link_thumbnailer 2.5.0 → 2.5.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: e82e842f04354aa42e0d379f3bb7f1260f9d12ef
-  data.tar.gz: 472731d1a762495f3deef46028d68481a77f1eb3
+  metadata.gz: 8ebb211bdc1846b8bd87073ba32ad911ce291294
+  data.tar.gz: 0681efbd2d6315670e5b9888c66d27b898aa9401
 SHA512:
-  metadata.gz: 615f8cf840ce82c6080340730f33146732c14d1c32ab11d9c8d71228c9c0cec05f84a55a846f3e55e4b31df91efb0f84854c4e02f033626b95a65888fb8ebd2d
-  data.tar.gz: bb113d233c1296b407d44091d64dd50d6c031eae0eb28f5d816415abdcfea5ffb8c861ea53e36b6be9914c202254db5689f1e545d645cf457ac312f1c5dfddd5
+  metadata.gz: 7b15cff3407a6e288a97f9923279b4f926b5f8d08bd00db52881dec032501198bd8dda0c635b14fd89054c2b83f87ce4a3557a9fe57d5490af8ed6f5b3905f74
+  data.tar.gz: 576533b10d2f694ce8bbd974bc76af780a566227d37fc05d0ff82aca54c9807b521cdf56c457af06b8c9d54496aef04d0a51c21787af4479e7930f603847e48c

data/CHANGELOG.md

@@ -1,3 +1,7 @@
+# 2.5.1
+
+- Implement `Set-Cookie` header between http redirections to set cookies when site requires it. Fixes #55.
+
 # 2.5.0
 
 - Handles seamlessly `og:image` and `og:image:url`

data/README.md

@@ -83,7 +83,7 @@ object.to_json
 
 ## Configuration
 
-LinkThumnailer comes with default configuration values. You can change default value by overriding them in a rails initializer:
+LinkThumbnailer comes with default configuration values. You can change default value by overriding them in a rails initializer:
 
 In `config/initializers/link_thumbnailer.rb`
 
@@ -107,7 +107,7 @@ LinkThumbnailer.configure do |config|
   #
   # See http://www.ruby-doc.org/stdlib-2.1.1/libdoc/net/http/rdoc/Net/HTTP.html#open_timeout
   #
-  # config.http_timeout = 5
+  # config.http_open_timeout = 5
 
   # List of blacklisted urls you want to skip when searching for images.
   #
@@ -177,6 +177,16 @@ LinkThumbnailer defines a list of custom exceptions you may want to rescue in yo
 * `RedirectLimit` -- raised when redirection threshold defined in config is reached
 * `BadUriFormat` -- raised when url given is not a valid HTTP url
 
+You can rescue from any LinkThumbnailer exceptions using the following code:
+
+```ruby
+begin
+  LinkThumbnailer.generate('http://foo.com')
+rescue LinkThumbnailer::Exceptions => e
+  # do something
+end
+```
+
 ## Contributing
 
 1. Fork it

data/lib/generators/templates/initializer.rb

@@ -18,7 +18,7 @@ LinkThumbnailer.configure do |config|
   #
   # See http://www.ruby-doc.org/stdlib-2.1.1/libdoc/net/http/rdoc/Net/HTTP.html#open_timeout
   #
-  # config.http_timeout = 5
+  # config.http_open_timeout = 5
 
   # List of blacklisted urls you want to skip when searching for images.
   #

data/lib/link_thumbnailer/configuration.rb

@@ -23,10 +23,13 @@ module LinkThumbnailer
 	class Configuration
 
     attr_accessor :redirect_limit, :blacklist_urls, :user_agent,
-                  :verify_ssl, :http_timeout, :attributes, :graders,
-                  :description_min_length, :positive_regex, :negative_regex,
+                  :verify_ssl, :http_open_timeout, :http_read_timeout, :attributes,
+                  :graders, :description_min_length, :positive_regex, :negative_regex,
                   :image_limit, :image_stats
 
+    alias_method :http_timeout, :http_open_timeout
+    alias_method :http_timeout=, :http_open_timeout=
+
     # Create a new instance.
     #
     # @return [LinkThumbnailer::Configuration]
@@ -34,7 +37,8 @@ module LinkThumbnailer
       @redirect_limit         = 3
       @user_agent             = 'link_thumbnailer'
       @verify_ssl             = true
-      @http_timeout           = 5
+      @http_open_timeout      = 5
+      @http_read_timeout      = 5
       @blacklist_urls         = [
         %r{^http://ad\.doubleclick\.net/},
         %r{^http://b\.scorecardresearch\.com/},

data/lib/link_thumbnailer/model.rb

@@ -9,7 +9,7 @@ module LinkThumbnailer
 
     def sanitize(str)
       return unless str
-      str.strip.gsub(/[\r\n\f]+/, "\n")
+      str.encode('UTF-8', 'binary', invalid: :replace, undef: :replace, replace: '').strip.gsub(/[\r\n\f]+/, "\n")
     end
 
   end

data/lib/link_thumbnailer/processor.rb

@@ -15,14 +15,14 @@ module LinkThumbnailer
       super(config)
     end
 
-    def call(url = '', redirect_count = 0)
+    def call(url = '', redirect_count = 0, headers = {})
       self.url        = url
       @redirect_count = redirect_count
 
       raise ::LinkThumbnailer::RedirectLimit if too_many_redirections?
 
       with_valid_url do
-        set_http_headers
+        set_http_headers(headers)
         set_http_options
         perform_request
       end
@@ -35,23 +35,32 @@ module LinkThumbnailer
       yield if block_given?
     end
 
