Parse spdx id from known sources

4 files changed, 46 insertions, 8 deletions

diff --git a/Gemfile.lock b/Gemfile.lock
index 9f6acfe..4459fda 100644
--- a/Gemfile.lock
+++ b/Gemfile.lock
@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    license-management (1.7.0)
+    license-management (1.7.3)
       license_finder (= 5.6.2)
 
 GEM
diff --git a/lib/license/management/repository.rb b/lib/license/management/repository.rb
index 0c428dd..fdff30d 100644
--- a/lib/license/management/repository.rb
+++ b/lib/license/management/repository.rb
@@ -25,6 +25,8 @@ module License
       attr_reader :spdx_data, :compatibility_data
 
       def spdx_data_for(id)
+        return if blank?(id)
+
         data = spdx_data[id]
         if data
           {
@@ -40,7 +42,7 @@ module License
 
       def id_for(license)
         ids = compatibility_data['ids']
-        ids[license.send(:short_name)] || ids[license.url]
+        ids[license.send(:short_name)] || ids[license.url] || known_sources(license.url)
       end
 
       # When `license_finder` is unable to determine the license it will use the full
@@ -54,9 +56,10 @@ module License
 
       def generate_item_for(license)
         log_info("detected unknown license named `#{license.send(:short_name)}`")
+        name = take_first_line_from(license.name)
         {
-          'id' => 'unknown',
-          'name' => take_first_line_from(license.name),
+          'id' => name.downcase,
+          'name' => name,
           'url' => present?(license.url) ? license.url : ''
         }
       end
@@ -71,6 +74,17 @@ module License
           memo
         end
       end
+
+      def known_sources(url)
+        return unless url =~ /\A#{::URI::DEFAULT_PARSER.make_regexp(['http', 'https'])}\z/
+
+        uri = URI.parse(url)
+        return unless ['opensource.org', 'licenses.nuget.org'].include?(uri.host)
+
+        uri.path.split('/')[-1]
+      rescue
+        nil
+      end
     end
   end
 end
diff --git a/normalized-licenses.yml b/normalized-licenses.yml
index 8b7640f..3fbef8d 100644
--- a/normalized-licenses.yml
+++ b/normalized-licenses.yml
@@ -7,14 +7,11 @@ ids:
   BSD: BSD-4-Clause
   CC0 1.0 Universal: CC0-1.0
   CC01: CC0-1.0
+  Common Public License Version 1.0: CPL-1.0
   EPL1: EPL-1.0
   GPLv2: GPL-2.0
   GPLv3: GPL-3.0
-  http://opensource.org/licenses/Apache-2.0: Apache-2.0
-  http://opensource.org/licenses/LGPL-2.1: LGPL-2.1
-  https://licenses.nuget.org/MIT: MIT
   http://www.apache.org/licenses/LICENSE-2.0: Apache-2.0
-  http://www.opensource.org/licenses/mit-license.php: MIT
   ISC: ISC
   LGPL 2.1: LGPL-2.1
   LGPL2_1: LGPL-2.1
diff --git a/spec/license/management/repository_spec.rb b/spec/license/management/repository_spec.rb
new file mode 100644
index 0000000..dbc0a22
--- /dev/null
+++ b/spec/license/management/repository_spec.rb
@@ -0,0 +1,27 @@
+RSpec.describe License::Management::Repository do
+  describe "#item_for" do
+    let(:spdx_licenses) { JSON.parse(IO.read('spdx-licenses.json'))['licenses'] }
+
+    context "when mapping a license that refers to opensource.org" do
+      it 'parses the SPDX id from the url' do
+        spdx_licenses.each do |license|
+          spdx_id = license['licenseId']
+          url = "https://opensource.org/licenses/#{spdx_id}"
+          license = LicenseFinder::License.new(short_name: url, matcher: LicenseFinder::License::NoneMatcher.new, url: url)
+          expect(subject.item_for(license)['id']).to eql(spdx_id)
+        end
+      end
+    end
+
+    context "when mapping a license that refers to nuget.org" do
+      it 'parses the SPDX id from the url' do
+        spdx_licenses.each do |license|
+          spdx_id = license['licenseId']
+          url = "https://licenses.nuget.org/#{spdx_id}"
+          license = LicenseFinder::License.new(short_name: url, matcher: LicenseFinder::License::NoneMatcher.new, url: url)
+          expect(subject.item_for(license)['id']).to eql(spdx_id)
+        end
+      end
+    end
+  end
+end