morgoth-alexa 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.document +5 -0
- data/.gitignore +5 -0
- data/LICENSE +20 -0
- data/README.rdoc +16 -0
- data/Rakefile +58 -0
- data/VERSION +1 -0
- data/alexa.gemspec +51 -0
- data/lib/alexa/url_info.rb +62 -0
- data/lib/alexa.rb +21 -0
- data/test/alexa_test.rb +96 -0
- data/test/fixtures/polsl.xml +1275 -0
- data/test/test_helper.rb +16 -0
- metadata +75 -0
data/.document
ADDED
data/LICENSE
ADDED
@@ -0,0 +1,20 @@
|
|
1
|
+
Copyright (c) 2009 Wojciech Wnętrzak
|
2
|
+
|
3
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
4
|
+
a copy of this software and associated documentation files (the
|
5
|
+
"Software"), to deal in the Software without restriction, including
|
6
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
7
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
8
|
+
permit persons to whom the Software is furnished to do so, subject to
|
9
|
+
the following conditions:
|
10
|
+
|
11
|
+
The above copyright notice and this permission notice shall be
|
12
|
+
included in all copies or substantial portions of the Software.
|
13
|
+
|
14
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
15
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
16
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
17
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
18
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
19
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
20
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.rdoc
ADDED
@@ -0,0 +1,16 @@
|
|
1
|
+
= Alexa Web Information Service
|
2
|
+
|
3
|
+
== Usage
|
4
|
+
Alexa.url_info(:access_key_id => 'key', :secret_access_key => 'secret', :host => 'site.com')
|
5
|
+
# also you can specify option :response_group => 'Rank,ContactInfo' or any other valid group,
|
6
|
+
# see: http://docs.amazonwebservices.com/AlexaWebInfoService/2005-07-11/
|
7
|
+
# default response group takes all the available options
|
8
|
+
|
9
|
+
returns object with methods:
|
10
|
+
:xml_response, :rank, :data_url, :site_title, :site_description, :language_locale, :language_encoding,
|
11
|
+
:links_in_count, :keywords, :related_links, :speed_median_load_time, :speed_percentile,
|
12
|
+
:rank_by_country, :rank_by_city, :usage_statistics
|
13
|
+
|
14
|
+
== Copyright
|
15
|
+
|
16
|
+
Copyright (c) 2009 Wojciech Wnętrzak. See LICENSE for details.
|
data/Rakefile
ADDED
@@ -0,0 +1,58 @@
|
|
1
|
+
# encoding: UTF-8
|
2
|
+
require 'rubygems'
|
3
|
+
require 'rake'
|
4
|
+
|
5
|
+
begin
|
6
|
+
require 'jeweler'
|
7
|
+
Jeweler::Tasks.new do |gem|
|
8
|
+
gem.name = "alexa"
|
9
|
+
gem.summary = %Q{Alexa Web Information Service library}
|
10
|
+
gem.email = "w.wnetrzak@gmail.com"
|
11
|
+
gem.homepage = "http://github.com/morgoth/alexa"
|
12
|
+
gem.authors = ["Wojciech Wnętrzak"]
|
13
|
+
gem.add_dependency('xml-simple')
|
14
|
+
# gem is a Gem::Specification... see http://www.rubygems.org/read/chapter/20 for additional settings
|
15
|
+
end
|
16
|
+
|
17
|
+
rescue LoadError
|
18
|
+
puts "Jeweler (or a dependency) not available. Install it with: sudo gem install jeweler"
|
19
|
+
end
|
20
|
+
|
21
|
+
require 'rake/testtask'
|
22
|
+
Rake::TestTask.new(:test) do |test|
|
23
|
+
test.libs << 'lib' << 'test'
|
24
|
+
test.pattern = 'test/**/*_test.rb'
|
25
|
+
test.verbose = true
|
26
|
+
end
|
27
|
+
|
28
|
+
begin
|
29
|
+
