fluent-plugin-woothee 0.1.1 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: a023e5af86e6a0c8e61e478271210b67f49a0cb9
-  data.tar.gz: 47c03e7e2554c15a2839b4e78f207cce867ebdd2
+  metadata.gz: 7d68a5e296eb04df8c7a1234cf96db2a097c52f8
+  data.tar.gz: cd1d52ccf24f32dc9e3cb5f47c774e654da3ed17
 SHA512:
-  metadata.gz: 4d4ec145c669b50764a57b546e2697d25237789c0e5e9339f6e35c9908db5bfcfd99cfe55225e9b305e80464e5becdbb1fabbffff21d5ae8c936ef66f75f3503
-  data.tar.gz: 629c34ed3cc1c175df26a3f06409fd1c0c3923163f0f7673b72a1fb967dc14db38f16ee3fb675b85320bcfb2d49f319742e5bd02457b50444d92348d21bcfa3b
+  metadata.gz: 06a22c57d945e94c124a64a066fd202784d489f5e5b46d55277c1289edacf404a018a9002ab93758c5df2c8c49cc1f1838159110aefbe1884e70362bb75bf0b6
+  data.tar.gz: fce1881e0f0c538e48f4d86de8f07db25df84445381a086927e03c96a67f477fd7e7518d18c452ab5ccc4758edec4778a6869929b442529cb1db9c6a7f754fb3

data/README.md

@@ -69,6 +69,65 @@ If you want to drop __almost__ all of messages with crawler's user-agent, and no
 'fluent-plugin-woothee' uses 'Woothee.is_crawler' of woothee with this configuration, fast and incomplete method to judge user-agent is crawler or not.
 If you want to drop all of crawlers completely, specify 'type woothee' and 'drop_categories crawler'.
 
+## WootheeFilter
+
+This is filter version of 'fluent-plugin-woothee'.
+Note that this filter version does not have rewrite tag functionality.
+
+## Configuration
+
+To add woothee parser result into filtered messages:
+
+    <filter input.**>
+      type woothee
+      key_name agent
+      merge_agent_info yes
+    </filter>
+
+Filtered messages with non-modified tag has attributes like 'agent\_name', 'agent\_category' and 'agent\_os' from woothee parser result. If you want to change attribute names, or want to merge more attributes of browser vendor and its version, write configurations as below:
+
+    <filter input.**>
+      type woothee
+      key_name agent
+      merge_agent_info yes
+      out_key_name ua_name
+      out_key_category ua_category
+      out_key_os ua_os
+      out_key_os_version ua_os_version
+      out_key_version ua_version
+      out_key_vendor ua_vendor
+    </filter>
+
+To filter messages with specified user-agent categories (and merge woothee parser result), configure like this:
+
+    <filter input.**>
+      type woothee
+      key_name agent
+      filter_categories pc,smartphone,mobilephone,appliance
+      merge_agent_info yes
+    </filter>
+
+Or, you can specify categories to drop (and not to merge woothee result):
+
+    <filter input.**>
+      type woothee
+      key_name agent
+      drop_categories crawler
+      merge_agent_info false # default
+    </filter>
+
+### Fast Crawler Filter
+
+If you want to drop __almost__ all of messages with crawler's user-agent, and not to merge woothee result, you just specify plugin type:
+
+    <filter input.**>
+      type woothee_fast_crawler_filter
+      key_name useragent
+    </filter>
+
+'fluent-plugin-woothee' uses 'Woothee.is_crawler' of woothee with this configuration, fast and incomplete method to judge user-agent is crawler or not.
+If you want to drop all of crawlers completely, specify 'type woothee' and 'drop_categories crawler'.
+
 ## TODO
 
 * patches welcome!

