Carbon fuzzing 3/3: added actual fuzzer implementation and a fuzzverter utility for investigating crashing protos (#1156)

* finished fuzzer and added fuzzverter util * fixed typo * renamed cmd line params * fixed libproto_mutator download path * small fixes * small fixes * small fixes * renamed sample corpus proto * small fixes * try building on github with LIBCPP_DEBUG enabled * temporarily marked proto fuzzer as a manual test * code review * use a dedicated proto-fuzzer feature to work around LIBCPP_DEBUG=1 crash in proto code * code review comments, added README.md * minor fixes to the text * Update bazel/cc_toolchains/clang_cc_toolchain_config.bzl Co-authored-by: Jon Meow <jperkins@google.com> * use Carbon source representation for "empty Main()" instead of text format proto representation * fixed typo * made FuzzerUtil produce the full carbon source (proto converted + Main if needed) to decrease code duplication a bit * typo * switched to text proto format per code review * Update executable_semantics/prelude.h Co-authored-by: Jon Meow <jperkins@google.com> * Update executable_semantics/fuzzing/README.md Co-authored-by: Jon Meow <jperkins@google.com> * Update executable_semantics/fuzzing/README.md Co-authored-by: Jon Meow <jperkins@google.com> * Update executable_semantics/fuzzing/README.md Co-authored-by: Jon Meow <jperkins@google.com> * Update executable_semantics/fuzzing/README.md Co-authored-by: Jon Meow <jperkins@google.com> * Update executable_semantics/fuzzing/README.md Co-authored-by: Jon Meow <jperkins@google.com> * review comments * removed unnecessary file mode variables * Update executable_semantics/fuzzing/fuzzverter.cpp Co-authored-by: Jon Meow <jperkins@google.com> * Update executable_semantics/fuzzing/README.md Co-authored-by: Jon Meow <jperkins@google.com> * Update executable_semantics/fuzzing/README.md Co-authored-by: Jon Meow <jperkins@google.com> * code review comments * Update executable_semantics/syntax/BUILD Co-authored-by: Jon Meow <jperkins@google.com> * buildifier Co-authored-by: Jon Meow <jperkins@google.com>
2026-10-04 19:01:05 +01:00 · 2022-04-11 16:47:03 -04:00
parent db0de1004f
commit 22462a0d7f
18 changed files with 497 additions and 20 deletions
@@ -0,0 +1,136 @@
+// Part of the Carbon Language project, under the Apache License v2.0 with LLVM
+// Exceptions. See /LICENSE for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+// An utility for converting between fuzzer protos and Carbon sources.
+//
+// For example, to convert a crashing input in text proto to carbon source:
+// `fuzzverter --mode=proto_to_carbon --input file.textproto`
+//
+// To generate a new text proto from carbon source for seeding the corpus:
+// `fuzzverter --mode=carbon_to_proto --input file.carbon`
+
+#include <google/protobuf/text_format.h>
+
+#include <cstdlib>
+#include <fstream>
+#include <ios>
+#include <sstream>
+
+#include "common/error.h"
+#include "common/fuzzing/carbon.pb.h"
+#include "executable_semantics/common/error.h"
+#include "executable_semantics/fuzzing/ast_to_proto.h"
+#include "executable_semantics/fuzzing/fuzzer_util.h"
+#include "executable_semantics/syntax/parse.h"
+#include "llvm/Support/CommandLine.h"
+#include "llvm/Support/InitLLVM.h"
+
+namespace Carbon {
+
+// Reads a file and returns its contents as a string.
+static auto ReadFile(std::string_view file_name) -> ErrorOr<std::string> {
+  std::ifstream file(file_name, std::ios::in);
+  if (!file.is_open()) {
+    return ErrorBuilder() << "Could not open " << file_name << " for reading";
+  }
+  std::stringstream ss;
+  ss << file.rdbuf();
+  return ss.str();
+}
+
+// Writes string `s` to `file_name`.