require 'rcov/rcovtask'
|
30
|
+
Rcov::RcovTask.new do |test|
|
31
|
+
test.libs << 'test'
|
32
|
+
test.pattern = 'test/**/*_test.rb'
|
33
|
+
test.verbose = true
|
34
|
+
end
|
35
|
+
rescue LoadError
|
36
|
+
task :rcov do
|
37
|
+
abort "RCov is not available. In order to run rcov, you must: sudo gem install spicycode-rcov"
|
38
|
+
end
|
39
|
+
end
|
40
|
+
|
41
|
+
|
42
|
+
task :default => :test
|
43
|
+
|
44
|
+
require 'rake/rdoctask'
|
45
|
+
Rake::RDocTask.new do |rdoc|
|
46
|
+
if File.exist?('VERSION.yml')
|
47
|
+
config = YAML.load(File.read('VERSION.yml'))
|
48
|
+
version = "#{config[:major]}.#{config[:minor]}.#{config[:patch]}"
|
49
|
+
else
|
50
|
+
version = ""
|
51
|
+
end
|
52
|
+
|
53
|
+
rdoc.rdoc_dir = 'rdoc'
|
54
|
+
rdoc.title = "alexa #{version}"
|
55
|
+
rdoc.rdoc_files.include('README*')
|
56
|
+
rdoc.rdoc_files.include('lib/**/*.rb')
|
57
|
+
end
|
58
|
+
|
data/VERSION
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
0.0.1
|
data/alexa.gemspec
ADDED
@@ -0,0 +1,51 @@
|
|
1
|
+
# -*- encoding: utf-8 -*-
|
2
|
+
|
3
|
+
Gem::Specification.new do |s|
|
4
|
+
s.name = %q{alexa}
|
5
|
+
s.version = "0.0.1"
|
6
|
+
|
7
|
+
s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
|
8
|
+
s.authors = ["Wojciech Wnętrzak"]
|
9
|
+
s.date = %q{2009-07-03}
|
10
|
+
s.email = %q{w.wnetrzak@gmail.com}
|
11
|
+
s.extra_rdoc_files = [
|
12
|
+
"LICENSE",
|
13
|
+
"README.rdoc"
|
14
|
+
]
|
15
|
+
s.files = [
|
16
|
+
".document",
|
17
|
+
".gitignore",
|
18
|
+
"LICENSE",
|
19
|
+
"README.rdoc",
|
20
|
+
"Rakefile",
|
21
|
+
"VERSION",
|
22
|
+
"alexa.gemspec",
|
23
|
+
"lib/alexa.rb",
|
24
|
+
"lib/alexa/url_info.rb",
|
25
|
+
"test/alexa_test.rb",
|
26
|
+
"test/fixtures/polsl.xml",
|
27
|
+
"test/test_helper.rb"
|
28
|
+
]
|
29
|
+
s.homepage = %q{http://github.com/morgoth/alexa}
|
30
|
+
s.rdoc_options = ["--charset=UTF-8"]
|
31
|
+
s.require_paths = ["lib"]
|
32
|
+
s.rubygems_version = %q{1.3.4}
|
33
|
+
s.summary = %q{Alexa Web Information Service library}
|
34
|
+
s.test_files = [
|
35
|
+
"test/alexa_test.rb",
|
36
|
+
"test/test_helper.rb"
|
37
|
+
]
|
38
|
+
|
39
|
+
if s.respond_to? :specification_version then
|
40
|
+
current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
|
41
|
+
s.specification_version = 3
|
42
|
+
|
43
|
+
if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
|
44
|
+
s.add_runtime_dependency(%q<xml-simple>, [">= 0"])
|
45
|
+
else
|
46
|
+
s.add_dependency(%q<xml-simple>, [">= 0"])
|
47
|
+
end
|
48
|
+
else
|
49
|
+
s.add_dependency(%q<xml-simple>, [">= 0"])
|
50
|
+
end
|
51
|
+
end
|
@@ -0,0 +1,62 @@
|
|
1
|
+
module Alexa
|
2
|
+
class UrlInfo
|
3
|
+
RESPONSE_GROUP = "Rank,ContactInfo,AdultContent,Speed,Language,Keywords,OwnedDomains,LinksInCount,SiteData,RelatedLinks,RankByCountry,RankByCity,UsageStats"
|
4
|
+
attr_accessor :access_key_id, :secret_access_key, :host, :response_group, :xml_response,
|
5
|
+
:rank, :data_url, :site_title, :site_description, :language_locale, :language_encoding,
|
6
|
+
:links_in_count, :keywords, :related_links, :speed_median_load_time, :speed_percentile,
|
7
|
+
:rank_by_country, :rank_by_city, :usage_statistics
|
8
|
+
|
9
|
+
def initialize(options = {} )
|
10
|
+
@access_key_id = options[:access_key_id]
|
11
|
+
@secret_access_key = options[:secret_access_key]
|
12
|
+
@host = options[:host]
|
13
|
+
@response_group = options[:response_group] || RESPONSE_GROUP