data/fluent-plugin-woothee.gemspec

@@ -2,13 +2,13 @@
 
 Gem::Specification.new do |gem|
   gem.name          = "fluent-plugin-woothee"
-  gem.version       = "0.1.1"
+  gem.version       = "0.2.0"
   gem.authors       = ["TAGOMORI Satoshi"]
   gem.email         = ["tagomoris@gmail.com"]
   gem.description   = %q{parsing by Project Woothee. See https://github.com/woothee/woothee }
   gem.summary       = %q{Fluentd plugin to parse UserAgent strings with woothee parser. It adds device information or filter records with specific device types.}
   gem.homepage      = "https://github.com/tagomoris/fluent-plugin-woothee"
-  gem.license       = "APLv2"
+  gem.license       = "Apache-2.0"
 
   gem.files         = `git ls-files`.split($\)
   gem.executables   = gem.files.grep(%r{^bin/}).map{ |f| File.basename(f) }

data/lib/fluent/plugin/filter_woothee.rb

@@ -0,0 +1,107 @@
+class Fluent::WootheeFilter < Fluent::Filter
+  Fluent::Plugin.register_filter('woothee', self)
+  Fluent::Plugin.register_filter('woothee_fast_crawler_filter', self)
+
+  config_param :fast_crawler_filter_mode, :bool, :default => false
+
+  config_param :key_name, :string
+
+  config_param :filter_categories, :default => [] do |val|
+    val.split(',').map(&:to_sym)
+  end
+  config_param :drop_categories, :default => [] do |val|
+    val.split(',').map(&:to_sym)
+  end
+  attr_accessor :mode
+
+  config_param :merge_agent_info, :bool, :default => false
+  config_param :out_key_name, :string, :default => 'agent_name'
+  config_param :out_key_category, :string, :default => 'agent_category'
+  config_param :out_key_os, :string, :default => 'agent_os'
+  config_param :out_key_os_version, :string, :default => nil # supress output
+  config_param :out_key_version, :string, :default => nil # supress output
+  config_param :out_key_vendor, :string, :default => nil # supress output
+
+  def initialize
+    super
+    require 'woothee'
+  end
+
+  def configure(conf)
+    super
+
+    if conf['type'] == 'woothee_fast_crawler_filter' or @fast_crawler_filter_mode
+      @fast_crawler_filter_mode = true
+
+      if @filter_categories.size > 0 or @drop_categories.size > 0 or @merge_agent_info
+        raise Fluent::ConfigError, "fast_crawler_filter cannot be specified with filter/drop/merge options"
+      end
+
+      return
+    end
+
+    if @filter_categories.size > 0 and @drop_categories.size > 0
+      raise Fluent::ConfigError, "both of 'filter' and 'drop' categories specified"
+    elsif @filter_categories.size > 0
+      unless @filter_categories.reduce(true){|r,i| r and Woothee::CATEGORY_LIST.include?(i)}
+        raise Fluent::ConfigError, "filter_categories has invalid category name"
+      end
+      @mode = :filter
+    elsif @drop_categories.size > 0
+      unless @drop_categories.reduce(true){|r,i| r and Woothee::CATEGORY_LIST.include?(i)}
+        raise Fluent::ConfigError, "drop_categories has invalid category name"
+      end
+      @mode = :drop
+    else
+      @mode = :through
+    end
+
+    if @mode == :through and not @merge_agent_info
+      raise Fluent::ConfigError, "configured not to do nothing (not to do either filter/drop nor addition of parser result)"
+    end
+  end
+
+  def fast_crawler_filter_stream(tag, es)
+    new_es = Fluent::MultiEventStream.new
+
+    es.each do |time,record|
+      unless Woothee.is_crawler(record[@key_name] || '')
+        new_es.add(time, record.dup)
+      end
+    end
+    new_es
+  end
+
+  def normal_filter_stream(tag, es)
+    new_es = Fluent::MultiEventStream.new
+
+    es.each do |time,record|
+      parsed = Woothee.parse(record[@key_name] || '')
+
+      category = parsed[Woothee::ATTRIBUTE_CATEGORY]
+      next if @mode == :filter and not @filter_categories.include?(category)
+      next if @mode == :drop and @drop_categories.include?(category)
+
+      if @merge_agent_info
+        record = record.merge({
+            @out_key_name => parsed[Woothee::ATTRIBUTE_NAME],
+            @out_key_category => parsed[Woothee::ATTRIBUTE_CATEGORY].to_s,
+            @out_key_os => parsed[Woothee::ATTRIBUTE_OS]
+          })
+        record[@out_key_os_version] = parsed[Woothee::ATTRIBUTE_OS_VERSION] if @out_key_os_version
+        record[@out_key_version] = parsed[Woothee::ATTRIBUTE_VERSION] if @out_key_version
+        record[@out_key_vendor] = parsed[Woothee::ATTRIBUTE_VENDOR] if @out_key_vendor
+      end
+      new_es.add(time, record.dup)
+    end
+    new_es
+  end
+
+  def filter_stream(tag, es)
+    if @fast_crawler_filter_mode
+      fast_crawler_filter_stream(tag, es)
+    else
+      normal_filter_stream(tag, es)
+    end
+  end
+end if defined?(Fluent::Filter)

data/lib/fluent/plugin/out_woothee.rb

@@ -2,6 +2,11 @@ class Fluent::WootheeOutput < Fluent::Output
   Fluent::Plugin.register_output('woothee', self)
   Fluent::Plugin.register_output('woothee_fast_crawler_filter', self)
 
