hog 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitignore +22 -0
- data/Gemfile +4 -0
- data/LICENSE.txt +22 -0
- data/README.md +173 -0
- data/Rakefile +2 -0
- data/bin/hog +5 -0
- data/example/runner.pig +13 -0
- data/example/tuple_processor.rb +29 -0
- data/example/tuple_processor_udf.rb +22 -0
- data/hog.gemspec +23 -0
- data/lib/hog/field.rb +52 -0
- data/lib/hog/testing/pig_stubs.rb +16 -0
- data/lib/hog/tuple.rb +41 -0
- data/lib/hog/utils.rb +12 -0
- data/lib/hog/version.rb +3 -0
- data/lib/hog.rb +19 -0
- metadata +95 -0
data/.gitignore
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
*.gem
|
2
|
+
*.rbc
|
3
|
+
.bundle
|
4
|
+
.config
|
5
|
+
.yardoc
|
6
|
+
Gemfile.lock
|
7
|
+
InstalledFiles
|
8
|
+
_yardoc
|
9
|
+
coverage
|
10
|
+
doc/
|
11
|
+
lib/bundler/man
|
12
|
+
pkg
|
13
|
+
rdoc
|
14
|
+
spec/reports
|
15
|
+
test/tmp
|
16
|
+
test/version_tmp
|
17
|
+
tmp
|
18
|
+
*.bundle
|
19
|
+
*.so
|
20
|
+
*.o
|
21
|
+
*.a
|
22
|
+
mkmf.log
|
data/Gemfile
ADDED
data/LICENSE.txt
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
Copyright (c) 2014 jondot
|
2
|
+
|
3
|
+
MIT License
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
6
|
+
a copy of this software and associated documentation files (the
|
7
|
+
"Software"), to deal in the Software without restriction, including
|
8
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
9
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
10
|
+
permit persons to whom the Software is furnished to do so, subject to
|
11
|
+
the following conditions:
|
12
|
+
|
13
|
+
The above copyright notice and this permission notice shall be
|
14
|
+
included in all copies or substantial portions of the Software.
|
15
|
+
|
16
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
17
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
18
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
19
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
20
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
21
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
22
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,173 @@
|
|
1
|
+
# Hog
|
2
|
+
|
3
|
+
Supercharge your Ruby Pig UDFs with Hog.
|
4
|
+
|
5
|
+
|
6
|
+
## Why Use Hog?
|
7
|
+
|
8
|
+
Use Hog when you want to process (map) tuples with Hadoop, and generally
|
9
|
+
belive that:
|
10
|
+
|
11
|
+
* It's too much overhead to write custom M/R code
|
12
|
+
* You're too lazy to process with Pig data pipes and sprinkled custom UDFs, and they're not powerful
|
13
|
+
enough anyway.
|
14
|
+
* Data processing with Hive queries is wrong
|
15
|
+
* Cascading is an overkill for your case
|
16
|
+
* Streaming Hadoop (unix processes) is the last resort
|
17
|
+
|
18
|
+
And instead want to develop within your familiar Ruby ecosystem but
|
19
|
+
also:
|
20
|
+
|
21
|
+
* Control and tweak job parameters via Pig
|
22
|
+
* Use Pig's job management
|
23
|
+
* Use Pig's I/O (serdes) easily
|
24
|
+
|
25
|
+
And you also want to use gems, jars, and everything Ruby (JRuby) has to offer for a really quick turn-around and pleasent development experience.
|
26
|
+
|
27
|
+
|
28
|
+
## Quickstart
|
29
|
+
|
30
|
+
Let's say you want to perform geolocation, detect a device's form
|
31
|
+
factor, and shove bits and fields from the nested structure into a Hive table to be able to query
|
32
|
+
it efficiently.
|
33
|
+
|
34
|
+
This is your raw event:
|
35
|
+
|
36
|
+
```javascript
|
37
|
+
{
|
38
|
+
"x-forwarded-for":"8.8.8.8",
|
39
|
+
"user-agent": "123",
|
40
|
+
"pageview": {
|
41
|
+
"userid":"ud-123",
|
42
|
+
"timestamp":123123123,
|
43
|
+
"headers":{
|
44
|
+
"user-agent":"...",
|
45
|
+
"cookies":"..."
