url_trimmer 0.0.1 → 0.0.2
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/url_trimmer.rb +5 -3
- data/lib/url_trimmer/version.rb +1 -1
- data/spec/lib/url_trimmer_spec.rb +4 -4
- 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: 208db8066a5128f36d5d2a23435569e278346eb4
|
|
4
|
+
data.tar.gz: 353abe12c38767112e252f3d709fc78ce1a0fa2e
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 626f3ad8f74050eb434382f8f40cdc2830dfac10812044db58ba99d0e9a0662eed8e78f705c25d5a32ce2b1ee6e405e6accc37e5d36df59a38f7ec71291a668c
|
|
7
|
+
data.tar.gz: b20cafc045eb1cf0cad0b8b2cb9e6fe60bb6c36de85efb51d0571959b9c41f008a1d49a044c1f087d30e7d35c5492a90646590de40c400b98f2a621c91f5a349
|
data/lib/url_trimmer.rb
CHANGED
|
@@ -3,6 +3,8 @@ require "domain_name"
|
|
|
3
3
|
|
|
4
4
|
module URLTrimmer
|
|
5
5
|
class Worker
|
|
6
|
+
URL_REGEXP = %r(\Ahttps?://([^/]+))
|
|
7
|
+
|
|
6
8
|
def self.uniq_by_domain(urls)
|
|
7
9
|
urls.map! do |url|
|
|
8
10
|
begin
|
|
@@ -11,9 +13,9 @@ module URLTrimmer
|
|
|
11
13
|
url.encode("UTF-8", invalid: :replace, undef: :replace, replace: "").downcase
|
|
12
14
|
end
|
|
13
15
|
end
|
|
14
|
-
urls.
|
|
15
|
-
urls.
|
|
16
|
-
urls.
|
|
16
|
+
urls.select! { |url| url =~ URL_REGEXP }
|
|
17
|
+
urls.uniq! { |url| DomainName(url[URL_REGEXP, 1]).domain }
|
|
18
|
+
urls.sort!
|
|
17
19
|
urls
|
|
18
20
|
end
|
|
19
21
|
end
|
data/lib/url_trimmer/version.rb
CHANGED
|
@@ -5,16 +5,16 @@ module URLTrimmer
|
|
|
5
5
|
describe Worker do
|
|
6
6
|
let(:urls) do
|
|
7
7
|
["http://www.google.com.ar/blah1", "https://www.google.com.ar/blah2", "https://www.google.com.br/blah3",
|
|
8
|
-
"http://www.google.com/blah4" "https://plus.google.com/blah5"]
|
|
8
|
+
"http://www.google.com/blah4", "https://plus.google.com/blah5"]
|
|
9
9
|
end
|
|
10
10
|
|
|
11
11
|
it "returns a list of unique URLs by domain" do
|
|
12
12
|
unique_urls = Worker.uniq_by_domain(urls)
|
|
13
13
|
|
|
14
14
|
unique_urls.size.must_equal 3
|
|
15
|
-
unique_urls.must_include("http://www.google.com.ar")
|
|
16
|
-
unique_urls.must_include("https://www.google.com.br")
|
|
17
|
-
unique_urls.must_include("http://www.google.com")
|
|
15
|
+
unique_urls.must_include("http://www.google.com.ar/blah1")
|
|
16
|
+
unique_urls.must_include("https://www.google.com.br/blah3")
|
|
17
|
+
unique_urls.must_include("http://www.google.com/blah4")
|
|
18
18
|
end
|
|
19
19
|
end
|
|
20
20
|
end
|
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: url_trimmer
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.0.
|
|
4
|
+
version: 0.0.2
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Cristian Rasch
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: bin
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2014-06-
|
|
11
|
+
date: 2014-06-30 00:00:00.000000000 Z
|
|
12
12
|
dependencies:
|
|
13
13
|
- !ruby/object:Gem::Dependency
|
|
14
14
|
name: domain_name
|