+static auto WriteFile(std::string_view s, std::string_view file_name)
+    -> ErrorOr<Success> {
+  std::ofstream file(file_name, std::ios::out);
+  if (!file.is_open()) {
+    return ErrorBuilder() << "Could not open " << file_name << " for writing";
+  }
+  file << s;
+  return Success();
+}
+
+// Converts text proto to Carbon source.
+static auto TextProtoToCarbon(std::string_view input_file_name,
+                              std::string_view output_file_name)
+    -> ErrorOr<Success> {
+  ASSIGN_OR_RETURN(const std::string input_contents, ReadFile(input_file_name));
+  Fuzzing::Carbon carbon_proto;
+  if (!google::protobuf::TextFormat::ParseFromString(input_contents,
+                                                     &carbon_proto)) {
+    return Error("Could not parse text proto");
+  }
+  const std::string carbon_source =
+      ProtoToCarbonWithMain(carbon_proto.compilation_unit());
+  return WriteFile(carbon_source, output_file_name);
+}
+
+// Converts Carbon source to text proto.
+static auto CarbonToTextProto(std::string_view input_file_name,
+                              std::string_view output_file_name)
+    -> ErrorOr<Success> {
+  Carbon::Arena arena;
+  const ErrorOr<AST> ast = Carbon::Parse(&arena, input_file_name,
+                                         /*trace=*/false);
+  if (!ast.ok()) {
+    return ErrorBuilder() << "Parsing failed: " << ast.error().message();
+  }
+  Fuzzing::Carbon carbon_proto;
+  *carbon_proto.mutable_compilation_unit() = AstToProto(*ast);
+
+  std::string proto_string;
+  google::protobuf::TextFormat::Printer p;
+  if (!p.PrintToString(carbon_proto, &proto_string)) {
+    return Error("Failed to convert to text proto");
+  }
+  return WriteFile(proto_string, output_file_name);
+}
+
+// Command line options for defining input/output format.
+enum class ConversionMode { TextProtoToCarbon, CarbonToTextProto };
+
+// Returns string representation of an enum option.
+static auto GetEnumString(llvm::cl::opt<ConversionMode>& o) -> llvm::StringRef {
+  // TODO: is there a better way?
+  return o.getParser().getOption(static_cast<int>(ConversionMode(o)));
+}
+
+// Performs the conversion specified by `mode`.
+auto Convert(const ConversionMode mode, std::string_view input_file_name,
+             std::string_view output_file_name) -> ErrorOr<Success> {
+  switch (mode) {
+    case ConversionMode::TextProtoToCarbon:
+      return TextProtoToCarbon(input_file_name, output_file_name);
+    case ConversionMode::CarbonToTextProto:
+      return CarbonToTextProto(input_file_name, output_file_name);
+  }
+}
+
+}  // namespace Carbon
+
+auto main(int argc, char* argv[]) -> int {
+  llvm::InitLLVM init_llvm(argc, argv);
+
+  llvm::cl::opt<Carbon::ConversionMode> mode(
+      "mode", llvm::cl::desc("Conversion mode"),
+      llvm::cl::values(
+          clEnumValN(Carbon::ConversionMode::TextProtoToCarbon,
+                     "proto_to_carbon", "Convert text proto to Carbon source"),
+          clEnumValN(Carbon::ConversionMode::CarbonToTextProto,
+                     "carbon_to_proto",
+                     "Convert Carbon source to text proto")));
+  llvm::cl::opt<std::string> input_file_name(
+      "input", llvm::cl::desc("<input file>"), llvm::cl::init("/dev/stdin"));
+  llvm::cl::opt<std::string> output_file_name(
+      "output", llvm::cl::desc("<output file>"), llvm::cl::init("/dev/stdout"));
+  llvm::cl::ParseCommandLineOptions(argc, argv);
+
+  if (const auto result =
+          Carbon::Convert(mode, input_file_name, output_file_name);
+      !result.ok()) {
+    llvm::errs() << GetEnumString(mode)
+                 << " conversion failed: " << result.error().message() << "\n";
+    return EXIT_FAILURE;
+  }
+  return EXIT_SUCCESS;
+}