|
46
|
+
}
|
47
|
+
}
|
48
|
+
}
|
49
|
+
```
|
50
|
+
|
51
|
+
To do this with Pig alone, you would need:
|
52
|
+
|
53
|
+
* A pig UDF for geolocation
|
54
|
+
* A pig UDF for form-factor detection
|
55
|
+
* A way to extract JSON fields from a tuple in Pig
|
56
|
+
* Combine everything in Pig code
|
57
|
+
|
58
|
+
And you'll give up
|
59
|
+
|
60
|
+
* Testability
|
61
|
+
* Maintainability
|
62
|
+
* Development speed
|
63
|
+
* Eventually, sanity
|
64
|
+
|
65
|
+
|
66
|
+
#### Using Hog
|
67
|
+
|
68
|
+
With Hog, you'll mostly feel you're _describing_ how you want the data to be
|
69
|
+
shaped, and it'll do the heavy lifting.
|
70
|
+
|
71
|
+
You always have the option to 'drop' to code with the `prepare`
|
72
|
+
block.
|
73
|
+
|
74
|
+
|
75
|
+
```ruby
|
76
|
+
# gem install hog
|
77
|
+
require 'hog'
|
78
|
+
|
79
|
+
TupleProcessor = Hog.tuple "mytuple" do
|
80
|
+
|
81
|
+
# Your own custom feature extraction. This can be any free style
|
82
|
+
# Ruby or Java code
|
83
|
+
#
|
84
|
+
prepare do |hsh|
|
85
|
+
loc = ip2location(hsh["x-forwarded-for"])
|
86
|
+
hsh.merge!(loc)
|
87
|
+
|
88
|
+
form_factor = formfactor(hsh["user-agent"])
|
89
|
+
hsh.merge!(form_factor)
|
90
|
+
end
|
91
|
+
|
92
|
+
|
93
|
+
# Describe your data columns (i.e. for Hive), and how you'd like
|
94
|
+
# to pull them out of your nested data.
|
95
|
+
#
|
96
|
+
# You can also put a fixed value, or serialize a sub-nested
|
97
|
+
# structure as-is for later use with Hive's json_object
|
98
|
+
#
|
99
|
+
chararray :id, :path => 'pageview.userid'
|
100
|
+
chararray :created_at, :path => 'pageview.timestamp', :with_nil => ''
|
101
|
+
float :sampling, :value => 1.0
|
102
|
+
chararray :ev_type, :value => 'pageview'
|
103
|
+
chararray :ev_info, :json => 'pageview.headers', :default => {}
|
104
|
+
end
|
105
|
+
```
|
106
|
+
|
107
|
+
As you'll notice - there's very little code, and you specify the bare
|
108
|
+
essentials -- which you would have had to specify (types and fields) with Pig in any way you'd try to use it.
|
109
|
+
|
110
|
+
|
111
|
+
Hog will generate a `TupleProcessor` that conforms to your description and logic.
|
112
|
+
|
113
|
+
|
114
|
+
#### Testing
|
115
|
+
|
116
|
+
To test your mapper, just use `TupleProcessor` within your specs and
|
117
|
+
give it a raw event. It's plain Ruby.
|
118
|
+
|
119
|
+
|
120
|
+
#### Hadoop
|
121
|
+
|
122
|
+
To hookup with Hadoop, rig `TupleProcessor` within a Pig JRuby UDF shell like so:
|
123
|
+
|
124
|
+
```
|
125
|
+
require 'tuple_processor'
|
126
|
+
require 'pigudf'
|
127
|
+
|
128
|
+
class TupleProcessorUdf < PigUdf
|
129
|
+
# TupleProcessor will automatically generate your UDF schema!, no matter how complex or
|
130
|
+
# convoluted.
|
131
|
+
outputSchema TupleProcessor.schema
|
132
|
+
|
133
|
+
# Use TupleProcessor as the processing logic.
|
134
|
+
def process(line)
|
135
|
+
# since 1 raw json can output several rows, 'process' returns
|
136
|
+
# an array. we pass that to Pig as Pig's own 'DataBag'.
|
137
|
+
# You can also do without and just return whatever TupleProcessor#process returns.
|
138
|
+
databag = DataBag.new
|
139
|
+
TupleProcessor.process(line).each do |res|
|
140
|
+
databag.add(res)
|
141
|
+
end
|
142
|
+
|
143
|
+
databag
|
144
|
+
end
|
145
|
+
end
|
146
|
+
```
|
147
|
+
|
148
|
+
|
149
|
+
## Gems and Jars
|
150
|
+
|
151
|
+
So you can't really use gems or jars with Pig Ruby UDFs unless all machines have
|
152
|
+
them; and then it becomes ugly to manage really.