-    def set_http_headers
+    def set_http_headers(headers = {})
+      headers.each { |k, v| http.headers[k] = v }
       http.headers['User-Agent']               = user_agent
       http.override_headers['Accept-Encoding'] = 'none'
     end
 
     def set_http_options
       http.verify_mode  = ::OpenSSL::SSL::VERIFY_NONE unless ssl_required?
-      http.open_timeout = http_timeout
+      http.open_timeout = http_open_timeout
+      http.read_timeout = http_read_timeout
       http.proxy = :ENV
     end
 
     def perform_request
-      response = http.request(url)
+      response          = http.request(url)
+      headers           = {}
+      headers['Cookie'] = response['Set-Cookie'] if response['Set-Cookie'].present?
+
       case response
       when ::Net::HTTPSuccess then response.body
       when ::Net::HTTPRedirection
-        call resolve_relative_url(response['location']), redirect_count + 1
+        call(
+          resolve_relative_url(response['location']),
+          redirect_count + 1,
+          headers
+        )
       else
         response.error!
       end
@@ -62,7 +71,7 @@ module LinkThumbnailer
     end
 
     def build_absolute_url_for(relative_url)
-      URI("#{url.scheme}://#{url.host}#{relative_url}")
+      ::URI.parse("#{url.scheme}://#{url.host}#{relative_url}")
     end
 
     def redirect_limit
@@ -73,8 +82,12 @@ module LinkThumbnailer
       config.user_agent
     end
 
-    def http_timeout
-      config.http_timeout
+    def http_open_timeout
+      config.http_open_timeout
+    end
+
+    def http_read_timeout
+      config.http_read_timeout
     end
 
     def ssl_required?

data/lib/link_thumbnailer/version.rb

@@ -1,3 +1,3 @@
 module LinkThumbnailer
-  VERSION = '2.5.0'
+  VERSION = '2.5.1'
 end

data/spec/configuration_spec.rb

@@ -7,7 +7,8 @@ describe LinkThumbnailer::Configuration do
   it { expect(instance.redirect_limit).to         eq(3) }
   it { expect(instance.user_agent).to             eq('link_thumbnailer') }
   it { expect(instance.verify_ssl).to             eq(true) }
-  it { expect(instance.http_timeout).to           eq(5) }
+  it { expect(instance.http_open_timeout).to      eq(5) }
+  it { expect(instance.http_read_timeout).to      eq(5) }
   it { expect(instance.blacklist_urls).to_not     be_empty }
   it { expect(instance.attributes).to             eq([:title, :images, :description, :videos, :favicon]) }
   it { expect(instance.graders).to_not            be_empty }
@@ -17,6 +18,11 @@ describe LinkThumbnailer::Configuration do
   it { expect(instance.image_limit).to            eq(5) }
   it { expect(instance.image_stats).to            eq(true) }
 
+  describe "#http_timeout" do
+    it { expect(instance.method(:http_timeout)).to eq(instance.method(:http_open_timeout)) }
+    it { expect(instance.method(:http_timeout=)).to eq(instance.method(:http_open_timeout=)) }
+  end
+
   describe '.config' do
 
     it { expect(LinkThumbnailer.config).to be_a(described_class) }

data/spec/processor_spec.rb

@@ -151,17 +151,29 @@ describe LinkThumbnailer::Processor do
 
     describe 'open_timeout' do
 
-      let(:http_timeout) { 1 }
+      let(:http_open_timeout) { 1 }
 
       before do
-        instance.stub(:http_timeout).and_return(http_timeout)
+        instance.stub(:http_open_timeout).and_return(http_open_timeout)
         action
       end
 
-      it { expect(http.open_timeout).to eq(http_timeout) }
+      it { expect(http.open_timeout).to eq(http_open_timeout) }
 
     end
 
+    describe 'read_timeout' do
+
+      let(:http_read_timeout) { 1 }
+
+      before do
+        instance.stub(:http_read_timeout).and_return(http_read_timeout)
+        action
+      end
+
+      it { expect(http.read_timeout).to eq(http_read_timeout) }
+
+    end
   end
 
   describe '#perform_request' do
@@ -200,7 +212,7 @@ describe LinkThumbnailer::Processor do
 
       it 'calls call method' do
         expect(instance).to receive(:resolve_relative_url)
-        expect(instance).to receive(:call).with(new_url, instance.redirect_count + 1)
+        expect(instance).to receive(:call).with(new_url, instance.redirect_count + 1, {})
         action
       end
 
@@ -250,16 +262,29 @@ describe LinkThumbnailer::Processor do
 
   end
 
-  describe '#http_timeout' do
+  describe '#http_open_timeout' do
+
+    let(:http_open_timeout)  { 1 }
+    let(:action)        { instance.send(:http_open_timeout) }
+
+    before do
+      instance.config.http_open_timeout = http_open_timeout
+    end
+
+    it { expect(action).to eq(http_open_timeout) }
+
+  end
+
+  describe '#http_read_timeout' do
 
-    let(:http_timeout)  { 1 }
-    let(:action)        { instance.send(:http_timeout) }
+    let(:http_read_timeout)  { 1 }
+    let(:action)        { instance.send(:http_read_timeout) }
 
     before do
-      instance.config.http_timeout = http_timeout
+      instance.config.http_read_timeout = http_read_timeout
     end
 
-    it { expect(action).to eq(http_timeout) }
+    it { expect(action).to eq(http_read_timeout) }
 
   end
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: link_thumbnailer
 version: !ruby/object:Gem::Version
-  version: 2.5.0
+  version: 2.5.1
 platform: ruby
 authors:
 - Pierre-Louis Gottfrois
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-02-08 00:00:00.000000000 Z
+date: 2015-05-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport
@@ -230,7 +230,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.3
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: Ruby gem ranking images from a given URL returning an object containing images