filters_spam 0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/filters_spam.rb +138 -0
- data/readme.md +44 -0
- metadata +70 -0
data/lib/filters_spam.rb
ADDED
@@ -0,0 +1,138 @@
|
|
1
|
+
# Just call filters_spam in your model with any of the options you want.
|
2
|
+
def filters_spam(options = {})
|
3
|
+
options = {
|
4
|
+
:message_field => :message,
|
5
|
+
:email_field => :email,
|
6
|
+
:author_field => :author,
|
7
|
+
:other_fields => [],
|
8
|
+
:extra_spam_words => []
|
9
|
+
}.update(options)
|
10
|
+
|
11
|
+
self.module_eval do
|
12
|
+
scope :ham, lambda {{:conditions => {:spam => false}, :order => 'created_at DESC'}}
|
13
|
+
scope :spam, lambda {{:conditions => {:spam => true}, :order => 'created_at DESC'}}
|
14
|
+
before_validation(:on => :create) { calculate_spam_score }
|
15
|
+
|
16
|
+
cattr_accessor :spam_words
|
17
|
+
self.spam_words = %w{
|
18
|
+
-online 4u 4-u acne adipex advicer baccarrat blackjack bllogspot booker buy byob carisoprodol
|
19
|
+
casino chatroom cialis coolhu credit-card-debt cwas cyclen cyclobenzaprine orgy
|
20
|
+
day-trading debt-consolidation discreetordering duty-free dutyfree equityloans fioricet
|
21
|
+
freenet free\s*shipping gambling- hair-loss homefinance holdem incest jrcreations leethal levitra macinstruct
|
22
|
+
mortgagequotes nemogs online-gambling ottawavalleyag ownsthis paxil penis pharmacy phentermine
|
23
|
+
poker poze pussy ringtones roulette shemale shoes -site slot-machine thorcarlson
|
24
|
+
tramadol trim-spa ultram valeofglamorganconservatives viagra vioxx xanax zolus
|
25
|
+
} | options[:extra_spam_words]
|
26
|
+
end
|
27
|
+
|
28
|
+
self.module_eval %{
|
29
|
+
def ham?
|
30
|
+
not spam?
|
31
|
+
end
|
32
|
+
|
33
|
+
def ham!
|
34
|
+
self.update_attribute(:spam, false)
|
35
|
+
end
|
36
|
+
|
37
|
+
def spam!
|
38
|
+
self.update_attribute(:spam, true)
|
39
|
+
end
|
40
|
+
|
41
|
+
protected
|
42
|
+
|
43
|
+
def score_for_message_links
|
44
|
+
link_count = self.#{options[:message_field]}.to_s.scan(/http:/).size
|
45
|
+
link_count > 2 ? -link_count : 2
|
46
|
+
end
|
47
|
+
|
48
|
+
def score_for_message_length
|
49
|
+
if self.#{options[:message_field]}.to_s.length > 20 and self.#{options[:message_field]}.to_s.scan(/http:/).size.zero?
|
50
|
+
2
|
51
|
+
else
|
52
|
+
-1
|
53
|
+
end
|
54
|
+
end
|
55
|
+
|
56
|
+
def score_for_previous_submissions
|
57
|
+
current_score = 0
|
58
|
+
|
59
|
+
self.class.find(:all, :conditions => {:#{options[:email_field]} => #{options[:email_field]}}).each do |i|
|
60
|
+
if i.spam?
|
61
|
+
current_score -= 1
|
62
|
+
else
|
63
|
+
current_score += 1
|
64
|
+
end
|
65
|
+
end
|
66
|
+
|
67
|
+
current_score
|
68
|
+
end
|
69
|
+
|
70
|
+
def score_for_spam_words
|
71
|
+
current_score = 0
|
72
|
+
|
73
|
+
spam_words.each do |word|
|
74
|
+
regex = /\#{word}/i
|
75
|
+
if #{options[:message_field]} =~ regex ||
|
76
|
+
#{options[:author_field]} =~ regex #{" || #{options[:other_fields].join(' =~ regex ')} =~ regex" if options[:other_fields].any?}
|
77
|
+
current_score -= 1
|
78
|
+
end
|
79
|
+
end
|
80
|
+
|
81
|
+
current_score
|
82
|
+
end
|
83
|
+
|
84
|
+
def score_for_suspect_url
|
85
|
+
current_score = 0
|
86
|
+
|
87
|
+
regex = /http:\\/\\/\\S*(\\.html|\\.info|\\?|&|free)/i
|
88
|
+
current_score =- (1 * #{options[:message_field]}.to_s.scan(regex).size)
|
89
|
+
end
|
90
|
+
|
91
|
+
def score_for_suspect_tld
|
92
|
+
regex = /http:\\/\\/\\S*\\.(de|pl|cn)/i
|
93
|
+
#{options[:message_field]}.to_s.scan(regex).size * -1
|
94
|
+
end
|
95
|
+
|
96
|
+
def score_for_lame_message_start
|
97
|
+
#{options[:message_field]}.to_s.strip =~ /^(interesting|sorry|nice|cool)/i ? -10 : 0
|
98
|
+
end
|
99
|
+
|
100
|
+
def score_for_author_link
|
101
|
+
#{options[:author_field]}.to_s.scan(/http:/).size * -2
|
102
|
+
end
|
103
|
+
|
104
|
+
def score_for_same_message
|
105
|
+
self.class.count(:conditions => {:#{options[:message_field]} => #{options[:message_field]}}) * -1
|
106
|
+
end
|
107
|
+
|
108
|
+
def score_for_consonant_runs
|
109
|
+
current_score = 0
|
110
|
+
|
111
|
+
[#{([options[:author_field], options[:message_field], options[:email_field]] | options[:other_fields]).join(', ')}].each do |field|
|
112
|
+
field.to_s.scan(/[bcdfghjklmnpqrstvwxz]{5,}/).each do |run|
|
113
|
+
current_score =- run.size - 4
|
114
|
+
end
|
115
|
+
end
|
116
|
+
|
117
|
+
current_score
|
118
|
+
end
|
119
|
+
|
120
|
+
def calculate_spam_score
|
121
|
+
score = 0
|
122
|
+
score += score_for_message_links
|
123
|
+
score += score_for_message_length
|
124
|
+
score += score_for_previous_submissions
|
125
|
+
score += score_for_spam_words
|
126
|
+
score += score_for_suspect_tld
|
127
|
+
score += score_for_lame_message_start
|
128
|
+
score += score_for_author_link
|
129
|
+
score += score_for_same_message
|
130
|
+
score += score_for_consonant_runs
|
131
|
+
self.spam = (score < 0)
|
132
|
+
|
133
|
+
logger.info("spam score was \#{score}")
|
134
|
+
|
135
|
+
true
|
136
|
+
end
|
137
|
+
}
|
138
|
+
end
|
data/readme.md
ADDED
@@ -0,0 +1,44 @@
|
|
1
|
+
# FiltersSpam
|
2
|
+
|
3
|
+
This is a small Ruby on Rails plugin that can be installed as a gem in your ``Gemfile``
|
4
|
+
that allows models to attach to it to provide spam filtering functionality.
