rippersnapper 0.0.2 → 0.0.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/lib/rippersnapper/domain_parser.rb +2 -3
- data/lib/rippersnapper/suffix_file_reader.rb +2 -4
- data/lib/rippersnapper/url.rb +4 -5
- data/lib/rippersnapper/version.rb +1 -1
- data/spec/rippersnapper/domain_parser_spec.rb +8 -0
- data/spec/rippersnapper/url_spec.rb +11 -0
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: f532fc959d773f20cf0542ec71061e82a3d32808
|
4
|
+
data.tar.gz: a092b51bae916693b046b012816316d69f390179
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 49eaf5932ee7b4611f78c4083201bf7007984726445e5dbe7d37f26c84feb2156a1105924f59887bbc94c861c39d8de766e7eb16e27b2a2e8470c445302cbea5
|
7
|
+
data.tar.gz: 7df9d142edf0e98bae8d3db17045f2771d5a3510971937be8cf42d01a2824353f6e9ead1a701edc4e4769f5c3167a2d469572da3dc348263fda23c03c2acf3a7
|
@@ -1,6 +1,5 @@
|
|
1
1
|
module Rippersnapper
|
2
2
|
class DomainParser
|
3
|
-
attr_reader :domain, :subdomain
|
4
3
|
|
5
4
|
def initialize url
|
6
5
|
@url = url
|
@@ -13,14 +12,14 @@ module Rippersnapper
|
|
13
12
|
test = url_parts.last(iteration + 1).join(".")
|
14
13
|
found = test if suffix_exists?(test)
|
15
14
|
end
|
16
|
-
found
|
15
|
+
found.to_s
|
17
16
|
end
|
18
17
|
end
|
19
18
|
|
20
19
|
def domain
|
21
20
|
@domain ||= begin
|
22
21
|
remaining = url_parts - suffix_parts
|
23
|
-
remaining.last
|
22
|
+
remaining.last.to_s
|
24
23
|
end
|
25
24
|
end
|
26
25
|
|
@@ -9,8 +9,6 @@ module Rippersnapper
|
|
9
9
|
|
10
10
|
def initialize file = nil
|
11
11
|
@file = file || File.open(File.dirname(__FILE__) + "/public_suffix.dat", 'r')
|
12
|
-
# Recursive hash of hashes http://stackoverflow.com/a/170240/973860
|
13
|
-
#@public_suffixes = Hash.new(&(p=lambda{|h,k| h[k] = Hash.new(&p)}))
|
14
12
|
@public_suffixes = Hash.new { false }
|
15
13
|
parse_file
|
16
14
|
end
|
@@ -26,6 +24,8 @@ module Rippersnapper
|
|
26
24
|
|
27
25
|
private
|
28
26
|
|
27
|
+
# NOTE: This could be extracted into an object if something more robust is
|
28
|
+
# needed in the future
|
29
29
|
def parse_file
|
30
30
|
file.each_line do |line|
|
31
31
|
line = line.strip
|
@@ -34,7 +34,5 @@ module Rippersnapper
|
|
34
34
|
end
|
35
35
|
end
|
36
36
|
end
|
37
|
-
|
38
37
|
end
|
39
|
-
|
40
38
|
end
|
data/lib/rippersnapper/url.rb
CHANGED
@@ -2,10 +2,9 @@ require 'URI'
|
|
2
2
|
|
3
3
|
module Rippersnapper
|
4
4
|
class Url
|
5
|
-
attr_accessor :subdomain, :domain
|
6
5
|
|
7
6
|
def initialize url
|
8
|
-
@url = url
|
7
|
+
@url = url.to_s
|
9
8
|
end
|
10
9
|
|
11
10
|
def uri
|
@@ -13,7 +12,7 @@ module Rippersnapper
|
|
13
12
|
end
|
14
13
|
|
15
14
|
def url
|
16
|
-
return @url if @url =~ /:\/\//
|
15
|
+
return @url if @url =~ /:\/\// || @url.empty?
|
17
16
|
"http://#{@url}"
|
18
17
|
end
|
19
18
|
|
@@ -30,11 +29,11 @@ module Rippersnapper
|
|
30
29
|
end
|
31
30
|
|
32
31
|
def scheme
|
33
|
-
uri.scheme
|
32
|
+
uri.scheme || ""
|
34
33
|
end
|
35
34
|
|
36
35
|
def host
|
37
|
-
uri.host
|
36
|
+
uri.host || ""
|
38
37
|
end
|
39
38
|
|
40
39
|
def path
|
@@ -17,5 +17,13 @@ module Rippersnapper
|
|
17
17
|
its(:domain) { should eq "google" }
|
18
18
|
its(:suffix) { should eq "whatever.sapporo.jp" }
|
19
19
|
end
|
20
|
+
|
21
|
+
context "empty url" do
|
22
|
+
subject { DomainParser.new "" }
|
23
|
+
its(:subdomain) { should eq "" }
|
24
|
+
its(:domain) { should eq "" }
|
25
|
+
its(:suffix) { should eq "" }
|
26
|
+
end
|
27
|
+
|
20
28
|
end
|
21
29
|
end
|
@@ -42,5 +42,16 @@ module Rippersnapper
|
|
42
42
|
its(:suffix) { should eq "com" }
|
43
43
|
its(:domain) { should eq "google" }
|
44
44
|
end
|
45
|
+
|
46
|
+
context "when url is nil" do
|
47
|
+
subject {Url.new nil}
|
48
|
+
its(:url) { should eq "" }
|
49
|
+
its(:path) { should eq "" }
|
50
|
+
its(:scheme) { should eq "" }
|
51
|
+
its(:host) { should eq "" }
|
52
|
+
its(:suffix) { should eq "" }
|
53
|
+
its(:domain) { should eq "" }
|
54
|
+
its(:subdomain) { should eq "" }
|
55
|
+
end
|
45
56
|
end
|
46
57
|
end
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: rippersnapper
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0.
|
4
|
+
version: 0.0.3
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Micah Woods
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2013-04-
|
11
|
+
date: 2013-04-19 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|