|
14
|
+
end
|
15
|
+
|
16
|
+
def connect
|
17
|
+
action = "UrlInfo"
|
18
|
+
timestamp = ( Time::now ).utc.strftime("%Y-%m-%dT%H:%M:%S.000Z")
|
19
|
+
signature = generate_signature(secret_access_key, action, timestamp)
|
20
|
+
url = generate_url(action, access_key_id, signature, timestamp, response_group, host)
|
21
|
+
@xml_response = Net::HTTP.get(url)
|
22
|
+
end
|
23
|
+
|
24
|
+
def parse_xml(xml)
|
25
|
+
xml = XmlSimple.xml_in(xml.force_encoding(Encoding::UTF_8), 'ForceArray' => false)
|
26
|
+
@rank = xml['Response']['UrlInfoResult']['Alexa']['TrafficData']['Rank'].to_i
|
27
|
+
@data_url = xml['Response']['UrlInfoResult']['Alexa']['TrafficData']['DataUrl']['content']
|
28
|
+
@site_title = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['SiteData']['Title']
|
29
|
+
@site_description = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['SiteData']['Description']
|
30
|
+
@language_locale = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['Language']['Locale']
|
31
|
+
@language_encoding = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['Language']['Encoding']
|
32
|
+
@links_in_count = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['LinksInCount'].to_i
|
33
|
+
@keywords = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['Keywords']['Keyword']
|
34
|
+
@related_links = xml['Response']['UrlInfoResult']['Alexa']['Related']['RelatedLinks']['RelatedLink']
|
35
|
+
@speed_median_load_time = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['Speed']['MedianLoadTime'].to_i
|
36
|
+
@speed_percentile = xml['Response']['UrlInfoResult']['Alexa']['ContentData']['Speed']['Percentile'].to_i
|
37
|
+
@rank_by_country = xml['Response']['UrlInfoResult']['Alexa']['TrafficData']['RankByCountry']['Country']
|
38
|
+
@rank_by_city = xml['Response']['UrlInfoResult']['Alexa']['TrafficData']['RankByCity']['City']
|
39
|
+
@usage_statistics = xml['Response']['UrlInfoResult']['Alexa']['TrafficData']['UsageStatistics']["UsageStatistic"]
|
40
|
+
end
|
41
|
+
|
42
|
+
private
|
43
|
+
|
44
|
+
def generate_signature(secret_acces_key, action, timestamp)
|
45
|
+
Base64.encode64( OpenSSL::HMAC.digest( OpenSSL::Digest::Digest.new( "sha1" ), secret_access_key, action + timestamp)).strip
|
46
|
+
end
|
47
|
+
|
48
|
+
def generate_url(action, access_key_id, signature, timestamp, response_group, host)
|
49
|
+
url = URI.parse(
|
50
|
+
"http://awis.amazonaws.com/?" +
|
51
|
+
{
|
52
|
+
"Action" => action,
|
53
|
+
"AWSAccessKeyId" => access_key_id,
|
54
|
+
"Signature" => signature,
|
55
|
+
"Timestamp" => timestamp,
|
56
|
+
"ResponseGroup" => response_group,
|
57
|
+
"Url" => host
|
58
|
+
}.to_a.collect{|item| item.first + "=" + CGI::escape(item.last) }.join("&") # Put key value pairs into http GET format
|
59
|
+
)
|
60
|
+
end
|
61
|
+
end
|
62
|
+
end
|
data/lib/alexa.rb
ADDED
@@ -0,0 +1,21 @@
|
|
1
|
+
#/usr/bin/ruby
|
2
|
+
require "cgi"
|
3
|
+
require "base64"
|
4
|
+
require "openssl"
|
5
|
+
require "digest/sha1"
|
6
|
+
require "uri"
|
7
|
+
require "net/https"
|
8
|
+
require "xmlsimple"
|
9
|
+
require "time"
|
10
|
+
|
11
|
+
require 'alexa/url_info'
|
12
|
+
|
13
|
+
module Alexa
|
14
|
+
def self.url_info(options = {})
|
15
|
+
url_info = Alexa::UrlInfo.new(options)
|
16
|
+
xml = url_info.connect
|
17
|
+
url_info.parse_xml(xml)
|
18
|
+
url_info
|
19
|
+
end
|
20
|
+
end
|
21
|
+