|
5
|
+
|
6
|
+
## Rails Quickstart
|
7
|
+
|
8
|
+
# Add to Gemfile
|
9
|
+
gem 'filters_spam', '~> 0.1'
|
10
|
+
|
11
|
+
bundle install
|
12
|
+
|
13
|
+
## Usage
|
14
|
+
|
15
|
+
Once you have the plugin installed, you can use it by calling the function in your model like so:
|
16
|
+
|
17
|
+
filters_spam
|
18
|
+
|
19
|
+
If you want to change the default fields that are used by ``filters_spam``
|
20
|
+
then you can pass them in to the method as options.
|
21
|
+
|
22
|
+
All of the possible options are outlined below with the default values for each:
|
23
|
+
|
24
|
+
filters_spam({
|
25
|
+
:message_field => :message,
|
26
|
+
:email_field => :email,
|
27
|
+
:author_field => :author,
|
28
|
+
:other_fields => [],
|
29
|
+
:extra_spam_words => %w()
|
30
|
+
})
|
31
|
+
|
32
|
+
So, say you wanted to mark 'ruby' and 'rails' as spam words you simply pass them
|
33
|
+
in using the ``:extra_spam_words`` option:
|
34
|
+
|
35
|
+
filters_spam({
|
36
|
+
:extra_spam_words => %w(ruby rails)
|
37
|
+
})
|
38
|
+
|
39
|
+
Enjoy a life with less spam.
|
40
|
+
|
41
|
+
## Credits
|
42
|
+
|
43
|
+
This code was inspired by Russel Norris' [acts_as_snook plugin](http://github.com/rsl/acts_as_snook)
|
44
|
+
and ideas presented by [Jonathan Snook](http://snook.ca/archives/other/effective_blog_comment_spam_blocker)
|
metadata
ADDED
@@ -0,0 +1,70 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: filters_spam
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
hash: 9
|
5
|
+
prerelease: false
|
6
|
+
segments:
|
7
|
+
- 0
|
8
|
+
- 1
|
9
|
+
version: "0.1"
|
10
|
+
platform: ruby
|
11
|
+
authors:
|
12
|
+
- Philip Arndt
|
13
|
+
- David Jones
|
14
|
+
autorequire:
|
15
|
+
bindir: bin
|
16
|
+
cert_chain: []
|
17
|
+
|
18
|
+
date: 2010-08-27 00:00:00 +12:00
|
19
|
+
default_executable:
|
20
|
+
dependencies: []
|
21
|
+
|
22
|
+
description: This is a small Ruby on Rails plugin that can be installed as a gem in your Gemfile that allows models to attach to it to provide spam filtering functionality.
|
23
|
+
email: info@resolvedigital.co.nz
|
24
|
+
executables: []
|
25
|
+
|
26
|
+
extensions: []
|
27
|
+
|
28
|
+
extra_rdoc_files: []
|
29
|
+
|
30
|
+
files:
|
31
|
+
- readme.md
|
32
|
+
- lib/filters_spam.rb
|
33
|
+
has_rdoc: true
|
34
|
+
homepage: http://www.resolvedigital.co.nz
|
35
|
+
licenses: []
|
36
|
+
|
37
|
+
post_install_message:
|
38
|
+
rdoc_options: []
|
39
|
+
|
40
|
+
require_paths:
|
41
|
+
- lib
|
42
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
43
|
+
none: false
|
44
|
+
requirements:
|
45
|
+
- - ">="
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
hash: 57
|
48
|
+
segments:
|
49
|
+
- 1
|
50
|
+
- 8
|
51
|
+
- 7
|
52
|
+
version: 1.8.7
|
53
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
54
|
+
none: false
|
55
|
+
requirements:
|
56
|
+
- - ">="
|
57
|
+
- !ruby/object:Gem::Version
|
58
|
+
hash: 3
|
59
|
+
segments:
|
60
|
+
- 0
|
61
|
+
version: "0"
|
62
|
+
requirements:
|
63
|
+
- none
|
64
|
+
rubyforge_project:
|
65
|
+
rubygems_version: 1.3.7
|
66
|
+
signing_key:
|
67
|
+
specification_version: 3
|
68
|
+
summary: Attach to your model to have this filter out the spam using scoring techniques.
|
69
|
+
test_files: []
|
70
|
+
|