logstash-filter-split 0.1.0

checksums.yaml

@@ -0,0 +1,15 @@
+---
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    NTczYTg0Nzk1ZTcwYzQ5OWU2MjEyOTY4NzdlMjJmY2NmNWRiNjU0Zg==
+  data.tar.gz: !binary |-
+    MzE0ZGQzMDIwZTFmNmJjZTQwMzI1ZmVjYzVhNjE2Y2M1ZTcyOTViMQ==
+SHA512:
+  metadata.gz: !binary |-
+    NWUyYTQ3MDhjMmEzMGJjM2I0ZTlkMTUyYWI0MGU3N2U4ZjQzNzNhNTUwNzZm
+    YTk3NDVmYjQ0NTZkMDViYjcxYTIwODFlNzFhOWUzNTk0YTZkMTA5OWUwNzll
+    YmMwZDFkMWIwZDJjOWJhMWY1NGEyNWQ2ZTU0OWYxZGUzMGJjYmY=
+  data.tar.gz: !binary |-
+    ZGVhY2I4NTM0YTY2MTNlOWYzMzBlNTM5NTMzMzllZDI2NWE5NmVmZWM5MjVj
+    N2FmN2M0OGYzZTg0YmM4MjY5YTBiNjc3NGNlNjJkZjE1ZDZlMGFkYWZjNjUy
+    YzY3NDUwMWNhYjljMTc5YTdjNTMyMjBiYjAyMTE2NDA1OTM1ZTk=

data/.gitignore

@@ -0,0 +1,4 @@
+*.gem
+Gemfile.lock
+.bundle
+vendor

data/Gemfile

@@ -0,0 +1,4 @@
+source 'http://rubygems.org'
+gem 'rake'
+gem 'gem_publisher'
+gem 'archive-tar-minitar'

data/Rakefile

@@ -0,0 +1,6 @@
+@files=[]
+
+task :default do
+  system("rake -T")
+end
+

data/lib/logstash/filters/split.rb

@@ -0,0 +1,62 @@
+# encoding: utf-8
+require "logstash/filters/base"
+require "logstash/namespace"
+
+# The split filter is for splitting multiline messages into separate events.
+#
+# An example use case of this filter is for taking output from the 'exec' input
+# which emits one event for the whole output of a command and splitting that
+# output by newline - making each line an event.
+#
+# The end result of each split is a complete copy of the event
+# with only the current split section of the given field changed.
+class LogStash::Filters::Split < LogStash::Filters::Base
+
+  config_name "split"
+  milestone 2
+
+  # The string to split on. This is usually a line terminator, but can be any
+  # string.
+  config :terminator, :validate => :string, :default => "\n"
+
+  # The field which value is split by the terminator
+  config :field, :validate => :string, :default => "message"
+
+  public
+  def register
+    # Nothing to do
+  end # def register
+
+  public
+  def filter(event)
+    return unless filter?(event)
+
+    original_value = event[@field]
+
+    # If for some reason the field is an array of values, take the first only.
+    original_value = original_value.first if original_value.is_a?(Array)
+
+    # Using -1 for 'limit' on String#split makes ruby not drop trailing empty
+    # splits.
+    splits = original_value.split(@terminator, -1)
+
+    # Skip filtering if splitting this event resulted in only one thing found.
+    return if splits.length == 1
+    #or splits[1].empty?
+
+    splits.each do |value|
+      next if value.empty?
+
+      event_split = event.clone
+      @logger.debug("Split event", :value => value, :field => @field)
+      event_split[@field] = value
+      filter_matched(event_split)
+
+      # Push this new event onto the stack at the LogStash::FilterWorker
+      yield event_split
+    end
+
+    # Cancel this event, we'll use the newly generated ones above.
+    event.cancel
+  end # def filter
+end # class LogStash::Filters::Split

data/logstash-filter-split.gemspec

@@ -0,0 +1,26 @@
+Gem::Specification.new do |s|
+
+  s.name            = 'logstash-filter-split'
+  s.version         = '0.1.0'
+  s.licenses        = ['Apache License (2.0)']
+  s.summary         = "The split filter is for splitting multiline messages into separate events."
+  s.description     = "The split filter is for splitting multiline messages into separate events."
+  s.authors         = ["Elasticsearch"]
+  s.email           = 'richard.pijnenburg@elasticsearch.com'
+  s.homepage        = "http://logstash.net/"
+  s.require_paths = ["lib"]
+
+  # Files
+  s.files = `git ls-files`.split($\)
+
+  # Tests
+  s.test_files = s.files.grep(%r{^(test|spec|features)/})
+
+  # Special flag to let us know this is actually a logstash plugin
+  s.metadata = { "logstash_plugin" => "true", "group" => "filter" }
+
+  # Gem dependencies
+  s.add_runtime_dependency 'logstash', '>= 1.4.0', '< 2.0.0'
+
+end
+

