licensee 8.7.0 → 8.8.0

data/spec/integration_spec.rb

@@ -37,6 +37,22 @@ RSpec.describe 'integration test' do
           end
         end
 
+        context 'with CC-BY-NC-SA' do
+          let(:fixture) { 'cc-by-nc-sa' }
+
+          it 'matches nothing' do
+            expect(subject.license).to eql(nil)
+          end
+        end
+
+        context 'with CC-BY-ND' do
+          let(:fixture) { 'cc-by-nd' }
+
+          it 'matches nothing' do
+            expect(subject.license).to eql(nil)
+          end
+        end
+
         context 'MPL with HRs removed' do
           let(:license) { Licensee::License.find('mpl-2.0') }
           let(:fixture) { 'mpl-without-hrs' }
@@ -45,6 +61,15 @@ RSpec.describe 'integration test' do
             expect(subject.license).to eql(license)
           end
         end
+
+        context 'GPL3 with instructions removed' do
+          let(:license) { Licensee::License.find('gpl-3.0') }
+          let(:fixture) { 'gpl3-without-instructions' }
+
+          it 'matches to GPL3' do
+            expect(subject.license).to eql(license)
+          end
+        end
       end
 
       context 'with the license file stubbed' do

data/spec/licensee/content_helper_spec.rb

@@ -3,6 +3,8 @@ class ContentHelperTestHelper
   attr_accessor :content
 
   DEFAULT_CONTENT = <<-EOS.freeze
+The MIT License
+
 Copyright 2016 Ben Balter
 
 The made
@@ -32,12 +34,12 @@ RSpec.describe Licensee::ContentHelper do
   end
 
   it 'knows the length delta' do
-    expect(subject.length_delta(mit)).to eql(1012)
+    expect(subject.length_delta(mit)).to eql(1000)
     expect(subject.length_delta(subject)).to eql(0)
   end
 
   it 'knows the similarity' do
-    expect(subject.similarity(mit)).to be_within(1).of(4)
+    expect(subject.similarity(mit)).to be_within(1).of(2)
     expect(subject.similarity(subject)).to eql(100.0)
   end
 
@@ -70,6 +72,33 @@ RSpec.describe Licensee::ContentHelper do
       expect(normalized_content).to_not match(/\n/)
     end
 
+    Licensee::License.all(hidden: true).each do |license|
+      context license.name do
+        it 'strips the title' do
+          regex = /\A#{license.name_without_version}/i
+          expect(license.content_normalized).to_not match(regex)
+        end
+
+        it 'strips the version' do
+          expect(license.content_normalized).to_not match(/\Aversion/i)
+        end
+
+        it 'strips the copyright' do
+          expect(license.content_normalized).to_not match(/\Acopyright/i)
+        end
+
+        it 'strips the implementation instructions' do
+          end_terms_regex = /END OF TERMS AND CONDITIONS/i
+          expect(license.content_normalized).to_not match(end_terms_regex)
+          expect(license.content_normalized).to_not match(/How to apply/i)
+        end
+      end
+    end
+
+    it 'strips the title' do
+      expect(normalized_content).to_not match('MIT')
+    end
+
     it 'normalize the content' do
       expect(normalized_content).to eql 'the made up license.'
     end

data/spec/licensee/license_spec.rb

@@ -168,6 +168,11 @@ RSpec.describe Licensee::License do
       expect(mit).to_not be_gpl
       expect(gpl).to be_gpl
     end
+
+    it 'knows if a license is CC' do
+      expect(gpl).to_not be_creative_commons
+      expect(cc_by).to be_creative_commons
+    end
   end
 
   context 'content' do
@@ -180,7 +185,7 @@ RSpec.describe Licensee::License do
     end
 
     it 'computes the hash' do
-      expect(mit.hash).to eql('750260c322080bab4c19fd55eb78bc73e1ae8f11')
+      expect(mit.hash).to eql('d64f3bb4282a97b37454b5bb96a8a264a3363dc3')
     end
 
     context 'with content stubbed' do

data/spec/licensee/matchers/copyright_matcher_spec.rb

@@ -25,14 +25,15 @@ RSpec.describe Licensee::Matchers::Copyright do
     'UTF-8 Encoded'         => 'Copyright (c) 2010-2014 Simon Hürlimann',
     'Comma-separated date'  => 'Copyright (c) 2003, 2004 Ben Balter',
     'Hyphen-separated date' => 'Copyright (c) 2003-2004 Ben Balter',
-    'ASCII-8BIT encoded'    => "Copyright \xC2\xA92015 Ben Balter`"
+    'ASCII-8BIT encoded'    => "Copyright \xC2\xA92015 Ben Balter`",
+    'No year'               => 'Copyright Ben Balter'
       .force_encoding('ASCII-8BIT')
   }.each do |description, notice|
     context "with a #{description} notice" do
       let(:content) { notice }
 
