Merge pull request #2232 from pchaigno/cython-generated-files

Detect Cython generated files

lib/linguist/generated.rb

@@ -62,6 +62,7 @@ module Linguist
       generated_parser? ||
       generated_net_docfile? ||
       generated_postscript? ||
+      compiled_cython_file? ||
       generated_protocol_buffer_go? ||
       generated_protocol_buffer? ||
       generated_jni_header? ||
@@ -270,5 +271,17 @@ module Linguist
       # VCR Cassettes have "recorded_with: VCR" in the second last line.
       return lines[-2].include?("recorded_with: VCR")
     end
+
+    # Internal: Is this a compiled C/C++ file from Cython?
+    #
+    # Cython-compiled C/C++ files typically contain:
+    # /* Generated by Cython x.x.x on ... */
+    # on the first line.
+    #
+    # Return true or false
+    def compiled_cython_file?
+      return false unless ['.c', '.cpp'].include? extname
+      return lines[0].include?("Generated by Cython")
+    end
   end
 end

test/test_blob.rb

@@ -240,6 +240,10 @@ class TestBlob < Minitest::Test
     # Godep saved dependencies
     assert sample_blob("Godeps/Godeps.json").generated?
     assert sample_blob("Godeps/_workspace/src/github.com/kr/s3/sign.go").generated?
+
+    # Cython-generated C/C++
+    assert sample_blob("C/sgd_fast.c").generated?
+    assert sample_blob("C++/wrapper_inner.cpp").generated?
   end
 
   def test_vendored