|
data/test/alexa_test.rb
ADDED
@@ -0,0 +1,96 @@
|
|
1
|
+
require 'test_helper'
|
2
|
+
|
3
|
+
class AlexaTest < Test::Unit::TestCase
|
4
|
+
context "Alexa::UrlInfo" do
|
5
|
+
setup do
|
6
|
+
@alexa = Alexa::UrlInfo.new(
|
7
|
+
:access_key_id => "12345678901234567890",
|
8
|
+
:secret_access_key => "qwertyuiopasdfghjklzxcvbnmQWERTYUIOPASDF",
|
9
|
+
:host => "some.host"
|
10
|
+
)
|
11
|
+
end
|
12
|
+
|
13
|
+
should "Generate signature" do
|
14
|
+
signature = @alexa.send :generate_signature, @alexa.secret_access_key, "UrlInfo", '2009-07-03T07:22:24.000Z'
|
15
|
+
assert_equal "I1mPdBy+flhhzqqUaamNq9gq190=", signature
|
16
|
+
end
|
17
|
+
|
18
|
+
should "Generatee url" do
|
19
|
+
url = @alexa.send( :generate_url,
|
20
|
+
"UrlInfo",
|
21
|
+
@alexa.access_key_id,
|
22
|
+
"I1mPdBy+flhhzqqUaamNq9gq190=",
|
23
|
+
'2009-07-03T07:22:24.000Z',
|
24
|
+
"Rank,ContactInfo,AdultContent,Speed,Language,Keywords,OwnedDomains,LinksInCount,SiteData,RelatedLinks",
|
25
|
+
"heroku.com"
|
26
|
+
)
|
27
|
+
expected_uri = "/?Action=UrlInfo&AWSAccessKeyId=12345678901234567890&Signature=I1mPdBy%2BflhhzqqUaamNq9gq190%3D&Timestamp=2009-07-03T07%3A22%3A24.000Z&ResponseGroup=Rank%2CContactInfo%2CAdultContent%2CSpeed%2CLanguage%2CKeywords%2COwnedDomains%2CLinksInCount%2CSiteData%2CRelatedLinks&Url=heroku.com"
|
28
|
+
assert_equal expected_uri, url.request_uri
|
29
|
+
assert_equal "awis.amazonaws.com", url.host
|
30
|
+
end
|
31
|
+
|
32
|
+
context "should parse xml and" do
|
33
|
+
setup do
|
34
|
+
@xml = fixture_file('polsl.xml')
|
35
|
+
@alexa.parse_xml(@xml)
|
36
|
+
end
|
37
|
+
|
38
|
+
should "return rank" do
|
39
|
+
assert_equal 86020, @alexa.rank
|
40
|
+
end
|
41
|
+
|
42
|
+
should "return data url" do
|
43
|
+
assert_equal "polsl.pl", @alexa.data_url
|
44
|
+
end
|
45
|
+
|
46
|
+
should "return site title" do
|
47
|
+
assert_equal "Silesian University of Technology", @alexa.site_title
|
48
|
+
end
|
49
|
+
|
50
|
+
should "return site description" do
|
51
|
+
assert_equal "About the university, studies, faculties and departments, photo gallery.", @alexa.site_description
|
52
|
+
end
|
53
|
+
|
54
|
+
should "return language locale" do
|
55
|
+
assert_equal "pl-PL", @alexa.language_locale
|
56
|
+
end
|
57
|
+
|
58
|
+
should "return language encoding" do
|
59
|
+
assert_equal "iso-8859-2", @alexa.language_encoding
|
60
|
+
end
|
61
|
+
|
62
|
+
should "return links in count" do
|
63
|
+
assert_equal 281, @alexa.links_in_count
|
64
|
+
end
|
65
|
+
|
66
|
+
should "return keywords" do
|
67
|
+
assert_equal ["Polska", "Regionalne", "Gliwice"], @alexa.keywords
|
68
|
+
end
|
69
|
+
|
70
|
+
should "return related links" do
|
71
|
+
assert_equal 10, @alexa.related_links.count
|
72
|
+
end
|
73
|
+
|
74
|
+
should "return speed_median load time" do
|
75
|
+
assert_equal 266, @alexa.speed_median_load_time
|
76
|
+
end
|
77
|
+
|
78
|
+
should "return speed percentile" do
|
79
|
+
assert_equal 98, @alexa.speed_percentile
|
80
|
+
end
|
81
|
+
|
82
|
+
should "return rank by country" do
|
83
|
+
assert_equal 3, @alexa.rank_by_country.count
|
84
|
+
end
|
85
|
+
|
86
|
+
should "return rank by city" do
|
87
|
+
assert_equal 68, @alexa.rank_by_city.count
|
88
|
+
end
|
89
|
+
|
90
|
+
should "return usage statistics" do
|
91
|
+
assert_equal 4, @alexa.usage_statistics.count
|
92
|
+
end
|
93
|
+
|
94
|
+
end
|
95
|
+
end
|
96
|
+
end
|