Support basic auth in solr extract service

lib/hydra/derivatives/processors/full_text.rb

@@ -22,16 +22,23 @@ def extract
       # TODO: this pulls the whole file into memory. We should stream it from Fedora instead
       # @return [String] the result of calling the extract service
       def fetch
-        req = Net::HTTP.new(uri.host, uri.port)
-        req.use_ssl = true if check_for_ssl
-        resp = req.post(uri.to_s, file_content, request_headers)
+        resp = http_request
         raise "Solr Extract service was unsuccessful. '#{uri}' returned code #{resp.code} for #{source_path}\n#{resp.body}" unless resp.code == '200'
         file_content.rewind if file_content.respond_to?(:rewind)
+        resp.body.force_encoding(resp.type_params['charset']) if resp.type_params['charset']
+        resp.body
+      end
 
-        if resp.type_params['charset']
-          resp.body.force_encoding(resp.type_params['charset'])
+      # Send the request to the extract service
+      # @return [Net::HttpResponse] the result of calling the extract service
+      def http_request
+        Net::HTTP.start(uri.host, uri.port) do |http|
+          req = Net::HTTP::Post.new(uri.request_uri, request_headers)
+          req.use_ssl = true if check_for_ssl
+          req.basic_auth uri.user, uri.password unless uri.password.nil?
+          req.body = file_content
+          http.request req
         end
-        resp.body
       end
 
       def file_content

spec/processors/full_text_spec.rb

@@ -18,21 +18,27 @@
   describe "fetch" do
     subject { processor.send(:fetch) }
 
-    let(:request)       { double }
     let(:response_body) { 'returned by Solr' }
     let(:uri)           { URI('https://example.com:99/solr/update') }
+    let(:req) do
+      Net::HTTP::Post.new(
+        '/solr/update',
+        "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244"
+      )
+    end
 
     before do
       allow(processor).to receive(:uri).and_return(uri)
-      allow(Net::HTTP).to receive(:new).with('example.com', 99).and_return(request)
     end
 
     context "when that is successful" do
       let(:resp) { double(code: '200', type_params: {}, body: response_body) }
 
       it "calls the extraction service" do
         expect(processor).to receive(:check_for_ssl)
-        expect(request).to receive(:post).with('https://example.com:99/solr/update', String, "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244").and_return(resp)
+        expect(Net::HTTP).to receive(:start).with('example.com', 99).and_yield(req)
+        expect(Net::HTTP::Post).to receive(:new).with('/solr/update', "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244").and_return(req)
+        expect(req).to receive(:request).and_return(resp)
         expect(subject).to eq response_body
       end
     end
@@ -44,7 +50,9 @@
 
       it "calls the extraction service" do
         expect(processor).to receive(:check_for_ssl)
-        expect(request).to receive(:post).with('https://example.com:99/solr/update', String, "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244").and_return(resp)
+        expect(Net::HTTP).to receive(:start).with('example.com', 99).and_yield(req)
+        expect(Net::HTTP::Post).to receive(:new).with('/solr/update', "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244").and_return(req)
+        expect(req).to receive(:request).and_return(resp)
         expect(subject).to eq response_utf8
       end
     end
@@ -54,10 +62,43 @@
 
       it "raises an error" do
         expect(processor).to receive(:check_for_ssl)
-        expect(request).to receive(:post).with('https://example.com:99/solr/update', String, "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244").and_return(resp)
+        expect(Net::HTTP).to receive(:start).with('example.com', 99).and_yield(req)
+        expect(Net::HTTP::Post).to receive(:new).with('/solr/update', "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244").and_return(req)
+        expect(req).to receive(:request).and_return(resp)
         expect { subject }.to raise_error(RuntimeError, %r{^Solr Extract service was unsuccessful. 'https://example\.com:99/solr/update' returned code 500})
       end
     end
+
+    context "with basic_auth" do
+      let(:resp) { double(code: '200', type_params: {}, body: response_body) }
+      let(:uri) { URI('https://user:password@example.com:99/solr/update') }
+
+      it "calls the extraction service with basic auth" do
+        expect(processor).to receive(:check_for_ssl)
+        expect(Net::HTTP).to receive(:start).with('example.com', 99).and_yield(req)
+        expect(Net::HTTP::Post).to receive(:new).with('/solr/update', "Content-Type" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "Content-Length" => "24244").and_return(req)
+        expect(req).to receive(:basic_auth).with('user', 'password')
+        expect(req).to receive(:request).and_return(resp)
+        expect(subject).to eq response_body
+      end
+    end
+
+    context "with basic_auth and incorrect credentials" do
+      let(:resp) { double(code: '401', type_params: {}, body: response_body) }
+      let(:uri) { URI('https://user:password@example.com:99/solr/update') }
+
+      before do
+        allow(req).to receive(:basic_auth).with('user', 'incorrectpassword')
+      end
+
+      it "raised a 401 error" do
+        req.basic_auth('user', 'incorrectpassword')
+        expect(processor).to receive(:check_for_ssl)
+        expect(Net::HTTP).to receive(:start).with('example.com', 99).and_yield(req)
+        expect(req).to receive(:request).and_return(resp)
+        expect { subject }.to raise_error(RuntimeError, %r{^Solr Extract service was unsuccessful. 'https://user:password@example.com:99/solr/update' returned code 401})
+      end
+    end
   end
 
   describe "uri" do