cf-swearjar 1.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitignore +2 -0
- data/Gemfile +3 -0
- data/README.rdoc +28 -0
- data/Rakefile +18 -0
- data/cf-swearjar.gemspec +37 -0
- data/lib/config/en.yml +1521 -0
- data/lib/swearjar/tester.rb +32 -0
- data/lib/swearjar/version.rb +3 -0
- data/lib/swearjar.rb +63 -0
- data/spec/data/swear.yml +2 -0
- data/spec/spec.opts +7 -0
- data/spec/spec_helper.rb +1 -0
- data/spec/swearjar_spec.rb +48 -0
- metadata +113 -0
@@ -0,0 +1,32 @@
|
|
1
|
+
require 'yaml'
|
2
|
+
require 'fuzzy_hash'
|
3
|
+
require 'bloomfilter'
|
4
|
+
|
5
|
+
class Swearjar
|
6
|
+
class Tester
|
7
|
+
|
8
|
+
def initialize(config_file)
|
9
|
+
data = YAML.load_file
|
10
|
+
|
11
|
+
@tester = FuzzyHash.new
|
12
|
+
|
13
|
+
data['regex'].each do |pattern, type|
|
14
|
+
@tester[Regexp.new(pattern)] = type
|
15
|
+
end
|
16
|
+
|
17
|
+
data['simple'].each do |test, type|
|
18
|
+
@tester[test] = type
|
19
|
+
end
|
20
|
+
|
21
|
+
end
|
22
|
+
|
23
|
+
def scan(string, &block)
|
24
|
+
string.scan(/\b[\b]+\b/, &block)
|
25
|
+
end
|
26
|
+
|
27
|
+
def profane?(string)
|
28
|
+
scan(string) {|w| return true}
|
29
|
+
end
|
30
|
+
|
31
|
+
end
|
32
|
+
end
|
data/lib/swearjar.rb
ADDED
@@ -0,0 +1,63 @@
|
|
1
|
+
require 'yaml'
|
2
|
+
require 'fuzzy_hash'
|
3
|
+
|
4
|
+
class Swearjar
|
5
|
+
|
6
|
+
def self.default
|
7
|
+
from_language
|
8
|
+
end
|
9
|
+
|
10
|
+
def self.from_language(language = 'en')
|
11
|
+
new(File.join(File.dirname(__FILE__), 'config', "#{language}.yml"))
|
12
|
+
end
|
13
|
+
|
14
|
+
attr_reader :tester, :hash
|
15
|
+
|
16
|
+
def initialize(file = nil)
|
17
|
+
@tester = FuzzyHash.new
|
18
|
+
@hash = {}
|
19
|
+
load_file(file) if file
|
20
|
+
end
|
21
|
+
|
22
|
+
def load_file(file)
|
23
|
+
data = YAML.load_file(file)
|
24
|
+
|
25
|
+
data['regex'].each do |pattern, type|
|
26
|
+
@tester[Regexp.new(pattern)] = type
|
27
|
+
end if data['regex']
|
28
|
+
|
29
|
+
data['simple'].each do |test, type|
|
30
|
+
@hash[test] = type
|
31
|
+
end if data['simple']
|
32
|
+
end
|
33
|
+
|
34
|
+
def scan(string, &block)
|
35
|
+
string = string.to_s
|
36
|
+
string.scan(/\b[a-zA-Z-]+\b/) do |word|
|
37
|
+
block.call(word, hash[word.downcase] || hash[word.downcase.gsub(/e?s$/,'')] )
|
38
|
+
end
|
39
|
+
if match = tester.match_with_result(string)
|
40
|
+
block.call(match.last, match.first)
|
41
|
+
end
|
42
|
+
end
|
43
|
+
|
44
|
+
def profane?(string)
|
45
|
+
string = string.to_s
|
46
|
+
scan(string) {|word, test| return true if !test.nil?}
|
47
|
+
return false
|
48
|
+
end
|
49
|
+
|
50
|
+
def scorecard(string)
|
51
|
+
string = string.to_s
|
52
|
+
scorecard = {}
|
53
|
+
scan(string) {|word, test| test.each { |type| scorecard.key?(type) ? scorecard[type] += 1 : scorecard[type] = 1} if test}
|
54
|
+
scorecard
|
55
|
+
end
|
56
|
+
|
57
|
+
def censor(string)
|
58
|
+
censored_string = string.to_s.dup
|
59
|
+
scan(string) {|word, test| censored_string.gsub!(word, block_given? ? yield(word) : word.gsub(/\S/, '*')) if test}
|
60
|
+
censored_string
|
61
|
+
end
|
62
|
+
|
63
|
+
end
|
data/spec/data/swear.yml
ADDED
data/spec/spec.opts
ADDED
data/spec/spec_helper.rb
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
require File.expand_path(File.join(File.dirname(__FILE__), '..', 'lib', 'swearjar'))
|
@@ -0,0 +1,48 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Swearjar do
|
4
|
+
|
5
|
+
it "should detect dirty words" do
|
6
|
+
Swearjar.default.profane?('fuck you jim henson').should be_true
|
7
|
+
end
|
8
|
+
|
9
|
+
it "should detect dirty words regardless of case" do
|
10
|
+
Swearjar.default.profane?('FuCk you jim henson').should be_true
|
11
|
+
end
|
12
|
+
|
13
|
+
it "should not detect non-dirty words" do
|
14
|
+
Swearjar.default.profane?('i love you jim henson').should be_false
|
15
|
+
end
|
16
|
+
|
17
|
+
it "should give us a scorecard" do
|
18
|
+
Swearjar.default.scorecard('fuck you jim henson').should == {'sexual'=>1}
|
19
|
+
end
|
20
|
+
|
21
|
+