       it 'matches' do
-        expect(subject.match).to eql(no_license)
+        expect(content).to be_detected_as(no_license)
       end
     end
   end

data/spec/licensee/matchers/dice_matcher_spec.rb

@@ -1,6 +1,9 @@
 RSpec.describe Licensee::Matchers::Dice do
+  let(:mit) { Licensee::License.find('mit') }
   let(:gpl) { Licensee::License.find('gpl-3.0') }
   let(:agpl) { Licensee::License.find('agpl-3.0') }
+  let(:cc_by) { Licensee::License.find('cc-by-4.0') }
+  let(:cc_by_sa) { Licensee::License.find('cc-by-sa-4.0') }
   let(:content) { sub_copyright_info(gpl.content) }
   let(:file) { Licensee::Project::LicenseFile.new(content, 'LICENSE.txt') }
   subject { described_class.new(file) }
@@ -19,12 +22,12 @@ RSpec.describe Licensee::Matchers::Dice do
 
   it 'sorts licenses by similarity' do
     expect(subject.licenses_by_similiarity[0]).to eql([gpl, 100.0])
-    expect(subject.licenses_by_similiarity[1]).to eql([agpl, 94.06571848945562])
+    expect(subject.licenses_by_similiarity[1]).to eql([agpl, 95.73361082206036])
   end
 
   it 'returns a list of licenses above the confidence threshold' do
     expect(subject.licenses_by_similiarity[0]).to eql([gpl, 100.0])
-    expect(subject.licenses_by_similiarity[1]).to eql([agpl, 94.06571848945562])
+    expect(subject.licenses_by_similiarity[1]).to eql([agpl, 95.73361082206036])
   end
 
   it 'returns the match confidence' do
@@ -40,4 +43,42 @@ RSpec.describe Licensee::Matchers::Dice do
       expect(subject.confidence).to eql(0)
     end
   end
+
+  context 'stacked licenses' do
+    let(:content) do
+      sub_copyright_info(mit.content) + "\n\n" + sub_copyright_info(gpl.content)
+    end
+
+    it "doesn't match" do
+      skip 'Stacked MIT + GPL not properly detected'
+      expect(content).to_not be_detected_as(gpl)
+      expect(subject.match).to eql(nil)
+      expect(subject.matches).to be_empty
+      expect(subject.confidence).to eql(0)
+    end
+  end
+
+  context 'CC false positive' do
+    context 'CC-BY' do
+      let(:content) { cc_by.content }
+
+      it 'matches' do
+        expect(content).to be_detected_as(cc_by)
+      end
+    end
+
+    context 'CC-ND' do
+      let(:project_path) { fixture_path('cc-by-nd') }
+      let(:license_path) { File.expand_path('LICENSE', project_path) }
+      let(:content) { File.read(license_path) }
+
+      it "doesn't match" do
+        expect(content).to_not be_detected_as(cc_by)
+        expect(content).to_not be_detected_as(cc_by_sa)
+        expect(subject.match).to be_nil
+        expect(subject.matches).to be_empty
+        expect(subject.confidence).to eql(0)
+      end
+    end
+  end
 end

data/spec/licensee/matchers/dist_zilla_matcher_spec.rb

@@ -0,0 +1,32 @@
+RSpec.describe Licensee::Matchers::DistZilla do
+  let(:mit) { Licensee::License.find('mit') }
+  let(:content) { 'license = MIT' }
+  let(:file) { Licensee::Project::LicenseFile.new(content, 'dist.ini') }
+  subject { described_class.new(file) }
+
+  it 'stores the file' do
+    expect(subject.file).to eql(file)
+  end
+
+  it 'has confidence' do
+    expect(subject.confidence).to eql(90)
+  end
+
+  it 'matches' do
+    expect(subject.match).to eql(mit)
+  end
+
+  {
+    'spdx name'     => ['license = MIT', 'mit'],
+    'non spdx name' => ['license = Mozilla_2_0', 'mpl-2.0']
+  }.each do |description, license_declaration_and_key|
+    context "with a #{description}" do
+      let(:content) { license_declaration_and_key[0] }
+      let(:license) { Licensee::License.find(license_declaration_and_key[1]) }
+
+      it 'matches' do
+        expect(subject.match).to eql(license)
+      end
+    end
+  end
+end

