fluent-plugin-twitter 0.1.1 → 0.2.1
Sign up to get free protection for your applications and to get access to all the features.
- data/README.md +6 -2
- data/fluent-plugin-twitter.gemspec +2 -2
- data/lib/fluent/plugin/in_twitter.rb +63 -38
- metadata +32 -13
data/README.md
CHANGED
@@ -2,7 +2,7 @@ fluent-plugin-twitter
|
|
2
2
|
=====================
|
3
3
|
|
4
4
|
## Component
|
5
|
-
Fluentd Input/Output plugin
|
5
|
+
Fluentd Input/Output plugin to process tweets with Twitter Streaming API.
|
6
6
|
|
7
7
|
## Installation
|
8
8
|
|
@@ -29,8 +29,9 @@ gem install fluent-plugin-twitter
|
|
29
29
|
oauth_token_secret YOUR_OAUTH_TOKEN_SECRET # Required
|
30
30
|
tag input.twitter.sampling # Required
|
31
31
|
timeline sampling # Required (sampling or userstream)
|
32
|
-
keyword Ruby,Python # Optional (
|
32
|
+
keyword Ruby,Python # Optional (only work with `timeline` is sampling)
|
33
33
|
lang ja,en # Optional
|
34
|
+
output_format nest # Optional (nest or flat or simple[default])
|
34
35
|
</source>
|
35
36
|
|
36
37
|
<match input.twitter.sampling>
|
@@ -71,6 +72,9 @@ $ curl http://localhost:8888/notify.twitter -F 'json={"message":"foo"}'
|
|
71
72
|
### Twitter OAuth Guide
|
72
73
|
http://pocketstudio.jp/log3/2012/02/12/how_to_get_twitter_apikey_and_token/
|
73
74
|
|
75
|
+
### Quick Tour to count a tweet with fluent-plugin-twitter and fluent-plugin-datacounter
|
76
|
+
http://qiita.com/items/fe4258b394190f23fece
|
77
|
+
|
74
78
|
## TODO
|
75
79
|
patches welcome!
|
76
80
|
|
@@ -3,11 +3,11 @@ $:.push File.expand_path("../lib", __FILE__)
|
|
3
3
|
|
4
4
|
Gem::Specification.new do |s|
|
5
5
|
s.name = "fluent-plugin-twitter"
|
6
|
-
s.version = "0.
|
6
|
+
s.version = "0.2.1"
|
7
7
|
s.authors = ["Kentaro Yoshida"]
|
8
8
|
s.email = ["y.ken.studio@gmail.com"]
|
9
9
|
s.homepage = "https://github.com/y-ken/fluent-plugin-twitter"
|
10
|
-
s.summary = %q{Fluentd Input/Output plugin
|
10
|
+
s.summary = %q{Fluentd Input/Output plugin to process tweets with Twitter Streaming API.}
|
11
11
|
|
12
12
|
s.files = `git ls-files`.split("\n")
|
13
13
|
s.test_files = `git ls-files -- {test,spec,features}/*`.split("\n")
|
@@ -1,6 +1,7 @@
|
|
1
1
|
module Fluent
|
2
2
|
class TwitterInput < Fluent::Input
|
3
3
|
TIMELINE_TYPE = %w(userstream sampling)
|
4
|
+
OUTPUT_FORMAT_TYPE = %w(nest flat simple)
|
4
5
|
Plugin.register_input('twitter', self)
|
5
6
|
|
6
7
|
config_param :consumer_key, :string
|
@@ -10,7 +11,9 @@ module Fluent
|
|
10
11
|
config_param :tag, :string
|
11
12
|
config_param :timeline, :string
|
12
13
|
config_param :keyword, :string, :default => nil
|
13
|
-
config_param :lang, :string, :default =>
|
14
|
+
config_param :lang, :string, :default => nil
|
15
|
+
config_param :output_format, :string, :default => 'simple'
|
16
|
+
config_param :flatten_separator, :string, :default => '_'
|
14
17
|
|
15
18
|
def initialize
|
16
19
|
super
|
@@ -19,7 +22,14 @@ module Fluent
|
|
19
22
|
|
20
23
|
def configure(conf)
|
21
24
|
super
|
22
|
-
|
25
|
+
|
26
|
+
if !TIMELINE_TYPE.include?(@timeline)
|
27
|
+
raise Fluent::ConfigError, "timeline value undefined #{@timeline}"
|
28
|
+
end
|
29
|
+
if !OUTPUT_FORMAT_TYPE.include?(@output_format)
|
30
|
+
raise Fluent::ConfigError, "output_format value undefined #{@output_format}"
|
31
|
+
end
|
32
|
+
|
23
33
|
TweetStream.configure do |config|
|
24
34
|
config.consumer_key = @consumer_key
|
25
35
|
config.consumer_secret = @consumer_secret
|
@@ -31,6 +41,9 @@ module Fluent
|
|
31
41
|
|
32
42
|
def start
|
33
43
|
@thread = Thread.new(&method(:run))
|
44
|
+
@any = Proc.new do |hash|
|
45
|
+
get_message(hash) if is_message?(hash)
|
46
|
+
end
|
34
47
|
end
|
35
48
|
|
36
49
|
def shutdown
|
@@ -38,57 +51,69 @@ module Fluent
|
|
38
51
|
end
|
39
52
|
|
40
53
|
def run
|
54
|
+
client = get_twitter_connection
|
41
55
|
if @timeline == 'sampling' && @keyword
|
42
|
-
|
56
|
+
client.track(@keyword)
|
43
57
|
elsif @timeline == 'sampling' && @keyword.nil?