data/rakelib/publish.rake

@@ -0,0 +1,9 @@
+require "gem_publisher"
+
+desc "Publish gem to RubyGems.org"
+task :publish_gem do |t|
+  gem_file = Dir.glob(File.expand_path('../*.gemspec',File.dirname(__FILE__))).first
+  gem = GemPublisher.publish_if_updated(gem_file, :rubygems)
+  puts "Published #{gem}" if gem
+end
+

data/rakelib/vendor.rake

@@ -0,0 +1,169 @@
+require "net/http"
+require "uri"
+require "digest/sha1"
+
+def vendor(*args)
+  return File.join("vendor", *args)
+end
+
+directory "vendor/" => ["vendor"] do |task, args|
+  mkdir task.name
+end
+
+def fetch(url, sha1, output)
+
+  puts "Downloading #{url}"
+  actual_sha1 = download(url, output)
+
+  if actual_sha1 != sha1
+    fail "SHA1 does not match (expected '#{sha1}' but got '#{actual_sha1}')"
+  end
+end # def fetch
+
+def file_fetch(url, sha1)
+  filename = File.basename( URI(url).path )
+  output = "vendor/#{filename}"
+  task output => [ "vendor/" ] do
+    begin
+      actual_sha1 = file_sha1(output)
+      if actual_sha1 != sha1
+        fetch(url, sha1, output)
+      end
+    rescue Errno::ENOENT
+      fetch(url, sha1, output)
+    end
+  end.invoke
+
+  return output
+end
+
+def file_sha1(path)
+  digest = Digest::SHA1.new
+  fd = File.new(path, "r")
+  while true
+    begin
+      digest << fd.sysread(16384)
+    rescue EOFError
+      break
+    end
+  end
+  return digest.hexdigest
+ensure
+  fd.close if fd
+end
+
+def download(url, output)
+  uri = URI(url)
+  digest = Digest::SHA1.new
+  tmp = "#{output}.tmp"
+  Net::HTTP.start(uri.host, uri.port, :use_ssl => (uri.scheme == "https")) do |http|
+    request = Net::HTTP::Get.new(uri.path)
+    http.request(request) do |response|
+      fail "HTTP fetch failed for #{url}. #{response}" if [200, 301].include?(response.code)
+      size = (response["content-length"].to_i || -1).to_f
+      count = 0
+      File.open(tmp, "w") do |fd|
+        response.read_body do |chunk|
+          fd.write(chunk)
+          digest << chunk
+          if size > 0 && $stdout.tty?
+            count += chunk.bytesize
+            $stdout.write(sprintf("\r%0.2f%%", count/size * 100))
+          end
+        end
+      end
+      $stdout.write("\r      \r") if $stdout.tty?
+    end
+  end
+
+  File.rename(tmp, output)
+
+  return digest.hexdigest
+rescue SocketError => e
+  puts "Failure while downloading #{url}: #{e}"
+  raise
+ensure
+  File.unlink(tmp) if File.exist?(tmp)
+end # def download
+
+def untar(tarball, &block)
+  require "archive/tar/minitar"
+  tgz = Zlib::GzipReader.new(File.open(tarball))
+  # Pull out typesdb
+  tar = Archive::Tar::Minitar::Input.open(tgz)
+  tar.each do |entry|
+    path = block.call(entry)
+    next if path.nil?
+    parent = File.dirname(path)
+    
+    mkdir_p parent unless File.directory?(parent)
+
+    # Skip this file if the output file is the same size
+    if entry.directory?
+      mkdir path unless File.directory?(path)
+    else
+      entry_mode = entry.instance_eval { @mode } & 0777
+      if File.exists?(path)
+        stat = File.stat(path)
+        # TODO(sissel): Submit a patch to archive-tar-minitar upstream to
+        # expose headers in the entry.
+        entry_size = entry.instance_eval { @size }
+        # If file sizes are same, skip writing.
+        next if stat.size == entry_size && (stat.mode & 0777) == entry_mode
+      end
+      puts "Extracting #{entry.full_name} from #{tarball} #{entry_mode.to_s(8)}"
+      File.open(path, "w") do |fd|
+        # eof? check lets us skip empty files. Necessary because the API provided by
+        # Archive::Tar::Minitar::Reader::EntryStream only mostly acts like an
+        # IO object. Something about empty files in this EntryStream causes
+        # IO.copy_stream to throw "can't convert nil into String" on JRuby
+        # TODO(sissel): File a bug about this.
+        while !entry.eof?
+          chunk = entry.read(16384)
+          fd.write(chunk)
+        end
+          #IO.copy_stream(entry, fd)
+      end
+      File.chmod(entry_mode, path)
+    end
+  end
+  tar.close
+  File.unlink(tarball) if File.file?(tarball)
+end # def untar
+
+def ungz(file)
+
+  outpath = file.gsub('.gz', '')
+  tgz = Zlib::GzipReader.new(File.open(file))
+  begin
+    File.open(outpath, "w") do |out|
+      IO::copy_stream(tgz, out)
+    end
+    File.unlink(file)
+  rescue
+    File.unlink(outpath) if File.file?(outpath)
+   raise
+  end
+  tgz.close
+end
+
+desc "Process any vendor files required for this plugin"
+task "vendor" do |task, args|
+
+  @files.each do |file| 
+    download = file_fetch(file['url'], file['sha1'])
+    if download =~ /.tar.gz/
+      prefix = download.gsub('.tar.gz', '').gsub('vendor/', '')
+      untar(download) do |entry|
+        if !file['files'].nil?
+          next unless file['files'].include?(entry.full_name.gsub(prefix, ''))
+          out = entry.full_name.split("/").last
+        end
+        File.join('vendor', out)
+      end
+    elsif download =~ /.gz/
+      ungz(download)
+    end
+  end
+
+end