+  # Define `router` method of v0.12 to support v0.10 or earlier
+  unless method_defined?(:router)
+    define_method("router") { Fluent::Engine }
+  end
+
   config_param :tag, :string, :default => nil
   config_param :remove_prefix, :string, :default => nil
   config_param :add_prefix, :string, :default => nil
@@ -103,7 +108,7 @@ class Fluent::WootheeOutput < Fluent::Output
   def fast_crawler_filter_emit(tag, es)
     es.each do |time,record|
       unless Woothee.is_crawler(record[@key_name] || '')
-        Fluent::Engine.emit(tag, time, record)
+        router.emit(tag, time, record)
       end
     end
   end
@@ -126,7 +131,7 @@ class Fluent::WootheeOutput < Fluent::Output
         record[@out_key_version] = parsed[Woothee::ATTRIBUTE_VERSION] if @out_key_version
         record[@out_key_vendor] = parsed[Woothee::ATTRIBUTE_VENDOR] if @out_key_vendor
       end
-      Fluent::Engine.emit(tag, time, record)
+      router.emit(tag, time, record)
     end
   end
 

data/test/helper.rb

@@ -23,6 +23,7 @@ unless ENV.has_key?('VERBOSE')
 end
 
 require 'fluent/plugin/out_woothee'
+require 'fluent/plugin/filter_woothee'
 
