fluent-plugin-rewrite-tag-filter 1.0.0 → 1.1.0

data/README.rdoc

@@ -22,7 +22,7 @@ user-agent, request-uri, regex-backreference and so on with regular expression.
 
     rewruterule<num:1-200> <attribute> <regex_pattern> <new_tag>
     
-    # Capitalize every matched regex backreference. (ex: $1, $2)
+    # Optional: Capitalize every matched regex backreference. (ex: $1, $2)
     capitalize_regex_backreference <yes/no(default)>
 
 === Usage
@@ -87,6 +87,15 @@ On starting td-agent, Logging supported like below.
     2012-09-16 18:10:51 +0900: adding rewrite_tag_filter rule: [4, "domain", /^(mail)\.(example)\.com$/, "site.$2$1"]
     2012-09-16 18:10:51 +0900: adding rewrite_tag_filter rule: [5, "domain", /.+/, "site.unmatched"]
 
+=== Tag placeholder
+
+Supporing these placeholder for rewrite tag.
+
+* ${hostname}
+* __HOSTNAME__
+* ${tag}
+* __TAG__
+
 == Example
 
 - Example1: how to analyze response_time, response_code and user_agent for each virtual domain websites.

data/fluent-plugin-rewrite-tag-filter.gemspec

@@ -3,7 +3,7 @@ $:.push File.expand_path("../lib", __FILE__)
 
 Gem::Specification.new do |s|
   s.name        = "fluent-plugin-rewrite-tag-filter"
-  s.version     = "1.0.0"
+  s.version     = "1.1.0"
   s.authors     = ["Kentaro Yoshida"]
   s.email       = ["y.ken.studio@gmail.com"]
   s.homepage    = "https://github.com/y-ken/fluent-plugin-rewrite-tag-filter"

data/lib/fluent/plugin/out_rewrite_tag_filter.rb

@@ -14,6 +14,7 @@ class Fluent::RewriteTagFilterOutput < Fluent::Output
 
     @rewriterules = []
     rewriterule_names = []
+    @hostname = `hostname`.chomp
 
     invalids = conf.keys.select{|k| k =~ /^rewriterule(\d+)$/}.select{|arg| arg =~ /^rewriterule(\d+)/ and not (1..PATTERN_MAX_NUM).include?($1.to_i)}
     if invalids.size > 0
@@ -39,18 +40,17 @@ class Fluent::RewriteTagFilterOutput < Fluent::Output
   end
 
   def emit(tag, es, chain)
+    placeholder = get_placeholder(tag)
     es.each do |time,record|
       rewrite = false
       @rewriterules.each do |index, rewritekey, regexp, rewritetag|
         rewritevalue = record[rewritekey]
         next if rewritevalue.nil?
         next unless (regexp && regexp.match(rewritevalue))
+        backreference_table = map_regex_table($~.captures)
         rewrite = true
-        tag = if rewritetag.include?('$')
-                rewritetag.gsub(/\$\d+/, map_regex_table($~.captures))
-              else
-                rewritetag
-              end
+        rewritetag.gsub!(/(\${[a-z]+}|__[A-Z]+__)/, placeholder)
+        tag = rewritetag.gsub(/\$\d+/, backreference_table)
         break
       end
       Fluent::Engine.emit(tag, time, record) if (rewrite)
@@ -76,5 +76,14 @@ class Fluent::RewriteTagFilterOutput < Fluent::Output
     end
     return hash_table
   end
+
+  def get_placeholder(tag)
+    return {
+      '__HOSTNAME__' => @hostname,
+      '${hostname}' => @hostname,
+      '__TAG__' => tag,
+      '${tag}' => tag,
+    }
+  end
 end
 

data/test/plugin/test_out_rewrite_tag_filter.rb

@@ -10,6 +10,7 @@ class RewriteTagFilterOutputTest < Test::Unit::TestCase
     rewriterule2 domain ^news\.google\.com$ site.GoogleNews
     rewriterule3 agent .* Mac OS X .* agent.MacOSX
     rewriterule4 agent (Googlebot|CustomBot)-([a-zA-Z]+) agent.$1-$2
+    rewriterule5 domain ^(tagtest)\.google\.com$ site.${tag}.$1
   ]
 
   # aggresive test
@@ -48,9 +49,10 @@ class RewriteTagFilterOutputTest < Test::Unit::TestCase
       d1.emit({'domain' => 'news.google.com', 'path' => '/', 'agent' => 'Googlebot-Mobile', 'response_time' => 900000})
       d1.emit({'domain' => 'map.google.com', 'path' => '/', 'agent' => 'Macintosh; Intel Mac OS X 10_7_4', 'response_time' => 900000})
       d1.emit({'domain' => 'labs.google.com', 'path' => '/', 'agent' => 'Mozilla/5.0 Googlebot-FooBar/2.1', 'response_time' => 900000})
+      d1.emit({'domain' => 'tagtest.google.com', 'path' => '/', 'agent' => 'Googlebot', 'response_time' => 900000})
     end
     emits = d1.emits
-    assert_equal 4, emits.length
+    assert_equal 5, emits.length
     p emits[0]
     assert_equal 'site.Google', emits[0][0] # tag
     p emits[1]
@@ -60,6 +62,8 @@ class RewriteTagFilterOutputTest < Test::Unit::TestCase
     assert_equal 'agent.MacOSX', emits[2][0] #tag
     p emits[3]
     assert_equal 'agent.Googlebot-FooBar', emits[3][0] #tag
+    p emits[4]
+    assert_equal 'site.input.access.tagtest', emits[4][0] #tag
   end
 
   def test_emit2

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: fluent-plugin-rewrite-tag-filter
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.1.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-09-18 00:00:00.000000000 Z
+date: 2012-11-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: fluentd
-  requirement: &11833100 !ruby/object:Gem::Requirement
+  requirement: &11336200 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,10 +21,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11833100
+  version_requirements: *11336200
 - !ruby/object:Gem::Dependency
   name: fluentd
-  requirement: &11832240 !ruby/object:Gem::Requirement
+  requirement: &11335760 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -32,7 +32,7 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *11832240
+  version_requirements: *11335760
 description: 
 email:
 - y.ken.studio@gmail.com