it "should detect multiword" do
|
22
|
+
Swearjar.default.scorecard('jim henson has a hard on').should == {'sexual'=>1}
|
23
|
+
end
|
24
|
+
|
25
|
+
it "should detect multiword plurals" do
|
26
|
+
Swearjar.default.scorecard('jim henson has a hard ons').should == {'sexual'=>1}
|
27
|
+
end
|
28
|
+
|
29
|
+
it "should detect simple dirty plurals" do
|
30
|
+
Swearjar.default.profane?('jim henson had two dicks').should be_true
|
31
|
+
Swearjar.default.profane?('jim henson has two asses').should be_true
|
32
|
+
end
|
33
|
+
|
34
|
+
it "should censor a string" do
|
35
|
+
Swearjar.default.censor('jim henson has a massive hard on he is gonna use to fuck everybody').should == 'jim henson has a massive **** ** he is gonna use to **** everybody'
|
36
|
+
end
|
37
|
+
|
38
|
+
it "should not do much when given a non-string" do
|
39
|
+
Swearjar.default.profane?(nil).should be_false
|
40
|
+
end
|
41
|
+
|
42
|
+
it "should allow you to load a new yaml file" do
|
43
|
+
sj = Swearjar.new
|
44
|
+
sj.load_file(File.expand_path('../data/swear.yml', __FILE__))
|
45
|
+
sj.censor("Python is the best language!").should == "****** is the best language!"
|
46
|
+
end
|
47
|
+
|
48
|
+
end
|
metadata
ADDED
@@ -0,0 +1,113 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: cf-swearjar
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 1.0.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Joshua Hull
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2012-06-08 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: fuzzyhash
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ~>
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: 0.0.11
|
22
|
+
type: :runtime
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ~>
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: 0.0.11
|
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: 0.8.7
|
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: 0.8.7
|
46
|
+
- !ruby/object:Gem::Dependency
|
47
|
+
name: rspec
|
48
|
+
requirement: !ruby/object:Gem::Requirement
|
49
|
+
none: false
|
50
|
+
requirements:
|
51
|
+
- - ~>
|
52
|
+
- !ruby/object:Gem::Version
|
53
|
+
version: 1.3.0
|
54
|
+
type: :development
|
55
|
+
prerelease: false
|
56
|
+
version_requirements: !ruby/object:Gem::Requirement
|
57
|
+
none: false
|
58
|
+
requirements:
|
59
|
+
- - ~>
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: 1.3.0
|
62
|
+
description: Put another nickel in the swearjar. Simple profanity detection with content
|
63
|
+
analysis.
|
64
|
+
email: joshbuddy@gmail.com
|
65
|
+
executables: []
|
66
|
+
extensions: []
|
67
|
+
extra_rdoc_files:
|
68
|
+
- README.rdoc
|
69
|
+
files:
|
70
|
+
- .gitignore
|
71
|
+
- Gemfile
|
72
|
+
- README.rdoc
|
73
|
+
- Rakefile
|
74
|
+
- cf-swearjar.gemspec
|
75
|
+
- lib/config/en.yml
|
76
|
+
- lib/swearjar.rb
|
77
|
+
- lib/swearjar/tester.rb
|
78
|
+
- lib/swearjar/version.rb
|
79
|
+
- spec/data/swear.yml
|
80
|
+
- spec/spec.opts
|
81
|
+
- spec/spec_helper.rb
|
82
|
+
- spec/swearjar_spec.rb
|
83
|
+
homepage: http://github.com/joshbuddy/swearjar
|
84
|
+
licenses: []
|
85
|
+
post_install_message:
|
86
|
+
rdoc_options:
|
87
|
+
- --charset=UTF-8
|
88
|
+
require_paths:
|
89
|
+
- lib
|
90
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
91
|
+
none: false
|
92
|
+
requirements:
|
93
|
+
- - ! '>='
|
94
|
+
- !ruby/object:Gem::Version
|
95
|
+
version: '0'
|
96
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
97
|
+
none: false
|
98
|
+
requirements:
|
99
|
+
- - ! '>='
|
100
|
+
- !ruby/object:Gem::Version
|
101
|
+
version: '0'
|
102
|
+
requirements: []
|
103
|
+
rubyforge_project: swearjar
|
104
|
+
rubygems_version: 1.8.24
|
105
|
+
signing_key:
|
106
|
+
specification_version: 3
|
107
|
+
summary: Put another nickel in the swearjar. Simple profanity detection with content
|
108
|
+
analysis
|
109
|
+
test_files:
|
110
|
+
- spec/data/swear.yml
|
111
|
+
- spec/spec.opts
|
112
|
+
- spec/spec_helper.rb
|
113
|
+
- spec/swearjar_spec.rb
|