data/spec/licensee/matchers/exact_matcher_spec.rb

@@ -9,7 +9,7 @@ RSpec.describe Licensee::Matchers::Exact do
   end
 
   it 'matches' do
-    expect(subject.match).to eql(mit)
+    expect(content).to be_detected_as(mit)
   end
 
   it 'is confident' do

data/spec/licensee/project_files/license_file_spec.rb

@@ -18,12 +18,12 @@ RSpec.describe Licensee::Project::LicenseFile do
   end
 
   it 'creates the wordset' do
-    expect(subject.wordset.count).to eql(93)
-    expect(subject.wordset.first).to eql('mit')
+    expect(subject.wordset.count).to eql(91)
+    expect(subject.wordset.first).to eql('permission')
   end
 
   it 'creates the hash' do
-    expect(subject.hash).to eql('750260c322080bab4c19fd55eb78bc73e1ae8f11')
+    expect(subject.hash).to eql('d64f3bb4282a97b37454b5bb96a8a264a3363dc3')
   end
 
   context 'filename scoring' do
@@ -119,4 +119,40 @@ RSpec.describe Licensee::Project::LicenseFile do
       end
     end
   end
+
+  context 'CC false positives' do
+    let(:regex) { Licensee::Project::LicenseFile::CC_FALSE_POSITIVE_REGEX }
+
+    it "knows MIT isn't a potential false positive" do
+      expect(subject.content).to_not match(regex)
+      expect(subject).to_not be_a_potential_false_positive
+    end
+
+    context 'a CC false positive without creative commons in the title' do
+      let(:content) { 'Creative Commons Attribution-NonCommercial 4.0' }
+
+      it "knows it's a potential false positive" do
+        expect(subject.content).to match(regex)
+        expect(subject).to be_a_potential_false_positive
+      end
+    end
+
+    context 'a CC false positive without creative commons in the title' do
+      let(:content) { 'Attribution-NonCommercial 4.0 International' }
+
+      it "knows it's a potential false positive" do
+        expect(subject.content).to match(regex)
+        expect(subject).to be_a_potential_false_positive
+      end
+    end
+
+    context 'CC-BY-ND' do
+      let(:content) { 'Attribution-NoDerivatives 4.0 International' }
+
+      it "knows it's a potential false positive" do
+        expect(subject.content).to match(regex)
+        expect(subject).to be_a_potential_false_positive
+      end
+    end
+  end
 end

data/spec/licensee/project_files/package_info_spec.rb

@@ -8,6 +8,7 @@ RSpec.describe Licensee::Project::PackageInfo do
       'licensee.gemspec' => 1.0,
       'package.json'     => 1.0,
       'DESCRIPTION'      => 0.9,
+      'dist.ini'         => 0.8,
       'bower.json'       => 0.75,
       'README.md'        => 0.0
     }.each do |filename, expected_score|
@@ -39,6 +40,14 @@ RSpec.describe Licensee::Project::PackageInfo do
       end
     end
 
+    context 'with dist.ini' do
+      let(:filename) { 'dist.ini' }
+
+      it 'returns the DistZilla matcher' do
+        expect(possible_matchers).to eql([Licensee::Matchers::DistZilla])
+      end
+    end
+
     context 'with DESCRIPTION' do
       let(:filename) { 'DESCRIPTION' }
       let(:content) { 'Package: test' }

data/spec/spec_helper.rb

@@ -38,10 +38,6 @@ end
 
 def wrap(text, line_width = 80)
   text = text.clone
-  copyright = /^#{Licensee::Matchers::Copyright::REGEX}$/i.match(text)
-  if copyright
-    text.gsub!(/^#{Licensee::Matchers::Copyright::REGEX}$/i, '[COPYRIGHT]')
-  end
   text.gsub!(/([^\n])\n([^\n])/, '\1 \2')
 
   text = text.split("\n").collect do |line|
@@ -51,7 +47,7 @@ def wrap(text, line_width = 80)
       line
     end
   end * "\n"
-  text.gsub! '[COPYRIGHT]', "\n#{copyright}\n" if copyright
+
   text.strip
 end
 
@@ -65,14 +61,52 @@ def add_random_words(string, count = 5)
   string
 end
 
+# Init git dir
+# Note: we disable gpgsign and restore it to its original setting to avoid
+# Signing commits during tests and slowing down / breaking specs
 def git_init(path)
   Dir.chdir path do
     `git init`
+    gpgsign = `git config --local commit.gpgsign`
+    `git config --local commit.gpgsign false`
     `git add .`
     `git commit -m 'initial commit'`
+    `git config --local commit.gpgsign #{gpgsign}`
   end
 end
 