data/spec/filters/split_spec.rb

@@ -0,0 +1,59 @@
+# encoding: utf-8
+require "spec_helper"
+require "logstash/filters/split"
+
+describe LogStash::Filters::Split do
+
+  describe "all defaults" do
+    config <<-CONFIG
+      filter {
+        split { }
+      }
+    CONFIG
+
+    sample "big\nbird\nsesame street" do
+      insist { subject.length } == 3
+      insist { subject[0]["message"] } == "big"
+      insist { subject[1]["message"] } == "bird"
+      insist { subject[2]["message"] } == "sesame street"
+    end
+  end
+
+  describe "custome terminator" do
+    config <<-CONFIG
+      filter {
+        split {
+          terminator => "\t"
+        }
+      }
+    CONFIG
+
+    sample "big\tbird\tsesame street" do
+      insist { subject.length } == 3
+      insist { subject[0]["message"] } == "big"
+      insist { subject[1]["message"] } == "bird"
+      insist { subject[2]["message"] } == "sesame street"
+    end
+  end
+
+  describe "custom field" do
+    config <<-CONFIG
+      filter {
+        split {
+          field => "custom"
+        }
+      }
+    CONFIG
+
+    sample("custom" => "big\nbird\nsesame street", "do_not_touch" => "1\n2\n3") do
+      insist { subject.length } == 3
+      subject.each do |s|
+         insist { s["do_not_touch"] } == "1\n2\n3"
+      end
+      insist { subject[0]["custom"] } == "big"
+      insist { subject[1]["custom"] } == "bird"
+      insist { subject[2]["custom"] } == "sesame street"
+    end
+  end
+
+end

metadata

@@ -0,0 +1,74 @@
+--- !ruby/object:Gem::Specification
+name: logstash-filter-split
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Elasticsearch
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-11-02 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: logstash
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: 1.4.0
+    - - <
+      - !ruby/object:Gem::Version
+        version: 2.0.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: 1.4.0
+    - - <
+      - !ruby/object:Gem::Version
+        version: 2.0.0
+description: The split filter is for splitting multiline messages into separate events.
+email: richard.pijnenburg@elasticsearch.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- Gemfile
+- Rakefile
+- lib/logstash/filters/split.rb
+- logstash-filter-split.gemspec
+- rakelib/publish.rake
+- rakelib/vendor.rake
+- spec/filters/split_spec.rb
+homepage: http://logstash.net/
+licenses:
+- Apache License (2.0)
+metadata:
+  logstash_plugin: 'true'
+  group: filter
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.4.1
+signing_key: 
+specification_version: 4
+summary: The split filter is for splitting multiline messages into separate events.
+test_files:
+- spec/filters/split_spec.rb