 class Test::Unit::TestCase
 end

data/test/plugin/test_filter_woothee.rb

@@ -0,0 +1,338 @@
+require 'helper'
+
+class Fluent::WootheeFilterTest < Test::Unit::TestCase
+  # fast crawler filter
+  CONFIG0 = %[
+type woothee_fast_crawler_filter
+key_name useragent
+]
+
+  # through & merge
+  CONFIG1 = %[
+type woothee
+key_name agent
+merge_agent_info yes
+]
+
+  # filter & merge
+  CONFIG2 = %[
+type woothee
+key_name agent
+filter_categories pc,smartphone,mobilephone,appliance
+merge_agent_info yes
+out_key_name ua_name
+out_key_category ua_category
+out_key_os ua_os
+out_key_os_version ua_os_version
+out_key_version ua_version
+out_key_vendor ua_vendor
+]
+
+  # drop & non-merge
+  CONFIG3 = %[
+type woothee
+key_name user_agent
+drop_categories crawler,misc
+]
+
+  def setup
+    omit("Use fluentd v0.12 or later") unless defined?(Fluent::Filter)
+
+    Fluent::Test.setup
+  end
+
+  def create_driver(conf=CONFIG1,tag='test')
+    Fluent::Test::FilterTestDriver.new(Fluent::WootheeFilter, tag).configure(conf)
+  end
+
+  class TestConfigure < self
+    def test_fast_crawer_filter
+      d = create_driver CONFIG0
+      assert_equal true, d.instance.fast_crawler_filter_mode
+      assert_equal 'useragent', d.instance.key_name
+    end
+
+    def test_through_and_merge
+      d = create_driver CONFIG1
+      assert_equal false, d.instance.fast_crawler_filter_mode
+      assert_equal 'agent', d.instance.key_name
+
+      assert_equal 0, d.instance.filter_categories.size
+      assert_equal 0, d.instance.drop_categories.size
+      assert_equal :through, d.instance.mode
+
+      assert_equal true, d.instance.merge_agent_info
+      assert_equal 'agent_name', d.instance.out_key_name
+      assert_equal 'agent_category', d.instance.out_key_category
+      assert_equal 'agent_os', d.instance.out_key_os
+      assert_nil d.instance.out_key_version
+      assert_nil d.instance.out_key_vendor
+    end
+
+    def test_filter_and_merge
+      d = create_driver CONFIG2
+      assert_equal false, d.instance.fast_crawler_filter_mode
+      assert_equal 'agent', d.instance.key_name
+
+      assert_equal 4, d.instance.filter_categories.size
+      assert_equal [:pc,:smartphone,:mobilephone,:appliance], d.instance.filter_categories
+      assert_equal 0, d.instance.drop_categories.size
+      assert_equal :filter, d.instance.mode
+
+      assert_equal true, d.instance.merge_agent_info
+      assert_equal 'ua_name', d.instance.out_key_name
+      assert_equal 'ua_category', d.instance.out_key_category
+      assert_equal 'ua_os', d.instance.out_key_os
+      assert_equal 'ua_os_version', d.instance.out_key_os_version
+      assert_equal 'ua_version', d.instance.out_key_version
+      assert_equal 'ua_vendor', d.instance.out_key_vendor
+    end
+
+    def test_drop_and_non_merge
+      d = create_driver CONFIG3
+      assert_equal false, d.instance.fast_crawler_filter_mode
+      assert_equal 'user_agent', d.instance.key_name
+
+      assert_equal 0, d.instance.filter_categories.size
+      assert_equal 2, d.instance.drop_categories.size
+      assert_equal [:crawler,:misc], d.instance.drop_categories
+      assert_equal :drop, d.instance.mode
+
+      assert_equal false, d.instance.merge_agent_info
+    end
+  end
+
+  def test_filter_fast_crawler_filter_stream
+    d = create_driver CONFIG0
+    time = Time.parse('2012-07-20 16:19:00').to_i
+    d.run do
+      d.filter({'useragent' => 'Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)', 'value' => 1}, time)
+      d.filter({'useragent' => 'Mozilla/5.0 (compatible; Yahoo! Slurp; http://help.yahoo.com/help/us/ysearch/slurp)', 'value' => 2}, time)
+      d.filter({'useragent' => 'Mozilla/5.0 (iPad; U; CPU OS 4_3_2 like Mac OS X; ja-jp) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8H7 Safari/6533.18.5', 'value' => 3}, time)
+      d.filter({'useragent' => 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; Trident/4.0)', 'value' => 4}, time)
+      d.filter({'useragent' => 'Mozilla/5.0 (compatible; bingbot/2.0; +http://www.bing.com/bingbot.htm)', 'value' => 5}, time)
