logstash-codec-oldlogstashjson 0.1.0

checksums.yaml

@@ -0,0 +1,15 @@
+---
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    YzA5MDVkOTM1YmI1N2ZlOGNkNGYzMThjN2UxMWY4M2FhZGExNTJkZg==
+  data.tar.gz: !binary |-
+    ZWIzNTc2OWQ2ZDA4MzRmNTFiNTA2NGE2NzU3MGYxMGE0ODNiODczYQ==
+SHA512:
+  metadata.gz: !binary |-
+    MjM2MmYzMmQzNTZiNDJiZDk5MmQ3ODI0NWY4ZjkzNDk0YTk0ZWM4ZGE4OWJm
+    ZTZmMGJhY2M4NDQ3MTI4ZmUxMTJiN2M0MmJlMGI4NmRlNGUyYmM0MGQyNjc3
+    ZDFmZWYyNGMzNTU1Mzg4OTQyMzM3NDAyOTlkZTc0ZmY1YzkwODg=
+  data.tar.gz: !binary |-
+    ODViODIwZjZlMzY5NmJhOGQxNTVmOWE2N2ZmNWI4YTRmMzYwZWJjZTUyNGI5
+    YmVkMTBmOWM3MGMyY2ZjMDI1ZmRjYWUzZTU4YmZjZGQwMWE3YjdmMGVlOTJi
+    YjFjZjcwNDQ2NjIyNGFjNzgzMDRlOWQ0MTEwNDI5MzU5MjA0ZTU=

data/.gitignore

@@ -0,0 +1,3 @@
+*.gem
+Gemfile.lock
+.bundle

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/codecs/oldlogstashjson.rb

@@ -0,0 +1,57 @@
+# encoding: utf-8
+require "logstash/codecs/base"
+require "logstash/json"
+
+class LogStash::Codecs::OldLogStashJSON < LogStash::Codecs::Base
+  config_name "oldlogstashjson"
+  milestone 2
+
+  # Map from v0 name to v1 name.
+  # Note: @source is gone and has no similar field.
+  V0_TO_V1 = {"@timestamp" => "@timestamp", "@message" => "message",
+              "@tags" => "tags", "@type" => "type",
+              "@source_host" => "host", "@source_path" => "path"}
+
+  public
+  def decode(data)
+    begin
+      obj = LogStash::Json.load(data.force_encoding(Encoding::UTF_8))
+    rescue LogStash::Json::ParserError => e
+      @logger.info("JSON parse failure. Falling back to plain-text", :error => e, :data => data)
+      yield LogStash::Event.new("message" => data)
+      return
+    end
+
+    h  = {}
+
+    # Convert the old logstash schema to the new one.
+    V0_TO_V1.each do |key, val|
+      h[val] = obj[key] if obj.include?(key)
+    end
+
+    h.merge!(obj["@fields"]) if obj["@fields"].is_a?(Hash)
+    yield LogStash::Event.new(h)
+  end # def decode
+
+  public
+  def encode(event)
+    h  = {}
+
+    # Convert the new logstash schema to the old one.
+    V0_TO_V1.each do |key, val|
+      h[key] = event[val] if event.include?(val)
+    end
+
+    event.to_hash.each do |field, val|
+      # TODO: might be better to V1_TO_V0 = V0_TO_V1.invert during
+      # initialization than V0_TO_V1.has_value? within loop
+      next if field == "@version" or V0_TO_V1.has_value?(field)
+      h["@fields"] ||= {}
+      h["@fields"][field] = val
+    end
+
+    # Tack on a \n because JSON outputs 1.1.x had them.
+    @on_event.call(LogStash::Json.dump(h) + NL)
+  end # def encode
+
+end # class LogStash::Codecs::OldLogStashJSON

data/logstash-codec-oldlogstashjson.gemspec

@@ -0,0 +1,26 @@
+Gem::Specification.new do |s|
+
+  s.name            = 'logstash-codec-oldlogstashjson'
+  s.version         = '0.1.0'
+  s.licenses        = ['Apache License (2.0)']
+  s.summary         = "Codec to convert v0 logstash format to v1"
+  s.description     = "Codec to convert v0 logstash format to v1"
+  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" => "codec" }
+
+  # 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/codecs/oldlogstashjson_spec.rb

@@ -0,0 +1,56 @@
+require "logstash/codecs/oldlogstashjson"
+require "logstash/event"
+require "logstash/json"
+require "insist"
+
+describe LogStash::Codecs::OldLogStashJSON do
+  subject do
+    next LogStash::Codecs::OldLogStashJSON.new
+  end
+
+  context "#decode" do
+    it "should return a new (v1) event from old (v0) json data" do
+      data = {"@message" => "bar", "@source_host" => "localhost",
+              "@tags" => ["a","b","c"]}
+      subject.decode(LogStash::Json.dump(data)) do |event|
+        insist { event.is_a? LogStash::Event }
+        insist { event["@timestamp"] } != nil
+        insist { event["type"] } == data["@type"]
+        insist { event["message"] } == data["@message"]
+        insist { event["host"] } == data["@source_host"]
+        insist { event["tags"] } == data["@tags"]
+        insist { event["path"] } == nil # @source_path not in v0 test data
+      end
+    end
+
+    it "should accept invalid json" do
+      subject.decode("some plain text") do |event|
+        insist { event["message"] } == "some plain text"
+      end
+    end
+  end
+
+  context "#encode" do
+    it "should return old (v0) json data" do
+      data = {"type" => "t", "message" => "wat!?",
+              "host" => "localhost", "path" => "/foo",
+              "tags" => ["a","b","c"],
+              "bah" => "baz"}
+      event = LogStash::Event.new(data)
+      got_event = false
+      subject.on_event do |d|
+        insist { LogStash::Json.load(d)["@timestamp"] } != nil
+        insist { LogStash::Json.load(d)["@type"] } == data["type"]
+        insist { LogStash::Json.load(d)["@message"] } == data["message"]
+        insist { LogStash::Json.load(d)["@source_host"] } == data["host"]
+        insist { LogStash::Json.load(d)["@source_path"] } == data["path"]
+        insist { LogStash::Json.load(d)["@tags"] } == data["tags"]
+        insist { LogStash::Json.load(d)["@fields"]["bah"] } == "baz"
+        insist { LogStash::Json.load(d)["@fields"]["@version"] } == nil
+        got_event = true
+      end
+      subject.encode(event)
+      insist { got_event }
+    end
+  end
+end

metadata

@@ -0,0 +1,74 @@
+--- !ruby/object:Gem::Specification
+name: logstash-codec-oldlogstashjson
+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: Codec to convert v0 logstash format to v1
+email: richard.pijnenburg@elasticsearch.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- Gemfile
+- Rakefile
+- lib/logstash/codecs/oldlogstashjson.rb
+- logstash-codec-oldlogstashjson.gemspec
+- rakelib/publish.rake
+- rakelib/vendor.rake
+- spec/codecs/oldlogstashjson_spec.rb
+homepage: http://logstash.net/
+licenses:
+- Apache License (2.0)
+metadata:
+  logstash_plugin: 'true'
+  group: codec
+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: Codec to convert v0 logstash format to v1
+test_files:
+- spec/codecs/oldlogstashjson_spec.rb