morito 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 CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA1:
3
- metadata.gz: 50f1f0b304f54a4d042d91ab748679af6aa8064d
4
- data.tar.gz: ada9db82752169ccfd43f7337bb97ec51d36827a
3
+ metadata.gz: 1c55decebb4c9e07810ea06f3f21d8f40088faa5
4
+ data.tar.gz: 1268725db5c75afe009299dbd5f3c29e3780ba80
5
5
  SHA512:
6
- metadata.gz: 7d864fa9b215e1b7b076b4fdba050e993ef52a402e4792dd8c882cc9f057eeb73c888eaeab624731393a2675b421801ffe7a32811719a1539f74f5d09ffcbdd3
7
- data.tar.gz: 50a9ed68a538e2f8b15caa8b26221a770a5507abc51e310d0453223eb1b313e217f6954d4c4cf432eb8c0a5c6e6f3bfd4cdcce4d93e27a9562a796088cfb83a4
6
+ metadata.gz: f36208dbab6355df0c691c3f4cef8783abbf10ae211c7b8440df5bd951380d4db4c0f452391f95e3b7a8e9b54c93f1a5aedea890edcaf495784ef2a1e0c120ee
7
+ data.tar.gz: 58dae02a4a09fa4529f6889a8da4597431a857285938c03e887813797347069ac34995dbde8ca756eec89537e4026c26f5929c4570bbc289fe27d55fa60af63c
data/README.md CHANGED
@@ -26,7 +26,8 @@ client.allowed?('http://example.com/some/path', cache: false) # => true / false
26
26
 
27
27
  ## TODO
28
28
 
29
- - Currently does not handle `Allow:` description.
29
+ - Handle `Allow:` description.
30
+ - Pattern matching like `Disallow: /*?`
30
31
 
31
32
  ## Contributing
32
33
 
data/lib/morito/client.rb CHANGED
@@ -9,11 +9,19 @@ module Morito
9
9
  def allowed?(requesting_url, cache: true)
10
10
  uri = URI.parse(requesting_url)
11
11
  robots_txt_body = robots_txt_body(uri, cache: cache)
12
- Morito::Processor.new(robots_txt_body).allowed?(@user_agent, uri.path)
12
+ Morito::Processor.new(robots_txt_body).allowed?(@user_agent, path(uri))
13
13
  end
14
14
 
15
15
  private
16
16
 
17
+ def path(uri)
18
+ if uri.query
19
+ "#{uri.path}?#{uri.query}"
20
+ else
21
+ uri.path
22
+ end
23
+ end
24
+
17
25
  def robots_txt_body(uri, cache: true)
18
26
  if cache
19
27
  with_cache(uri.host) do
@@ -41,7 +41,7 @@ module Morito
41
41
  end
42
42
 
43
43
  class LineParser
44
- attr_reader :user_agent, :disallow
44
+ attr_reader :user_agent
45
45
 
46
46
  def parse(line)
47
47
  case line
@@ -55,6 +55,10 @@ module Morito
55
55
  end
56
56
  end
57
57
 
58
+ def disallow
59
+ Regexp.escape(@disallow).gsub('\*', '.*').gsub('\$', '$')
60
+ end
61
+
58
62
  def disallow?
59
63
  @user_agent && @disallow
60
64
  end
@@ -1,3 +1,3 @@
1
1
  module Morito
2
- VERSION = "0.0.2"
2
+ VERSION = "0.0.3"
3
3
  end
@@ -13,6 +13,7 @@ describe Morito::Client do
13
13
  <<-EOS
14
14
  User-agent: *
15
15
  Disallow: /private
16
+ Disallow: /*?$
16
17
 
17
18
  # Comment
18
19
  User-agent: restricted agent # Comment
@@ -65,6 +66,20 @@ EOS
65
66
  it { should == true }
66
67
  end
67
68
  end
69
+
70
+ context 'with patternized path' do
71
+ let(:user_agent) { 'some agent' }
72
+
73
+ context 'for eol' do
74
+ let(:requesting_url) { 'http://example.com/some/path?' }
75
+ it { should == false }
76
+ end
77
+
78
+ context 'for no eol' do
79
+ let(:requesting_url) { 'http://example.com/some/path?param=1' }
80
+ it { should == true }
81
+ end
82
+ end
68
83
  end
69
84
 
70
85
  context 'with 404' do
metadata CHANGED
@@ -1,7 +1,7 @@
1
1
  --- !ruby/object:Gem::Specification
2
2
  name: morito
3
3
  version: !ruby/object:Gem::Version
4
- version: 0.0.2
4
+ version: 0.0.3
5
5
  platform: ruby
6
6
  authors:
7
7
  - negipo