|
153
|
+
|
154
|
+
The solution is neat: pack everything as you would with a JRuby project,
|
155
|
+
for example, use `warbler`; and your job code becomes a standalone-jar
|
156
|
+
which you deploy, version, and possibly generate via CI.
|
157
|
+
|
158
|
+
|
159
|
+
## Related Projects
|
160
|
+
|
161
|
+
* PigPen - https://github.com/Netflix/PigPen
|
162
|
+
* datafu - https://github.com/linkedin/datafu
|
163
|
+
|
164
|
+
|
165
|
+
# Contributing
|
166
|
+
|
167
|
+
Fork, implement, add tests, pull request, get my everlasting thanks and a respectable place here :).
|
168
|
+
|
169
|
+
# Copyright
|
170
|
+
|
171
|
+
Copyright (c) 2014 [Dotan Nahum](http://gplus.to/dotan) [@jondot](http://twitter.com/jondot). See MIT-LICENSE for further details.
|
172
|
+
|
173
|
+
|
data/Rakefile
ADDED
data/bin/hog
ADDED
data/example/runner.pig
ADDED
@@ -0,0 +1,13 @@
|
|
1
|
+
-- Since most work is in your TupleProcessor, this file serves as a simple shim
|
2
|
+
-- that loads data into raw lines, and sends them off to your UDF.
|
3
|
+
-- As a result, it will mostly stay the same.
|
4
|
+
-- You can tweak any M/R or Pig params here as well.
|
5
|
+
|
6
|
+
REGISTER 'your-warbled-dependencies-jar.jar'
|
7
|
+
REGISTER 'tuple_processor_udf.rb' USING jruby AS processor;
|
8
|
+
|
9
|
+
data = FOREACH ( LOAD '/raw-data' AS (line:CHARARRAY) )
|
10
|
+
GENERATE flatten(process.process(line));
|
11
|
+
|
12
|
+
store data into '/out-data' USING PARQUETFILE
|
13
|
+
|
@@ -0,0 +1,29 @@
|
|
1
|
+
require 'hog'
|
2
|
+
|
3
|
+
TupleProcessor = Hog.tuple "mytuple" do
|
4
|
+
|
5
|
+
# Your own custom feature extraction. This can be any free style
|
6
|
+
# Ruby or Java code
|
7
|
+
#
|
8
|
+
prepare do |hsh|
|
9
|
+
loc = ip2location(hsh["x-forwarded-for"])
|
10
|
+
hsh.merge!(loc)
|
11
|
+
|
12
|
+
form_factor = formfactor(hsh["user-agent"])
|
13
|
+
hsh.merge!(form_factor)
|
14
|
+
end
|
15
|
+
|
16
|
+
|
17
|
+
# Describe your data columns (i.e. for Hive), and how you'd like
|
18
|
+
# to pull them out of your nested data.
|
19
|
+
#
|
20
|
+
# You can also put a fixed value, or serialize a sub-nested
|
21
|
+
# structure as-is for later use with Hive's json_object
|
22
|
+
#
|
23
|
+
chararray :id, :path => 'pageview.userid'
|
24
|
+
chararray :created_at, :path => 'pageview.timestamp', :with_nil => ''
|
25
|
+
float :sampling, :value => 1.0
|
26
|
+
chararray :ev_type, :value => 'pageview'
|
27
|
+
chararray :ev_info, :json => 'pageview.headers', :default => {}
|
28
|
+
end
|
29
|
+
|
@@ -0,0 +1,22 @@
|
|
1
|
+
require 'pigudf'
|
2
|
+
require 'tuple_processor'
|
3
|
+
|
4
|
+
class TupleProcessorUdf < PigUdf
|
5
|
+
# TupleProcessor will automatically generate your UDF schema!, no matter how complex or
|
6
|
+
# convoluted.
|
7
|
+
outputSchema TupleProcessor.schema
|
8
|
+
|
9
|
+
# Use TupleProcessor as the processing logic.
|
10
|
+
def process(line)
