logstash-codec-json 0.1.0

Sign up to get free protection for your applications and to get access to all the features.
checksums.yaml ADDED
@@ -0,0 +1,15 @@
1
+ ---
2
+ !binary "U0hBMQ==":
3
+ metadata.gz: !binary |-
4
+ OGJmODk5Yzg2N2M4MTg1Nzg4MzZjZjM1YTM4M2ZmYmExZGZlOGNhYg==
5
+ data.tar.gz: !binary |-
6
+ MmNmYWYwYzBiOTlkY2E5ZDZlNzEyZDUyMTBmZjkwNjI4ODVkOWJlOQ==
7
+ SHA512:
8
+ metadata.gz: !binary |-
9
+ MjU1OGM2MDlmYjMwZmNmMDk3ZDFhOTAxNjk5MTQ2ODU3Njc3ZTk2Yzc2Mzhl
10
+ MjU2MWY2MmViODdjODZiMWI4ODY4OTFjYzFkYWJmNTNjNjUzOWYzYjMwMzBi
11
+ MzZhYzNkYjljN2E0NDU5Y2M2ODA1ZThjZGNkOTliZGUyZTZkZTI=
12
+ data.tar.gz: !binary |-
13
+ NDFmMWY5YWEwOGY4YWRiZmUxYTJhYjkwNDlkOWNjZjMzMzI0MDdkNzc1NjJm
14
+ NjE2MmNmNmVjMDMzYjVhOTBjYzdmOTdmY2RmYWQ2YzU1YjgwYzg5MThmZGI5
15
+ ZjRjMTU4NTljYWY0MWJmMWY2YmNmZWIxMGMyOGQ4NTJlNDAzMGQ=
data/.gitignore ADDED
@@ -0,0 +1,3 @@
1
+ *.gem
2
+ Gemfile.lock
3
+ .bundle
data/Gemfile ADDED
@@ -0,0 +1,4 @@
1
+ source 'http://rubygems.org'
2
+ gem 'rake'
3
+ gem 'gem_publisher'
4
+ gem 'archive-tar-minitar'
data/Rakefile ADDED
@@ -0,0 +1,6 @@
1
+ @files=[]
2
+
3
+ task :default do
4
+ system("rake -T")
5
+ end
6
+
@@ -0,0 +1,48 @@
1
+ # encoding: utf-8
2
+ require "logstash/codecs/base"
3
+ require "logstash/util/charset"
4
+ require "logstash/json"
5
+
6
+ # This codec may be used to decode (via inputs) and encode (via outputs)
7
+ # full JSON messages. If you are streaming JSON messages delimited
8
+ # by '\n' then see the `json_lines` codec.
9
+ # Encoding will result in a single JSON string.
10
+ class LogStash::Codecs::JSON < LogStash::Codecs::Base
11
+ config_name "json"
12
+
13
+ milestone 3
14
+
15
+ # The character encoding used in this codec. Examples include "UTF-8" and
16
+ # "CP1252".
17
+ #
18
+ # JSON requires valid UTF-8 strings, but in some cases, software that
19
+ # emits JSON does so in another encoding (nxlog, for example). In
20
+ # weird cases like this, you can set the `charset` setting to the
21
+ # actual encoding of the text and Logstash will convert it for you.
22
+ #
23
+ # For nxlog users, you'll want to set this to "CP1252".