+def format_percent(float)
+  "#{format('%.2f', float)}%"
+end
+
 RSpec::Matchers.define :be_an_existing_file do
   match { |path| File.exist?(path) }
 end
+
+RSpec::Matchers.define :be_detected_as do |expected|
+  match do |actual|
+    @expected_as_array = [expected.content]
+    license_file = Licensee::Project::LicenseFile.new(actual, 'LICENSE')
+    return false unless license_file.license
+    values_match? expected, license_file.license
+  end
+
+  failure_message do |actual|
+    license_file = Licensee::Project::LicenseFile.new(actual, 'LICENSE')
+    license_name = expected.meta['spdx-id'] || expected.key
+    similarity = expected.similarity(license_file)
+    msg = "Expected the content to match the #{license_name} license"
+    msg << " (#{format_percent(similarity)} similarity"
+    msg << "using the #{license_file.matcher} matcher)"
+  end
+
+  failure_message_when_negated do |actual|
+    license_file = Licensee::Project::LicenseFile.new(actual, 'LICENSE')
+    license_name = expected.meta['spdx-id'] || expected.key
+    similarity = expected.similarity(license_file)
+    msg = "Expected the content to *not* match the #{license_name} license"
+    msg << " (#{format_percent(similarity)} similarity)"
+  end
+
+  diffable
+end

data/spec/vendored_license_spec.rb

@@ -1,7 +1,7 @@
 RSpec.describe 'vendored licenes' do
   let(:filename) { 'LICENSE.txt' }
   let(:license_file) { Licensee::Project::LicenseFile.new(content, filename) }
-  let(:detected_license) { license_file.license }
+  let(:detected_license) { license_file.license if license_file }
   let(:wtfpl) { Licensee::License.find('wtfpl') }
 
   Licensee.licenses(hidden: true).each do |license|
@@ -12,11 +12,11 @@ RSpec.describe 'vendored licenes' do
       let(:content) { content_with_copyright }
 
       it 'detects the license' do
-        expect(detected_license).to eql(license)
+        expect(content).to be_detected_as(license)
       end
 
       context 'when modified' do
-        let(:line_length) { 50 }
+        let(:line_length) { 60 }
         let(:random_words) { 3 }
         let(:content_rewrapped) { wrap(content_with_copyright, line_length) }
         let(:content_with_random_words) do
@@ -24,23 +24,21 @@ RSpec.describe 'vendored licenes' do
         end
 
         context 'without the title' do
-          let(:content) do
-            content = content_with_copyright.sub(/\A.*license\n/i, '')
-            content.sub(/\A#{license.name}/i, '')
-          end
+          let(:content) { wtfpl.send :strip_title, content_with_copyright }
 
           it 'detects the license' do
-            # WTFPL is too short to be mofifed and still be detected
-            expect(detected_license).to eql(license) unless license == wtfpl
+            skip 'The WTFPL is too short to be modified' if license == wtfpl
+            expect(content).to be_detected_as(license)
           end
         end
 
         context 'with a double title' do
-          let(:content) { "#{license.name}\n\n#{content_with_copyright}" }
+          let(:content) do
+            "#{license.name.sub('*', 'u')}\n\n#{content_with_copyright}"
+          end
 
           it 'detects the license' do
-            # WTFPL is too short to be mofifed and still be detected
-            expect(detected_license).to eql(license) unless license == wtfpl
+            expect(content).to be_detected_as(license)
           end
         end
 
@@ -48,7 +46,7 @@ RSpec.describe 'vendored licenes' do
           let(:content) { content_rewrapped }
 
           it 'detects the license' do
-            expect(detected_license).to eql(license)
+            expect(content).to be_detected_as(license)
           end
         end
 
@@ -56,8 +54,8 @@ RSpec.describe 'vendored licenes' do
           let(:content) { content_with_random_words }
 
           it 'detects the license' do
-            # WTFPL is too short to be mofifed and still be detected
-            expect(detected_license).to eql(license) unless license == wtfpl
+            skip 'The WTFPL is too short to be modified' if license == wtfpl
+            expect(content).to be_detected_as(license)
           end
         end
 
@@ -65,8 +63,8 @@ RSpec.describe 'vendored licenes' do
           let(:content) { wrap(content_with_random_words, line_length) }
 
           it 'detects the license' do
-            # WTFPL is too short to be mofifed and still be detected
-            expect(detected_license).to eql(license) unless license == wtfpl
+            skip 'The WTFPL is too short to be modified' if license == wtfpl
+            expect(content).to be_detected_as(license)
           end
         end
       end