|
11
|
+
# since 1 raw json can output several rows, 'process' returns
|
12
|
+
# an array. we pass that to Pig as Pig's own 'DataBag'.
|
13
|
+
# You can also do without and just return whatever TupleProcessor#process returns.
|
14
|
+
databag = DataBag.new
|
15
|
+
TupleProcessor.process(line).each do |res|
|
16
|
+
databag.add(res)
|
17
|
+
end
|
18
|
+
|
19
|
+
databag
|
20
|
+
end
|
21
|
+
end
|
22
|
+
|
data/hog.gemspec
ADDED
@@ -0,0 +1,23 @@
|
|
1
|
+
# coding: utf-8
|
2
|
+
lib = File.expand_path('../lib', __FILE__)
|
3
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
|
+
require 'hog/version'
|
5
|
+
|
6
|
+
Gem::Specification.new do |spec|
|
7
|
+
spec.name = "hog"
|
8
|
+
spec.version = Hog::VERSION
|
9
|
+
spec.authors = ["jondot"]
|
10
|
+
spec.email = ["jondotan@gmail.com"]
|
11
|
+
spec.summary = %q{}
|
12
|
+
spec.description = %q{}
|
13
|
+
spec.homepage = ""
|
14
|
+
spec.license = "MIT"
|
15
|
+
|
16
|
+
spec.files = `git ls-files -z`.split("\x0")
|
17
|
+
spec.executables = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
|
18
|
+
spec.test_files = spec.files.grep(%r{^(test|spec|features)/})
|
19
|
+
spec.require_paths = ["lib"]
|
20
|
+
|
21
|
+
spec.add_development_dependency "bundler", "~> 1.6"
|
22
|
+
spec.add_development_dependency "rake"
|
23
|
+
end
|
data/lib/hog/field.rb
ADDED
@@ -0,0 +1,52 @@
|
|
1
|
+
module Hog
|
2
|
+
class Field
|
3
|
+
attr_reader :opts
|
4
|
+
|
5
|
+
def initialize(type, name, opts={})
|
6
|
+
@name = name
|
7
|
+
@type = type
|
8
|
+
@opts = opts
|
9
|
+
opts[:path] ||= name.to_s
|
10
|
+
end
|
11
|
+
|
12
|
+
def get(hash)
|
13
|
+
val = nil
|
14
|
+
if opts[:value]
|
15
|
+
val = opts[:value]
|
16
|
+
elsif opts[:json]
|
17
|
+
lkval = lookup(opts[:json],hash) || opts[:default]
|
18
|
+
val = JSON.dump(lkval)
|
19
|
+
else
|
20
|
+
val = lookup(opts[:path], hash)
|
21
|
+
end
|
22
|
+
get_value(val)
|
23
|
+
end
|
24
|
+
|
25
|
+
def to_s
|
26
|
+
"#{@name}:#{@type}"
|
27
|
+
end
|
28
|
+
|
29
|
+
private
|
30
|
+
|
31
|
+
def get_value(val)
|
32
|
+
# xxx optimize?
|
33
|
+
if @type == "int" && [TrueClass, FalseClass].include?(val.class)
|
34
|
+
val ? 1 : 0
|
35
|
+
elsif @type == "chararray"
|
36
|
+
val.to_s
|
37
|
+
else
|
38
|
+
val
|
39
|
+
end
|
40
|
+
end
|
41
|
+
|
42
|
+
def lookup(path, hash)
|
43
|
+
begin
|
44
|
+
path.split('.').inject(hash) {|acc, value| acc[value]}
|
45
|
+
rescue
|
46
|
+
return opts[:with_nil] if opts[:with_nil]
|
47
|
+
raise "No path: #{path} in hash: #{hash}"