24
+ config :charset, :validate => ::Encoding.name_list, :default => "UTF-8"
25
+
26
+ public
27
+ def register
28
+ @converter = LogStash::Util::Charset.new(@charset)
29
+ @converter.logger = @logger
30
+ end
31
+
32
+ public
33
+ def decode(data)
34
+ data = @converter.convert(data)
35
+ begin
36
+ yield LogStash::Event.new(LogStash::Json.load(data))
37
+ rescue LogStash::Json::ParserError => e
38
+ @logger.info("JSON parse failure. Falling back to plain-text", :error => e, :data => data)
39
+ yield LogStash::Event.new("message" => data)
40
+ end
41
+ end # def decode
42
+
43
+ public
44
+ def encode(event)
45
+ @on_event.call(event.to_json)
46
+ end # def encode
47
+
48
+ end # class LogStash::Codecs::JSON
@@ -0,0 +1,26 @@
1
+ Gem::Specification.new do |s|
2
+
3
+ s.name = 'logstash-codec-json'
4
+ s.version = '0.1.0'
5
+ s.licenses = ['Apache License (2.0)']
6
+ s.summary = "This codec may be used to decode (via inputs) and encode (via outputs) full JSON messages"
7
+ s.description = "This codec may be used to decode (via inputs) and encode (via outputs) full JSON messages"
8
+ s.authors = ["Elasticsearch"]
9
+ s.email = 'richard.pijnenburg@elasticsearch.com'
10
+ s.homepage = "http://logstash.net/"
11
+ s.require_paths = ["lib"]
12
+
13
+ # Files
14
+ s.files = `git ls-files`.split($\)
15
+
16
+ # Tests
17
+ s.test_files = s.files.grep(%r{^(test|spec|features)/})
18
+
19
+ # Special flag to let us know this is actually a logstash plugin
20
+ s.metadata = { "logstash_plugin" => "true", "group" => "codec" }
21
+
22
+ # Gem dependencies
23
+ s.add_runtime_dependency 'logstash', '>= 1.4.0', '< 2.0.0'
24
+
25
+ end
26
+
@@ -0,0 +1,9 @@
1
+ require "gem_publisher"
2
+
3
+ desc "Publish gem to RubyGems.org"
4
+ task :publish_gem do |t|
5
+ gem_file = Dir.glob(File.expand_path('../*.gemspec',File.dirname(__FILE__))).first
6
+ gem = GemPublisher.publish_if_updated(gem_file, :rubygems)
7
+ puts "Published #{gem}" if gem
8
+ end
9
+
@@ -0,0 +1,169 @@
1
+ require "net/http"
2
+ require "uri"
3
+ require "digest/sha1"
4
+
5
+ def vendor(*args)
6
+ return File.join("vendor", *args)
7
+ end
8
+
9
+ directory "vendor/" => ["vendor"] do |task, args|
10
+ mkdir task.name
11
+ end
12
+
13
+ def fetch(url, sha1, output)
14
+
15
+ puts "Downloading #{url}"
16
+ actual_sha1 = download(url, output)
17
+
18
+ if actual_sha1 != sha1
19
+ fail "SHA1 does not match (expected '#{sha1}' but got '#{actual_sha1}')"
20
+ end
21
+ end # def fetch
22
+
23
+ def file_fetch(url, sha1)
24
+ filename = File.basename( URI(url).path )
25
+ output = "vendor/#{filename}"
26
+ task output => [ "vendor/" ] do
27
+ begin
28
+ actual_sha1 = file_sha1(output)
29
+ if actual_sha1 != sha1
30
+ fetch(url, sha1, output)
31
+ end
32
+ rescue Errno::ENOENT
33
+ fetch(url, sha1, output)
34
+ end
35
+ end.invoke
36
+
37
+ return output
38
+ end
39
+
40
+ def file_sha1(path)
41
+ digest = Digest::SHA1.new
42
+ fd = File.new(path, "r")
43
+ while true
44
+ begin
45
+ digest << fd.sysread(16384)
46
+ rescue EOFError
47
+ break
48
+ end
49
+ end
50
+ return digest.hexdigest
51
+ ensure
52
+ fd.close if fd
53
+ end
54
+
55
+ def download(url, output)
56
+ uri = URI(url)
57
+ digest = Digest::SHA1.new
58
+ tmp = "#{output}.tmp"
59
+ Net::HTTP.start(uri.host, uri.port, :use_ssl => (uri.scheme == "https")) do |http|
60
+ request = Net::HTTP::Get.new(uri.path)
61
+ http.request(request) do |response|
62
+ fail "HTTP fetch failed for #{url}. #{response}" if [200, 301].include?(response.code)
63
+ size = (response["content-length"].to_i || -1).to_f
64
+ count = 0
65
+ File.open(tmp, "w") do |fd|
66
+ response.read_body do |chunk|
67
+ fd.write(chunk)
68
+ digest << chunk
69
+ if size > 0 && $stdout.tty?
70
+ count += chunk.bytesize
71
+ $stdout.write(sprintf("\r%0.2f%%", count/size * 100))
72
+ end
73
+ end
74
+ end
75
+ $stdout.write("\r \r") if $stdout.tty?
76
+ end
77
+ end
78
+
79
+ File.rename(tmp, output)
80
+
81
+ return digest.hexdigest
82
+ rescue SocketError => e
83
+ puts "Failure while downloading #{url}: #{e}"
84
+ raise
85
+ ensure
86
+ File.unlink(tmp) if File.exist?(tmp)
87
+ end # def download
88
+
89
+ def untar(tarball, &block)
90
+ require "archive/tar/minitar"
91
+ tgz = Zlib::GzipReader.new(File.open(tarball))
92
+ # Pull out typesdb
93
+ tar = Archive::Tar::Minitar::Input.open(tgz)
94
+ tar.each do |entry|
95
+ path = block.call(entry)
96
+ next if path.nil?
