: actions/checkout@v2
+    - uses: ruby/setup-ruby@v1
+      with:
+        ruby-version: ${{ matrix.ruby }}
+    - name: unit testing
+      env:
+        CI: true
+      run: |
+        gem install bundler rake
+        bundle install --jobs 4 --retry 3
+        bundle exec rake test

data/ChangeLog

@@ -1,3 +1,8 @@
+Release 0.16.0 - 2021/01/25
+
+	* input: Add `tag_source` and `record_tag_key` parameters for using record field as tag
+	* in_kafka_group: Use NumericParser for floating point
+
 Release 0.15.3 - 2020/12/08
 
 	* in_kafka: Fix `record_time_key` parameter not working

data/fluent-plugin-kafka.gemspec

@@ -13,7 +13,7 @@ Gem::Specification.new do |gem|
   gem.test_files    = gem.files.grep(%r{^(test|spec|features)/})
   gem.name          = "fluent-plugin-kafka"
   gem.require_paths = ["lib"]
-  gem.version       = '0.15.3'
+  gem.version       = '0.16.0'
   gem.required_ruby_version = ">= 2.1.0"
 
   gem.add_dependency "fluentd", [">= 0.10.58", "< 2"]
@@ -21,4 +21,5 @@ Gem::Specification.new do |gem|
   gem.add_dependency 'ruby-kafka', '>= 1.2.0', '< 2'
   gem.add_development_dependency "rake", ">= 0.9.2"
   gem.add_development_dependency "test-unit", ">= 3.0.8"
+  gem.add_development_dependency "webrick"
 end

data/lib/fluent/plugin/in_kafka.rb

@@ -31,6 +31,10 @@ class Fluent::KafkaInput < Fluent::Input
   config_param :add_suffix, :string, :default => nil,
                :desc => "tag suffix"
   config_param :add_offset_in_record, :bool, :default => false
+  config_param :tag_source, :enum, :list => [:topic, :record], :default => :topic,
+               :desc => "Source for the fluentd event tag"
+  config_param :record_tag_key, :string, :default => 'tag',
+               :desc => "Tag field when tag_source is 'record'"
 
   config_param :offset_zookeeper, :string, :default => nil
   config_param :offset_zk_root_node, :string, :default => '/fluent-plugin-kafka'
@@ -225,6 +229,8 @@ class Fluent::KafkaInput < Fluent::Input
         @kafka_message_key,
         @time_source,
         @record_time_key,
+        @tag_source,
+        @record_tag_key,
         opt)
     }
     @topic_watchers.each {|tw|
@@ -249,7 +255,7 @@ class Fluent::KafkaInput < Fluent::Input
   end
 
   class TopicWatcher < Coolio::TimerWatcher
-    def initialize(topic_entry, kafka, interval, parser, add_prefix, add_suffix, offset_manager, router, kafka_message_key, time_source, record_time_key, options={})
+    def initialize(topic_entry, kafka, interval, parser, add_prefix, add_suffix, offset_manager, router, kafka_message_key, time_source, record_time_key, tag_source, record_tag_key, options={})
       @topic_entry = topic_entry
       @kafka = kafka
       @callback = method(:consume)
@@ -262,6 +268,8 @@ class Fluent::KafkaInput < Fluent::Input
       @kafka_message_key = kafka_message_key
       @time_source = time_source
       @record_time_key = record_time_key
+      @tag_source = tag_source
+      @record_tag_key = record_tag_key
 
       @next_offset = @topic_entry.offset
       if @topic_entry.offset == -1 && offset_manager
@@ -298,6 +306,11 @@ class Fluent::KafkaInput < Fluent::Input
       messages.each { |msg|
         begin
           record = @parser.call(msg, @topic_entry)
+          if @tag_source == :record
+            tag = record[@record_tag_key]
+            tag = @add_prefix + "." + tag if @add_prefix
+            tag = tag + "." + @add_suffix if @add_suffix
+          end
           case @time_source
           when :kafka
             record_time = Fluent::EventTime.from_time(msg.create_time)

data/lib/fluent/plugin/in_kafka_group.rb

@@ -36,6 +36,10 @@ class Fluent::KafkaGroupInput < Fluent::Input
   config_param :get_kafka_client_log, :bool, :default => false
   config_param :time_format, :string, :default => nil,
                :desc => "Time format to be used to parse 'time' field."
+  config_param :tag_source, :enum, :list => [:topic, :record], :default => :topic,
+               :desc => "Source for the fluentd event tag"
+  config_param :record_tag_key, :string, :default => 'tag',
+               :desc => "Tag field when tag_source is 'record'"
   config_param :kafka_message_key, :string, :default => nil,
                :desc => "Set kafka's message key to this field"
   config_param :connect_timeout, :integer, :default => nil,
@@ -138,6 +142,10 @@ class Fluent::KafkaGroupInput < Fluent::Input
         @time_parser = Fluent::TextParser::TimeParser.new(@time_format)
       end
     end
+
+    if @time_source == :record && defined?(Fluent::NumericTimeParser)
+      @float_numeric_parse = Fluent::NumericTimeParser.new(:float)
+    end
   end
 
   def setup_parser(conf)
@@ -244,49 +252,104 @@ class Fluent::KafkaGroupInput < Fluent::Input
     end
   end
 
+  def process_batch_with_record_tag(batch)
+    es = {} 
+    batch.messages.each { |msg|
+      begin
+        record = @parser_proc.call(msg)
+        tag = record[@record_tag_key]
+        tag = @add_prefix + "." + tag if @add_prefix
+        tag = tag + "." + @add_suffix if @add_suffix
+        es[tag] ||= Fluent::MultiEventStream.new
+        case @time_source
+        when :kafka
+          record_time = Fluent::EventTime.from_time(msg.create_time)
+        when :now
+          record_time = Fluent::Engine.now
+        when :record
+          if @time_format
+            record_time = @time_parser.parse(record[@record_time_key].to_s)
+          else
+            record_time = record[@record_time_key]
+          end
+        else
+          log.fatal "BUG: invalid time_source: #{@time_source}"
+        end
+        if @kafka_message_key
+          record[@kafka_message_key] = msg.key
+        end
+        if @add_headers
+          msg.headers.each_pair { |k, v|
+            record[k] = v
+          }
+        end
+        es[tag].add(record_time, record)
+      rescue => e
+        log.warn "parser error in #{batch.topic}/#{batch.partition}", :error => e.to_s, :value => msg.value, :offset => msg.offset
+        log.debug_backtrace
+      end
+    }
+
+    unless es.empty?
+      es.each { |tag,es|
+        emit_events(tag, es)
+      }
+    end
+  end
+
+  def process_batch(batch)
+    es = Fluent::MultiEventStream.new
+    tag = batch.topic
+    tag = @add_prefix + "." + tag if @add_prefix
+    tag = tag + "." + @add_suffix if @add_suffix
+
+    batch.messages.each { |msg|
+      begin
+        record = @parser_proc.call(msg)
+        case @time_source
+        when :kafka
+          record_time = Fluent::EventTime.from_time(msg.create_time)
+        when :now
+          record_time = Fluent::Engine.now
+        when :record
+          record_time = record[@record_time_key]
+
+          if @time_format
+            record_time = @time_parser.parse(record_time.to_s)
+          elsif record_time.is_a?(Float) && @float_numeric_parse
+            record_time = @float_numeric_parse.parse(record_time)
+          end
+        else
+          log.fatal "BUG: invalid time_source: #{@time_source}"
+        end
+        if @kafka_message_key
+          record[@kafka_message_key] = msg.key
+        end
+        if @add_headers
+          msg.headers.each_pair { |k, v|
+            record[k] = v
+          }
+        end
+        es.add(record_time, record)
+      rescue => e
+        log.warn "parser error in #{batch.topic}/#{batch.partition}", :error => e.to_s, :value => msg.value, :offset => msg.offset
+        log.debug_backtrace
+      end
+    }
+
+    unless es.empty?
+      emit_events(tag, es)
+    end
+  end
+
   def run
     while @consumer
       begin
         @consumer.each_batch(@fetch_opts) { |batch|
-          es = Fluent::MultiEventStream.new
-          tag = batch.topic
-          tag = @add_prefix + "." + tag if @add_prefix
-          tag = tag + "." + @add_suffix if @add_suffix
-
-          batch.messages.each { |msg|
-            begin
-              record = @parser_proc.call(msg)
-              case @time_source
-              when :kafka
-                record_time = Fluent::EventTime.from_time(msg.create_time)
-              when :now
-                record_time = Fluent::Engine.now
-              when :record
-                if @time_format
-                  record_time = @time_parser.parse(record[@record_time_key].to_s)
-                else
-                  record_time = record[@record_time_key]
-                end
-              else
-                log.fatal "BUG: invalid time_source: #{@time_source}"
-              end
-              if @kafka_message_key
-                record[@kafka_message_key] = msg.key
-              end
-              if @add_headers
-                msg.headers.each_pair { |k, v|
-                  record[k] = v
-                }
-              end
-              es.add(record_time, record)
-            rescue => e
-              log.warn "parser error in #{batch.topic}/#{batch.partition}", :error => e.to_s, :value => msg.value, :offset => msg.offset
-              log.debug_backtrace
-            end
-          }
-
-          unless es.empty?
-            emit_events(tag, es)
+          if @tag_source == :record
+            process_batch_with_record_tag(batch)
+          else
+            process_batch(batch) 
           end
         }
       rescue ForShutdown

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: fluent-plugin-kafka
 version: !ruby/object:Gem::Version
-  version: 0.15.3
+  version: 0.16.0
 platform: ruby
 authors:
 - Hidemasa Togashi
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-12-08 00:00:00.000000000 Z
+date: 2021-01-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: fluentd
@@ -93,6 +93,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 3.0.8
+- !ruby/object:Gem::Dependency
+  name: webrick
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 description: Fluentd plugin for Apache Kafka > 0.8
 email:
 - togachiro@gmail.com
@@ -101,6 +115,7 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
+- ".github/workflows/linux.yml"
 - ".gitignore"
 - ".travis.yml"
 - ChangeLog