+      d.filter({'useragent' => 'Mozilla/5.0 (compatible; Rakutenbot/1.0; +http://dynamic.rakuten.co.jp/bot.html)', 'value' => 6}, time)
+      d.filter({'useragent' => 'Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_5_4; ja-jp) AppleWebKit/525.18 (KHTML, like Gecko) Version/3.1.2 Safari/525.20.1', 'value' => 7}, time)
+      d.filter({'useragent' => 'Yeti/1.0 (NHN Corp.; http://help.naver.com/robots/)', 'value' => 8}, time)
+    end
+
+    filtered = d.filtered_as_array
+    assert_equal 4, filtered.size
+
+    assert_equal 'test', filtered[0][0]
+    assert_equal time, filtered[0][1]
+    assert_equal 'Mozilla/5.0 (iPad; U; CPU OS 4_3_2 like Mac OS X; ja-jp) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8H7 Safari/6533.18.5', filtered[0][2]['useragent']
+    assert_equal 3, filtered[0][2]['value']
+    assert_equal 2, filtered[0][2].keys.size
+
+    assert_equal 4, filtered[1][2]['value']
+    assert_equal 6, filtered[2][2]['value']
+    assert_equal 7, filtered[3][2]['value']
+  end
+
+  # through & merge
+  def test_filter_through
+    d = create_driver(CONFIG1, 'test.message')
+    time = Time.parse('2012-07-20 16:40:30').to_i
+    d.run do
+      d.filter({'value' => 0, 'agent' => 'Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; x64; Trident/6.0)'}, time)
+      d.filter({'value' => 1, 'agent' => 'Mozilla/5.0 (Windows NT 6.0; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'}, time)
+      d.filter({'value' => 2, 'agent' => 'Mozilla/5.0 (Ubuntu; X11; Linux i686; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'}, time)
+      d.filter({'value' => 3, 'agent' => 'Mozilla/5.0 (Linux; U; Android 3.1; ja-jp; L-06C Build/HMJ37) AppleWebKit/534.13 (KHTML, like Gecko) Version/4.0 Safari/534.13'}, time)
+      d.filter({'value' => 4, 'agent' => 'DoCoMo/1.0/N505i/c20/TB/W24H12'}, time)
+      d.filter({'value' => 5, 'agent' => 'Mozilla/5.0 (PlayStation Vita 1.51) AppleWebKit/531.22.8 (KHTML, like Gecko) Silk/3.2'}, time)
+      d.filter({'value' => 6, 'agent' => 'Mozilla/5.0 (compatible; Google Desktop/5.9.1005.12335; http://desktop.google.com/)'}, time)
+      d.filter({'value' => 7, 'agent' => 'msnbot/1.1 (+http://search.msn.com/msnbot.htm)'}, time)
+    end
+
+    filtered = d.filtered_as_array
+    assert_equal 8, filtered.size
+    assert_equal 'test.message', filtered[0][0]
+    assert_equal time, filtered[0][1]
+
+    # 'agent' => 'Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; x64; Trident/6.0)'
+    m = filtered[0][2]
+    assert_equal 0, m['value']
+    assert_equal 'Internet Explorer', m['agent_name']
+    assert_equal 'pc', m['agent_category']
+    assert_equal 'Windows 8', m['agent_os']
+    assert_equal 5, m.keys.size
+
+    # 'agent' => 'Mozilla/5.0 (Windows NT 6.0; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'
+    m = filtered[1][2]
+    assert_equal 1, m['value']
+    assert_equal 'Firefox', m['agent_name']
+    assert_equal 'pc', m['agent_category']
+    assert_equal 'Windows Vista', m['agent_os']
+
+    # 'agent' => 'Mozilla/5.0 (Ubuntu; X11; Linux i686; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'
+    m = filtered[2][2]
+    assert_equal 2, m['value']
+    assert_equal 'Firefox', m['agent_name']
+    assert_equal 'pc', m['agent_category']
+    assert_equal 'Linux', m['agent_os']
+
+    # 'agent' => 'Mozilla/5.0 (Linux; U; Android 3.1; ja-jp; L-06C Build/HMJ37) AppleWebKit/534.13 (KHTML, like Gecko) Version/4.0 Safari/534.13'
+    m = filtered[3][2]
+    assert_equal 3, m['value']
+    assert_equal 'Safari', m['agent_name']
+    assert_equal 'smartphone', m['agent_category']
+    assert_equal 'Android', m['agent_os']
+
+    # 'agent' => 'DoCoMo/1.0/N505i/c20/TB/W24H12'
+    m = filtered[4][2]
+    assert_equal 4, m['value']
+    assert_equal 'docomo', m['agent_name']
+    assert_equal 'mobilephone', m['agent_category']
+    assert_equal 'docomo', m['agent_os']
+
+    # 'agent' => 'Mozilla/5.0 (PlayStation Vita 1.51) AppleWebKit/531.22.8 (KHTML, like Gecko) Silk/3.2'
+    m = filtered[5][2]