97
+ parent = File.dirname(path)
98
+
99
+ mkdir_p parent unless File.directory?(parent)
100
+
101
+ # Skip this file if the output file is the same size
102
+ if entry.directory?
103
+ mkdir path unless File.directory?(path)
104
+ else
105
+ entry_mode = entry.instance_eval { @mode } & 0777
106
+ if File.exists?(path)
107
+ stat = File.stat(path)
108
+ # TODO(sissel): Submit a patch to archive-tar-minitar upstream to
109
+ # expose headers in the entry.
110
+ entry_size = entry.instance_eval { @size }
111
+ # If file sizes are same, skip writing.
112
+ next if stat.size == entry_size && (stat.mode & 0777) == entry_mode
113
+ end
114
+ puts "Extracting #{entry.full_name} from #{tarball} #{entry_mode.to_s(8)}"
115
+ File.open(path, "w") do |fd|
116
+ # eof? check lets us skip empty files. Necessary because the API provided by
117
+ # Archive::Tar::Minitar::Reader::EntryStream only mostly acts like an
118
+ # IO object. Something about empty files in this EntryStream causes
119
+ # IO.copy_stream to throw "can't convert nil into String" on JRuby
120
+ # TODO(sissel): File a bug about this.
121
+ while !entry.eof?
122
+ chunk = entry.read(16384)
123
+ fd.write(chunk)
124
+ end
125
+ #IO.copy_stream(entry, fd)
126
+ end
127
+ File.chmod(entry_mode, path)
128
+ end
129
+ end
130
+ tar.close
131
+ File.unlink(tarball) if File.file?(tarball)
132
+ end # def untar
133
+
134
+ def ungz(file)
135
+
136
+ outpath = file.gsub('.gz', '')
137
+ tgz = Zlib::GzipReader.new(File.open(file))
138
+ begin
139
+ File.open(outpath, "w") do |out|
140
+ IO::copy_stream(tgz, out)
141
+ end
142
+ File.unlink(file)
143
+ rescue
144
+ File.unlink(outpath) if File.file?(outpath)
145
+ raise
146
+ end
147
+ tgz.close
148
+ end
149
+
150
+ desc "Process any vendor files required for this plugin"
151
+ task "vendor" do |task, args|
152
+
153
+ @files.each do |file|
154
+ download = file_fetch(file['url'], file['sha1'])
155
+ if download =~ /.tar.gz/
156
+ prefix = download.gsub('.tar.gz', '').gsub('vendor/', '')
157
+ untar(download) do |entry|
158
+ if !file['files'].nil?
159
+ next unless file['files'].include?(entry.full_name.gsub(prefix, ''))
160
+ out = entry.full_name.split("/").last
161
+ end
162
+ File.join('vendor', out)
163
+ end
164
+ elsif download =~ /.gz/
165
+ ungz(download)
166
+ end
167
+ end
168
+
169
+ end
@@ -0,0 +1,83 @@
1
+ require "logstash/codecs/json"
2
+ require "logstash/event"
3
+ require "logstash/json"
4
+ require "insist"
5
+
6
+ describe LogStash::Codecs::JSON do
7
+ subject do
8
+ next LogStash::Codecs::JSON.new
9
+ end
10
+
11
+ context "#decode" do
12
+ it "should return an event from json data" do
13
+ data = {"foo" => "bar", "baz" => {"bah" => ["a","b","c"]}}
14
+ subject.decode(LogStash::Json.dump(data)) do |event|
15
+ insist { event.is_a? LogStash::Event }
16
+ insist { event["foo"] } == data["foo"]
17
+ insist { event["baz"] } == data["baz"]
18
+ insist { event["bah"] } == data["bah"]
19
+ end
20
+ end
21
+
22
+ it "should be fast", :performance => true do
23
+ json = '{"message":"Hello world!","@timestamp":"2013-12-21T07:01:25.616Z","@version":"1","host":"Macintosh.local","sequence":1572456}'
24
+ iterations = 500000
25
+ count = 0
26
+
27
+ # Warmup
28
+ 10000.times { subject.decode(json) { } }
29
+
30
+ start = Time.now
31
+ iterations.times do
32
+ subject.decode(json) do |event|
33
+ count += 1
34
+ end
35
+ end
36
+ duration = Time.now - start
37
+ insist { count } == iterations
38
