logstash-codec-line 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +15 -0
- data/.gitignore +3 -0
- data/Gemfile +4 -0
- data/Rakefile +6 -0
- data/lib/logstash/codecs/line.rb +58 -0
- data/logstash-codec-line.gemspec +26 -0
- data/rakelib/publish.rake +9 -0
- data/rakelib/vendor.rake +169 -0
- data/spec/codecs/line_spec.rb +52 -0
- metadata +75 -0
checksums.yaml
ADDED
@@ -0,0 +1,15 @@
|
|
1
|
+
---
|
2
|
+
!binary "U0hBMQ==":
|
3
|
+
metadata.gz: !binary |-
|
4
|
+
NzU1OWQ3NjUxNDcxZWE1ZGVlOWVlNTg0ZjFhMGMyYmY2NGFhMjQzNg==
|
5
|
+
data.tar.gz: !binary |-
|
6
|
+
NDIyZDNhYmRiM2U0OWExMjY0MzNlOGViZjA1YzMzMmFmY2Q3YzliZQ==
|
7
|
+
SHA512:
|
8
|
+
metadata.gz: !binary |-
|
9
|
+
YTAxYWNlM2RmNTJhNTUxYjViMTkwZjRmZmEyMzZkMmJjY2FiYmJkODkzZGFj
|
10
|
+
ZTFjNTY0OWUxMzYxNDhhZjk2MDhjNGJiYjRmODcwNzNhMDliMTE0NDc1YzI4
|
11
|
+
OGU0YTM3Mjg0YWEwYzRmZjgxNjM4YjQ2YjhiYzU2NDU3MTY0MDY=
|
12
|
+
data.tar.gz: !binary |-
|
13
|
+
MGY0NWI0MTRhNjIyNTRkYzkzNTljMjRjZWY0OGE3MDRlOGQ2YzljZDhlYzZl
|
14
|
+
NGViNjBlYjAwNDNjZjFiNWQ5NGE5YmQ3MzlhMzZiMDQwN2ZlNjk5ZDMwZDFh
|
15
|
+
ZmNmYzU5NDUyOGQxYjcxOWIyYTVlNmI5OGJkOTFhNTM5Y2Q1YzI=
|
data/.gitignore
ADDED
data/Gemfile
ADDED
data/Rakefile
ADDED
@@ -0,0 +1,58 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
require "logstash/codecs/base"
|
3
|
+
require "logstash/util/charset"
|
4
|
+
|
5
|
+
# Line-oriented text data.
|
6
|
+
#
|
7
|
+
# Decoding behavior: Only whole line events will be emitted.
|
8
|
+
#
|
9
|
+
# Encoding behavior: Each event will be emitted with a trailing newline.
|
10
|
+
class LogStash::Codecs::Line < LogStash::Codecs::Base
|
11
|
+
config_name "line"
|
12
|
+
milestone 3
|
13
|
+
|
14
|
+
# Set the desired text format for encoding.
|
15
|
+
config :format, :validate => :string
|
16
|
+
|
17
|
+
# The character encoding used in this input. Examples include "UTF-8"
|
18
|
+
# and "cp1252"
|
19
|
+
#
|
20
|
+
# This setting is useful if your log files are in Latin-1 (aka cp1252)
|
21
|
+
# or in another character set other than UTF-8.
|
22
|
+
#
|
23
|
+
# This only affects "plain" format logs since json is UTF-8 already.
|
24
|
+
config :charset, :validate => ::Encoding.name_list, :default => "UTF-8"
|
25
|
+
|
26
|
+
public
|
27
|
+
def register
|
28
|
+
require "logstash/util/buftok"
|
29
|
+
@buffer = FileWatch::BufferedTokenizer.new
|
30
|
+
@converter = LogStash::Util::Charset.new(@charset)
|
31
|
+
@converter.logger = @logger
|
32
|
+
end
|
33
|
+
|
34
|
+
public
|
35
|
+
def decode(data)
|
36
|
+
@buffer.extract(data).each do |line|
|
37
|
+
yield LogStash::Event.new("message" => @converter.convert(line))
|
38
|
+
end
|
39
|
+
end # def decode
|
40
|
+
|
41
|
+
public
|
42
|
+
def flush(&block)
|
43
|
+
remainder = @buffer.flush
|
44
|
+
if !remainder.empty?