|
48
|
+
end
|
49
|
+
end
|
50
|
+
end
|
51
|
+
end
|
52
|
+
|
data/lib/hog/tuple.rb
ADDED
@@ -0,0 +1,41 @@
|
|
1
|
+
# todos
|
2
|
+
# - safe hash path lookup + provide default value?
|
3
|
+
|
4
|
+
|
5
|
+
module Hog
|
6
|
+
class Tuple
|
7
|
+
def initialize(name, block)
|
8
|
+
@name = name
|
9
|
+
@fields = []
|
10
|
+
self.instance_eval &block
|
11
|
+
end
|
12
|
+
|
13
|
+
def process(hsh)
|
14
|
+
@prepare.call(hsh)
|
15
|
+
res = []
|
16
|
+
@fields.each do |f|
|
17
|
+
res << f.get(hsh)
|
18
|
+
end
|
19
|
+
res
|
20
|
+
end
|
21
|
+
|
22
|
+
def prepare(&block)
|
23
|
+
@prepare = block
|
24
|
+
end
|
25
|
+
|
26
|
+
def to_s
|
27
|
+
"(#{@fields.map{|f| f.to_s }.join(',')})"
|
28
|
+
end
|
29
|
+
|
30
|
+
def method_missing(meth, *args, &block)
|
31
|
+
if [:chararray, :float, :double, :int, :long, :map].include?(meth)
|
32
|
+
f = Field.new(meth.to_s, *args)
|
33
|
+
@fields << f
|
34
|
+
return f
|
35
|
+
else
|
36
|
+
super
|
37
|
+
end
|
38
|
+
end
|
39
|
+
end
|
40
|
+
end
|
41
|
+
|
data/lib/hog/utils.rb
ADDED
data/lib/hog/version.rb
ADDED
data/lib/hog.rb
ADDED
@@ -0,0 +1,19 @@
|
|
1
|
+
require "hog/version"
|
2
|
+
require 'json'
|
3
|
+
require 'hog/tuple'
|
4
|
+
require 'hog/field'
|
5
|
+
require 'hog/utils'
|
6
|
+
|
7
|
+
module Hog
|
8
|
+
class << self
|
9
|
+
include Hog::Utils
|
10
|
+
end
|
11
|
+
|
12
|
+
def self.tuple(name, &block)
|
13
|
+
Tuple.new(name, block)
|
14
|
+
end
|
15
|
+
end
|
16
|
+
|
17
|
+
|
18
|
+
|
19
|
+
|
metadata
ADDED
@@ -0,0 +1,95 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: hog
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- jondot
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2014-06-18 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: bundler
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ~>
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: '1.6'
|
22
|
+
type: :development
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ~>
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: '1.6'
|
30
|
+
- !ruby/object:Gem::Dependency
|
31
|
+
name: rake
|
32
|
+
requirement: !ruby/object:Gem::Requirement
|
33
|
+
none: false
|
34
|
+
requirements:
|
35
|
+
- - ! '>='
|
36
|
+
- !ruby/object:Gem::Version
|
37
|
+
version: '0'
|
38
|
+
type: :development
|
39
|
+
prerelease: false
|
40
|
+
version_requirements: !ruby/object:Gem::Requirement
|
41
|
+
none: false
|
42
|
+
requirements:
|
43
|
+
- - ! '>='
|
44
|
+
- !ruby/object:Gem::Version
|
45
|
+
version: '0'
|
46
|
+
description: ''
|
47
|
+
email:
|
48
|
+
- jondotan@gmail.com
|
49
|
+
executables:
|
50
|
+
- hog
|
51
|
+
extensions: []
|
52
|
+
extra_rdoc_files: []
|
53
|
+
files:
|
54
|
+
- .gitignore
|
55
|
+
- Gemfile
|
56
|
+
- LICENSE.txt
|
57
|
+
- README.md
|
58
|
+
- Rakefile
|
59
|
+
- bin/hog
|
60
|
+
- example/runner.pig
|
61
|
+
- example/tuple_processor.rb
|
62
|
+
- example/tuple_processor_udf.rb
|
63
|
+
- hog.gemspec
|
64
|
+
- lib/hog.rb
|
65
|
+
- lib/hog/field.rb
|
66
|
+
- lib/hog/testing/pig_stubs.rb
|
67
|
+
- lib/hog/tuple.rb
|
68
|
+
- lib/hog/utils.rb
|
69
|
+
- lib/hog/version.rb
|
70
|
+
homepage: ''
|
71
|
+
licenses:
|
72
|
+
- MIT
|
73
|
+
post_install_message:
|
74
|
+
rdoc_options: []
|
75
|
+
require_paths:
|
76
|
+
- lib
|
77
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
78
|
+
none: false
|
79
|
+
requirements:
|
80
|
+
- - ! '>='
|
81
|
+
- !ruby/object:Gem::Version
|
82
|
+
version: '0'
|
83
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
84
|
+
none: false
|
85
|
+
requirements:
|
86
|
+
- - ! '>='
|
87
|
+
- !ruby/object:Gem::Version
|
88
|
+
version: '0'
|
89
|
+
requirements: []
|
90
|
+
rubyforge_project:
|
91
|
+
rubygems_version: 1.8.23
|
92
|
+
signing_key:
|
93
|
+
specification_version: 3
|
94
|
+
summary: ''
|
95
|
+
test_files: []
|