|
44
|
-
|
58
|
+
client.sample
|
45
59
|
elsif @timeline == 'userstream'
|
46
|
-
|
60
|
+
client.userstream
|
61
|
+
#elsif @timeline == 'follow'
|
62
|
+
# client.follow(@follow_ids)
|
47
63
|
end
|
48
64
|
end
|
49
65
|
|
50
|
-
def
|
51
|
-
|
66
|
+
def get_twitter_connection
|
67
|
+
notice = "twitter: starting Twitter Streaming API for #{@timeline}."
|
68
|
+
notice << " tag:#{@tag}"
|
69
|
+
notice << " lang:#{@lang}" unless @lang.nil?
|
70
|
+
notice << " keyword:#{@keyword}" unless @keyword.nil?
|
71
|
+
$log.info notice
|
52
72
|
client = TweetStream::Client.new
|
53
|
-
client.
|
54
|
-
|
55
|
-
|
56
|
-
get_message(status)
|
73
|
+
client.on_anything(&@any)
|
74
|
+
client.on_error do |message|
|
75
|
+
$log.info "twitter: unexpected error has occured. #{message}"
|
57
76
|
end
|
77
|
+
return client
|
58
78
|
end
|
59
79
|
|
60
|
-
def
|
61
|
-
|
62
|
-
|
63
|
-
|
64
|
-
|
65
|
-
next unless @lang.include?(status.user.lang)
|
66
|
-
get_message(status)
|
67
|
-
end
|
80
|
+
def is_message?(status)
|
81
|
+
return false if !status.include?(:text)
|
82
|
+
return false if !status.include?(:user)
|
83
|
+
return false if (!@lang.nil? && @lang != '') && !@lang.include?(status[:user][:lang])
|
84
|
+
return true
|
68
85
|
end
|
69
86
|
|
70
|
-
def
|
71
|
-
|
72
|
-
|
73
|
-
|
74
|
-
|
75
|
-
|
76
|
-
|
87
|
+
def get_message(status)
|
88
|
+
case @output_format
|
89
|
+
when 'nest'
|
90
|
+
record = status.inject({}){|f,(k,v)| f[k.to_s] = v; f}
|
91
|
+
when 'flat'
|
92
|
+
record = hash_flatten(status)
|
93
|
+
when 'simple'
|
94
|
+
record = Hash.new
|
95
|
+
record.store('message', status[:text])
|
96
|
+
record.store('geo', status[:geo])
|
97
|
+
record.store('place', status[:place])
|
98
|
+
record.store('created_at', status[:place])
|
99
|
+
record.store('user_name', status[:user][:name])
|
100
|
+
record.store('user_screen_name', status[:user][:screen_name])
|
101
|
+
record.store('user_profile_image_url', status[:user][:profile_image_url])
|
102
|
+
record.store('user_time_zone', status[:user][:time_zone])
|
103
|
+
record.store('user_lang', status[:user][:lang])
|
77
104
|
end
|
105
|
+
Engine.emit(@tag, Engine.now, record)
|
78
106
|
end
|
79
107
|
|
80
|
-
def
|
81
|
-
record
|
82
|
-
|
83
|
-
|
84
|
-
|
85
|
-
|
86
|
-
|
87
|
-
|
88
|
-
|
89
|
-
record.store('user_time_zone', status.user.time_zone)
|
90
|
-
record.store('user_lang', status.user.lang)
|
91
|
-
Engine.emit(@tag, Engine.now, record)
|
108
|
+
def hash_flatten(record, prefix = nil)
|
109
|
+
record.inject({}) do |d, (k, v)|
|
110
|
+
k = prefix.to_s + k.to_s
|
111
|
+
if v.is_a?(Hash)
|
112
|
+
d.merge(hash_flatten(v, k + @flatten_separator))
|
113
|
+
else
|
114
|
+
d.merge(k => v)
|
115
|
+
end
|
116
|
+
end
|
92
117
|
end
|
93
118
|
end
|
94
119
|
end
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: fluent-plugin-twitter
