Add GeneratedFile concept

ql/src/filters/ClassifyFiles.ql

@@ -8,17 +8,6 @@
 
 import go
 
-string generatorCommentRegex() {
-  result = "Generated By\\b.*\\bDo not edit" or
-  result = "This (file|class|interface|art[ei]fact) (was|is|(has been)) (?:auto[ -]?)?gener(e?)ated" or
-  result = "Any modifications to this file will be lost" or
-  result =
-    "This (file|class|interface|art[ei]fact) (was|is) (?:mechanically|automatically) generated" or
-  result = "The following code was (?:auto[ -]?)?generated (?:by|from)" or
-  result = "Autogenerated by Thrift" or
-  result = "(Code g|G)enerated from .* by ANTLR"
-}
-
 predicate classify(File f, string category) {
   // tests
   f instanceof TestFile and
@@ -29,13 +18,7 @@ predicate classify(File f, string category) {
   category = "library"
   or
   // generated code
-  exists(Comment c | c.getFile() = f |
-    c.getText().regexpMatch("(?i).*\\b(" + concat(generatorCommentRegex(), "|") + ")\\b.*")
-    or
-    // regular expression recommended for Go code generators
-    // (https://golang.org/pkg/cmd/go/internal/generate/)
-    c.getText().regexpMatch("^\\s*Code generated .* DO NOT EDIT\\.\\s*$")
-  ) and
+  f instanceof GeneratedFile and
   category = "generated"
 }
 

ql/src/semmle/go/Concepts.qll

@@ -6,8 +6,8 @@
 
 import go
 import semmle.go.dataflow.FunctionInputsAndOutputs
-
 import semmle.go.concepts.HTTP
+import semmle.go.concepts.GeneratedFile
 
 /**
  * A data-flow node that executes an operating system command,

ql/src/semmle/go/concepts/GeneratedFile.qll

@@ -0,0 +1,50 @@
+/** Provides a class for generated files. */
+
+import go
+
+/** Provides a class for generated files. */
+module GeneratedFile {
+  /**
+   * A file that has been generated.
+   *
+   * Extend this class to model new APIs. If you want to refine existing API models,
+   * extend `GeneratedFile` instead.
+   */
+  abstract class Range extends File { }
+
+  private string generatorCommentRegex() {
+    result = "Generated By\\b.*\\bDo not edit" or
+    result =
+      "This (file|class|interface|art[ei]fact) (was|is|(has been)) (?:auto[ -]?)?gener(e?)ated" or
+    result = "Any modifications to this file will be lost" or
+    result =
+      "This (file|class|interface|art[ei]fact) (was|is) (?:mechanically|automatically) generated" or
+    result = "The following code was (?:auto[ -]?)?generated (?:by|from)" or
+    result = "Autogenerated by Thrift" or
+    result = "(Code g|G)enerated from .* by ANTLR"
+  }
+
+  private class CommentHeuristicGeneratedFile extends Range {
+    CommentHeuristicGeneratedFile() {
+      exists(Comment c | c.getFile() = this |
+        c.getText().regexpMatch("(?i).*\\b(" + concat(generatorCommentRegex(), "|") + ")\\b.*")
+        or
+        // regular expression recommended for Go code generators
+        // (https://golang.org/pkg/cmd/go/internal/generate/)
+        c.getText().regexpMatch("^\\s*Code generated .* DO NOT EDIT\\.\\s*$")
+      )
+    }
+  }
+}
+
+/**
+ * A file that has been generated.
+ *
+ * Extend this class to refine existing API models. If you want to model new APIs,
+ * extend `GeneratedFile::Range` instead.
+ */
+class GeneratedFile extends File {
+  GeneratedFile::Range self;
+
+  GeneratedFile() { this = self }
+}