+ puts "codecs/json rate: #{"%02.0f/sec" % (iterations / duration)}, elapsed: #{duration}s"
39
+ end
40
+
41
+ context "processing plain text" do
42
+ it "falls back to plain text" do
43
+ decoded = false
44
+ subject.decode("something that isn't json") do |event|
45
+ decoded = true
46
+ insist { event.is_a?(LogStash::Event) }
47
+ insist { event["message"] } == "something that isn't json"
48
+ end
49
+ insist { decoded } == true
50
+ end
51
+ end
52
+
53
+ context "processing weird binary blobs" do
54
+ it "falls back to plain text and doesn't crash (LOGSTASH-1595)" do
55
+ decoded = false
56
+ blob = (128..255).to_a.pack("C*").force_encoding("ASCII-8BIT")
57
+ subject.decode(blob) do |event|
58
+ decoded = true
59
+ insist { event.is_a?(LogStash::Event) }
60
+ insist { event["message"].encoding.to_s } == "UTF-8"
61
+ end
62
+ insist { decoded } == true
63
+ end
64
+ end
65
+ end
66
+
67
+ context "#encode" do
68
+ it "should return json data" do
69
+ data = {"foo" => "bar", "baz" => {"bah" => ["a","b","c"]}}
70
+ event = LogStash::Event.new(data)
71
+ got_event = false
72
+ subject.on_event do |d|
73
+ insist { d.chomp } == LogStash::Event.new(data).to_json
74
+ insist { LogStash::Json.load(d)["foo"] } == data["foo"]
75
+ insist { LogStash::Json.load(d)["baz"] } == data["baz"]
76
+ insist { LogStash::Json.load(d)["bah"] } == data["bah"]
77
+ got_event = true
78
+ end
79
+ subject.encode(event)
80
+ insist { got_event }
81
+ end
82
+ end
83
+ end
metadata ADDED
@@ -0,0 +1,76 @@
1
+ --- !ruby/object:Gem::Specification
2
+ name: logstash-codec-json
3
+ version: !ruby/object:Gem::Version
4
+ version: 0.1.0
5
+ platform: ruby
6
+ authors:
7
+ - Elasticsearch
8
+ autorequire:
9
+ bindir: bin
10
+ cert_chain: []
11
+ date: 2014-10-30 00:00:00.000000000 Z
12
+ dependencies:
13
+ - !ruby/object:Gem::Dependency
14
+ name: logstash
15
+ requirement: !ruby/object:Gem::Requirement
16
+ requirements:
17
+ - - ! '>='
18
+ - !ruby/object:Gem::Version
19
+ version: 1.4.0
20
+ - - <
21
+ - !ruby/object:Gem::Version
22
+ version: 2.0.0
23
+ type: :runtime
24
+ prerelease: false
25
+ version_requirements: !ruby/object:Gem::Requirement
26
+ requirements:
27
+ - - ! '>='
28
+ - !ruby/object:Gem::Version
29
+ version: 1.4.0
30
+ - - <
31
+ - !ruby/object:Gem::Version
32
+ version: 2.0.0
33
+ description: This codec may be used to decode (via inputs) and encode (via outputs)
34
+ full JSON messages
35
+ email: richard.pijnenburg@elasticsearch.com
36
+ executables: []
37
+ extensions: []
38
+ extra_rdoc_files: []
39
+ files:
40
+ - .gitignore
41
+ - Gemfile
42
+ - Rakefile
43
+ - lib/logstash/codecs/json.rb
44
+ - logstash-codec-json.gemspec
45
+ - rakelib/publish.rake
46
+ - rakelib/vendor.rake
47
+ - spec/codecs/json_spec.rb
48
+ homepage: http://logstash.net/
49
+ licenses:
50
+ - Apache License (2.0)
51
+ metadata:
52
+ logstash_plugin: 'true'
53
+ group: codec
54
+ post_install_message:
55
+ rdoc_options: []
56
+ require_paths:
57
+ - lib
58
+ required_ruby_version: !ruby/object:Gem::Requirement
59
+ requirements:
60
+ - - ! '>='
61
+ - !ruby/object:Gem::Version
62
+ version: '0'
63
+ required_rubygems_version: !ruby/object:Gem::Requirement
64
+ requirements:
65
+ - - ! '>='
66
+ - !ruby/object:Gem::Version
67
+ version: '0'
68
+ requirements: []
69
+ rubyforge_project:
70
+ rubygems_version: 2.4.1
71
+ signing_key:
72
+ specification_version: 4
73
+ summary: This codec may be used to decode (via inputs) and encode (via outputs) full
74
+ JSON messages
75
+ test_files:
76
+ - spec/codecs/json_spec.rb