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.2.1
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -9,11 +9,11 @@ authors:
|
|
9
9
|
autorequire:
|
10
10
|
bindir: bin
|
11
11
|
cert_chain: []
|
12
|
-
date:
|
12
|
+
date: 2013-03-24 00:00:00.000000000 Z
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: fluentd
|
16
|
-
requirement:
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ! '>='
|
@@ -21,10 +21,15 @@ dependencies:
|
|
21
21
|
version: '0'
|
22
22
|
type: :development
|
23
23
|
prerelease: false
|
24
|
-
version_requirements:
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ! '>='
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: '0'
|
25
30
|
- !ruby/object:Gem::Dependency
|
26
31
|
name: fluentd
|
27
|
-
requirement:
|
32
|
+
requirement: !ruby/object:Gem::Requirement
|
28
33
|
none: false
|
29
34
|
requirements:
|
30
35
|
- - ! '>='
|
@@ -32,10 +37,15 @@ dependencies:
|
|
32
37
|
version: '0'
|
33
38
|
type: :runtime
|
34
39
|
prerelease: false
|
35
|
-
version_requirements:
|
40
|
+
version_requirements: !ruby/object:Gem::Requirement
|
41
|
+
none: false
|
42
|
+
requirements:
|
43
|
+
- - ! '>='
|
44
|
+
- !ruby/object:Gem::Version
|
45
|
+
version: '0'
|
36
46
|
- !ruby/object:Gem::Dependency
|
37
47
|
name: twitter
|
38
|
-
requirement:
|
48
|
+
requirement: !ruby/object:Gem::Requirement
|
39
49
|
none: false
|
40
50
|
requirements:
|
41
51
|
- - ! '>='
|
@@ -43,10 +53,15 @@ dependencies:
|
|
43
53
|
version: '0'
|
44
54
|
type: :runtime
|
45
55
|
prerelease: false
|
46
|
-
version_requirements:
|
56
|
+
version_requirements: !ruby/object:Gem::Requirement
|
57
|
+
none: false
|
58
|
+
requirements:
|
59
|
+
- - ! '>='
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '0'
|
47
62
|
- !ruby/object:Gem::Dependency
|
48
63
|
name: tweetstream
|
49
|
-
requirement:
|
64
|
+
requirement: !ruby/object:Gem::Requirement
|
50
65
|
none: false
|
51
66
|
requirements:
|
52
67
|
- - ! '>='
|
@@ -54,7 +69,12 @@ dependencies:
|
|
54
69
|
version: '0'
|
55
70
|
type: :runtime
|
56
71
|
prerelease: false
|
57
|
-
version_requirements:
|
72
|
+
version_requirements: !ruby/object:Gem::Requirement
|
73
|
+
none: false
|
74
|
+
requirements:
|
75
|
+
- - ! '>='
|
76
|
+
- !ruby/object:Gem::Version
|
77
|
+
version: '0'
|
58
78
|
description:
|
59
79
|
email:
|
60
80
|
- y.ken.studio@gmail.com
|
@@ -93,11 +113,10 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
93
113
|
version: '0'
|
94
114
|
requirements: []
|
95
115
|
rubyforge_project:
|
96
|
-
rubygems_version: 1.8.
|
116
|
+
rubygems_version: 1.8.23
|
97
117
|
signing_key:
|
98
118
|
specification_version: 3
|
99
|
-
summary: Fluentd Input/Output plugin
|
100
|
-
messaging system.
|
119
|
+
summary: Fluentd Input/Output plugin to process tweets with Twitter Streaming API.
|
101
120
|
test_files:
|
102
121
|
- test/helper.rb
|
103
122
|
- test/plugin/test_in_twitter.rb
|