|
45
|
+
block.call(LogStash::Event.new({"message" => remainder}))
|
46
|
+
end
|
47
|
+
end
|
48
|
+
|
49
|
+
public
|
50
|
+
def encode(event)
|
51
|
+
if event.is_a? LogStash::Event and @format
|
52
|
+
@on_event.call(event.sprintf(@format) + NL)
|
53
|
+
else
|
54
|
+
@on_event.call(event.to_s + NL)
|
55
|
+
end
|
56
|
+
end # def encode
|
57
|
+
|
58
|
+
end # class LogStash::Codecs::Plain
|
@@ -0,0 +1,26 @@
|
|
1
|
+
Gem::Specification.new do |s|
|
2
|
+
|
3
|
+
s.name = 'logstash-codec-line'
|
4
|
+
s.version = '0.1.0'
|
5
|
+
s.licenses = ['Apache License (2.0)']
|
6
|
+
s.summary = "Line-oriented text data."
|
7
|
+
s.description = "While decoding only whole line events will be emitted. When encoding each event will be emitted with a trailing newline."
|
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
|
+
|
data/rakelib/vendor.rake
ADDED
@@ -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 +1,52 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
require 'spec_helper'
|
4
|
+
require "logstash/codecs/line"
|
5
|
+
require "logstash/event"
|
6
|
+
|
7
|
+
describe LogStash::Codecs::Line do
|
8
|
+
subject do
|
9
|
+
next LogStash::Codecs::Line.new
|
10
|
+
end
|
11
|
+
|
12
|
+
context "#encode" do
|
13
|
+
let (:event) {LogStash::Event.new({"message" => "hello world", "host" => "test"})}
|
14
|
+
|
15
|
+
it "should return a default date formatted line" do
|
16
|
+
expect(subject).to receive(:on_event).once.and_call_original
|
17
|
+
subject.on_event do |d|
|
18
|
+
insist {d} == event.to_s + "\n"
|
19
|
+
end
|
20
|
+
subject.encode(event)
|
21
|
+
end
|
22
|
+
|
23
|
+
it "should respect the supplied format" do
|
24
|
+
format = "%{host}"
|
25
|
+
subject.format = format
|
26
|
+
expect(subject).to receive(:on_event).once.and_call_original
|
27
|
+
subject.on_event do |d|
|
28
|
+
insist {d} == event.sprintf(format) + "\n"
|
29
|
+
end
|
30
|
+
subject.encode(event)
|
31
|
+
end
|
32
|
+
end
|
33
|
+
|
34
|
+
context "#decode" do
|
35
|
+
it "should return an event from an ascii string" do
|
36
|
+
decoded = false
|
37
|
+
subject.decode("hello world\n") do |e|
|
38
|
+
decoded = true
|
39
|
+
insist { e.is_a?(LogStash::Event) }
|
40
|
+
insist { e["message"] } == "hello world"
|
41
|
+
end
|
42
|
+
insist { decoded } == true
|
43
|
+
end
|
44
|
+
|
45
|
+
it "should return an event from a valid utf-8 string" do
|
46
|
+
subject.decode("München\n") do |e|
|
47
|
+
insist { e.is_a?(LogStash::Event) }
|
48
|
+
insist { e["message"] } == "München"
|
49
|
+
end
|
50
|
+
end
|
51
|
+
end
|
52
|
+
end
|
metadata
ADDED
@@ -0,0 +1,75 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: logstash-codec-line
|
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: While decoding only whole line events will be emitted. When encoding
|
34
|
+
each event will be emitted with a trailing newline.
|
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/line.rb
|
44
|
+
- logstash-codec-line.gemspec
|
45
|
+
- rakelib/publish.rake
|
46
|
+
- rakelib/vendor.rake
|
47
|
+
- spec/codecs/line_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: Line-oriented text data.
|
74
|
+
test_files:
|
75
|
+
- spec/codecs/line_spec.rb
|