+    assert_equal 5, m['value']
+    assert_equal 'PlayStation Vita', m['agent_name']
+    assert_equal 'appliance', m['agent_category']
+    assert_equal 'PlayStation Vita', m['agent_os']
+
+    # 'agent' => 'Mozilla/5.0 (compatible; Google Desktop/5.9.1005.12335; http://desktop.google.com/)'
+    m = filtered[6][2]
+    assert_equal 6, m['value']
+    assert_equal 'Google Desktop', m['agent_name']
+    assert_equal 'misc', m['agent_category']
+    assert_equal 'UNKNOWN', m['agent_os']
+
+    # 'agent' => 'msnbot/1.1 (+http://search.msn.com/msnbot.htm)'
+    m = filtered[7][2]
+    assert_equal 7, m['value']
+    assert_equal 'msnbot', m['agent_name']
+    assert_equal 'crawler', m['agent_category']
+    assert_equal 'UNKNOWN', m['agent_os']
+  end
+
+  # filter & merge
+  def test_filter_stream
+    d = create_driver(CONFIG2, 'test.message')
+    time = Time.parse('2012-07-20 16:40:30').to_i
+    d.run do
+      d.filter({'value' => 0, 'agent' => 'Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; x64; Trident/6.0)'}, time)
+      d.filter({'value' => 1, 'agent' => 'Mozilla/5.0 (Windows NT 6.0; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'}, time)
+      d.filter({'value' => 2, 'agent' => 'Mozilla/5.0 (Ubuntu; X11; Linux i686; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'}, time)
+      d.filter({'value' => 3, 'agent' => 'Mozilla/5.0 (Linux; U; Android 3.1; ja-jp; L-06C Build/HMJ37) AppleWebKit/534.13 (KHTML, like Gecko) Version/4.0 Safari/534.13'}, time)
+      d.filter({'value' => 4, 'agent' => 'DoCoMo/1.0/N505i/c20/TB/W24H12'}, time)
+      d.filter({'value' => 5, 'agent' => 'Mozilla/5.0 (PlayStation Vita 1.51) AppleWebKit/531.22.8 (KHTML, like Gecko) Silk/3.2'}, time)
+      d.filter({'value' => 6, 'agent' => 'Mozilla/5.0 (compatible; Google Desktop/5.9.1005.12335; http://desktop.google.com/)'}, time)
+      d.filter({'value' => 7, 'agent' => 'msnbot/1.1 (+http://search.msn.com/msnbot.htm)'}, time)
+    end
+
+    filtered = d.filtered_as_array
+    assert_equal 6, filtered.size
+    assert_equal 'test.message', filtered[0][0]
+    assert_equal time, filtered[0][1]
+
+    # 'agent' => 'Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; x64; Trident/6.0)'
+    m = filtered[0][2]
+    assert_equal 8, m.keys.size
+    assert_equal 0, m['value']
+    assert_equal 'Internet Explorer', m['ua_name']
+    assert_equal 'pc', m['ua_category']
+    assert_equal 'Windows 8', m['ua_os']
+    assert_equal 'NT 6.2', m['ua_os_version']
+    assert_equal 'Microsoft', m['ua_vendor']
+    assert_equal '10.0', m['ua_version']
+
+    # 'agent' => 'Mozilla/5.0 (Windows NT 6.0; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'
+    m = filtered[1][2]
+    assert_equal 1, m['value']
+    assert_equal 'Firefox', m['ua_name']
+    assert_equal 'pc', m['ua_category']
+    assert_equal 'Windows Vista', m['ua_os']
+    assert_equal 'NT 6.0', m['ua_os_version']
+    assert_equal 'Mozilla', m['ua_vendor']
+    assert_equal '9.0.1', m['ua_version']
+
+    # 'agent' => 'Mozilla/5.0 (Ubuntu; X11; Linux i686; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'
+    m = filtered[2][2]
+    assert_equal 2, m['value']
+    assert_equal 'Firefox', m['ua_name']
+    assert_equal 'pc', m['ua_category']
+    assert_equal 'Linux', m['ua_os']
+    assert_equal 'UNKNOWN', m['ua_os_version']
+    assert_equal 'Mozilla', m['ua_vendor']
+    assert_equal '9.0.1', m['ua_version']
+
+    # 'agent' => 'Mozilla/5.0 (Linux; U; Android 3.1; ja-jp; L-06C Build/HMJ37) AppleWebKit/534.13 (KHTML, like Gecko) Version/4.0 Safari/534.13'
+    m = filtered[3][2]
+    assert_equal 3, m['value']
+    assert_equal 'Safari', m['ua_name']
+    assert_equal 'smartphone', m['ua_category']
+    assert_equal 'Android', m['ua_os']
+    assert_equal '3.1', m['ua_os_version']
+    assert_equal 'Apple', m['ua_vendor']
+    assert_equal '4.0', m['ua_version']
+
+    # 'agent' => 'DoCoMo/1.0/N505i/c20/TB/W24H12'
+    m = filtered[4][2]
+    assert_equal 4, m['value']
+    assert_equal 'docomo', m['ua_name']
