clang/tools/clang-fuzzer/dictionary/dictionary.c

   1 //===-- dictionary.c - Generate fuzzing dictionary for clang --------------===//
   2 //
   3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
   4 // See https://llvm.org/LICENSE.txt for license information.
   5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
   6 //
   7 //===----------------------------------------------------------------------===//
   8 //
   9 // This binary emits a fuzzing dictionary describing strings that are
  10 // significant to the clang parser: keywords and other tokens.
  11 //
  12 // The dictionary can be used by a fuzzer to reach interesting parser states
  13 // much more quickly.
  14 //
  15 // The output is a single-file dictionary supported by libFuzzer and AFL:
  16 // https://llvm.org/docs/LibFuzzer.html#dictionaries
  17 //
  18 //===----------------------------------------------------------------------===//
  19
  20 #include <stdio.h>
  21
  22 static void emit(const char *Name, const char *Spelling) {
  23   static char Hex[] = "0123456789abcdef";
  24
  25   printf("%s=\"", Name);
  26   unsigned char C;
  27   while ((C = *Spelling++)) {
  28     if (C < 32 || C == '"' || C == '\\')
  29       printf("\\x%c%c", Hex[C>>4], Hex[C%16]);
  30     else
  31       printf("%c", C);
  32   }
  33   printf("\"\n");
  34 }
  35
  36 int main(int argc, char **argv) {
  37 #define PUNCTUATOR(Name, Spelling) emit(#Name, Spelling);
  38 #define KEYWORD(Name, Criteria) emit(#Name, #Name);
  39 #define PPKEYWORD(Name) emit(#Name, #Name);
  40 #define CXX_KEYWORD_OPERATOR(Name, Equivalent) emit(#Name, #Name);
  41 #define OBJC_AT_KEYWORD(Name) emit(#Name, #Name);
  42 #define ALIAS(Spelling, Equivalent, Criteria) emit(Spelling, Spelling);
  43 #include "clang/Basic/TokenKinds.def"
  44   // Some other sub-token chunks significant to the lexer.
  45   emit("ucn16", "\\u0000");
  46   emit("ucn32", "\\U00000000");
  47   emit("rawstart", "R\"(");
  48   emit("rawend", ")\"");
  49   emit("quote", "\"");
  50   emit("squote", "'");
  51   emit("u8quote", "u8\"");
  52   emit("u16quote", "u\"");
  53   emit("u32quote", "U\"");
  54   emit("esc_nl", "\\\n");
  55   emit("hex", "0x");
  56 }
  57