+    assert_equal 'mobilephone', m['ua_category']
+    assert_equal 'docomo', m['ua_os']
+    assert_equal 'UNKNOWN', m['ua_os_version']
+    assert_equal 'docomo', m['ua_vendor']
+    assert_equal 'N505i', m['ua_version']
+
+    # 'agent' => 'Mozilla/5.0 (PlayStation Vita 1.51) AppleWebKit/531.22.8 (KHTML, like Gecko) Silk/3.2'
+    m = filtered[5][2]
+    assert_equal 5, m['value']
+    assert_equal 'PlayStation Vita', m['ua_name']
+    assert_equal 'appliance', m['ua_category']
+    assert_equal 'PlayStation Vita', m['ua_os']
+    assert_equal '1.51', m['ua_os_version']
+    assert_equal 'Sony', m['ua_vendor']
+    assert_equal 'UNKNOWN', m['ua_version']
+  end
+
+  # drop & non-merge
+  def test_filter_drop
+    d = create_driver(CONFIG3, 'test.message')
+    time = Time.parse('2012-07-20 16:40:30').to_i
+    d.run do
+      d.filter({'value' => 0, 'user_agent' => 'Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; x64; Trident/6.0)'}, time)
+      d.filter({'value' => 1, 'user_agent' => 'Mozilla/5.0 (Windows NT 6.0; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'}, time)
+      d.filter({'value' => 2, 'user_agent' => 'Mozilla/5.0 (Ubuntu; X11; Linux i686; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'}, time)
+      d.filter({'value' => 3, 'user_agent' => 'Mozilla/5.0 (Linux; U; Android 3.1; ja-jp; L-06C Build/HMJ37) AppleWebKit/534.13 (KHTML, like Gecko) Version/4.0 Safari/534.13'}, time)
+      d.filter({'value' => 4, 'user_agent' => 'DoCoMo/1.0/N505i/c20/TB/W24H12'}, time)
+      d.filter({'value' => 5, 'user_agent' => 'Mozilla/5.0 (PlayStation Vita 1.51) AppleWebKit/531.22.8 (KHTML, like Gecko) Silk/3.2'}, time)
+      d.filter({'value' => 6, 'user_agent' => 'Mozilla/5.0 (compatible; Google Desktop/5.9.1005.12335; http://desktop.google.com/)'}, time)
+      d.filter({'value' => 7, 'user_agent' => 'msnbot/1.1 (+http://search.msn.com/msnbot.htm)'}, time)
+    end
+
+    filtered = d.filtered_as_array
+    assert_equal 6, filtered.size
+    assert_equal 'test.message', filtered[0][0]
+    assert_equal time, filtered[0][1]
+
+    # 'agent' => 'Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; x64; Trident/6.0)'
+    m = filtered[0][2]
+    assert_equal 0, m['value']
+    assert_equal 2, m.keys.size
+
+    # 'agent' => 'Mozilla/5.0 (Windows NT 6.0; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'
+    m = filtered[1][2]
+    assert_equal 1, m['value']
+
+    # 'agent' => 'Mozilla/5.0 (Ubuntu; X11; Linux i686; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'
+    m = filtered[2][2]
+    assert_equal 2, m['value']
+
+    # 'agent' => 'Mozilla/5.0 (Linux; U; Android 3.1; ja-jp; L-06C Build/HMJ37) AppleWebKit/534.13 (KHTML, like Gecko) Version/4.0 Safari/534.13'
+    m = filtered[3][2]
+    assert_equal 3, m['value']
+
+    # 'agent' => 'DoCoMo/1.0/N505i/c20/TB/W24H12'
+    m = filtered[4][2]
+    assert_equal 4, m['value']
+
+    # 'agent' => 'Mozilla/5.0 (PlayStation Vita 1.51) AppleWebKit/531.22.8 (KHTML, like Gecko) Silk/3.2'
+    m = filtered[5][2]
+    assert_equal 5, m['value']
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: fluent-plugin-woothee
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.2.0
 platform: ruby
 authors:
 - TAGOMORI Satoshi
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-03-26 00:00:00.000000000 Z
+date: 2015-10-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -80,12 +80,14 @@ files:
 - README.md
 - Rakefile
 - fluent-plugin-woothee.gemspec
+- lib/fluent/plugin/filter_woothee.rb
 - lib/fluent/plugin/out_woothee.rb
 - test/helper.rb
+- test/plugin/test_filter_woothee.rb
 - test/plugin/test_out_woothee.rb
 homepage: https://github.com/tagomoris/fluent-plugin-woothee
 licenses:
-- APLv2
+- Apache-2.0
 metadata: {}
 post_install_message: 
 rdoc_options: []
@@ -110,5 +112,6 @@ summary: Fluentd plugin to parse UserAgent strings with woothee parser. It adds
   information or filter records with specific device types.
 test_files:
 - test/helper.rb
+- test/plugin/test_filter_woothee.rb
 - test/plugin/test_out_woothee.rb
 has_rdoc: