added fuzzing client

Author: firewave

.github/workflows/fuzz.yml

@@ -0,0 +1,64 @@
+# Syntax reference https://help.github.com/en/actions/reference/workflow-syntax-for-github-actions
+# Environment reference https://help.github.com/en/actions/reference/virtual-environments-for-github-hosted-runners
+name: fuzz
+on: [pull_request]
+
+permissions:
+  contents: read
+
+jobs:
+  fuzz:
+    runs-on: ubuntu-24.04
+    if: ${{ github.repository_owner == 'danmar' }}
+
+    steps:
+      - uses: actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      # the man-db trigger causes package installations to stall for several minutes at times. so just drop the package.
+      # see https://github.com/actions/runner/issues/4030
+      - name: Remove man-db package
+        run: |
+          sudo apt-get update
+          sudo apt-get remove man-db
+
+      - name: Install missing software
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y make
+
+      - name: Install clang
+        run: |
+          wget https://apt.llvm.org/llvm.sh
+          chmod +x llvm.sh
+          sudo ./llvm.sh 21
+
+      - name: Build fuzzer
+        id: build
+        run: |
+          # TODO: test O/LTO for best speed
+          # TODO: use -stdlib=libc++ -lc++
+          make -j$(nproc) CXX=clang++ CXXOPTS="-O3 -flto -fno-omit-frame-pointer -g -gline-tables-only -DFUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION -fsanitize=address,undefined -fsanitize-address-use-after-scope -fno-sanitize=integer -fno-sanitize-recover=undefined" LDOPTS="-flto" LIB_FUZZING_ENGINE="-fsanitize=fuzzer" fuzz
+        env:
+          CXX: clang-21
+
+      - name: Run fuzzer
+        run: |
+          mkdir corpus
+          mkdir artifacts
+          ./fuzz -only_ascii=1 -timeout=5 -fork=$(nproc) -use_value_profile=0 -max_total_time=60 -artifact_prefix=./artifacts/ corpus corpus_test
+
+      - name: Upload corpus
+        uses: actions/upload-artifact@v6
+        if: (success() || failure()) && steps.build.outcome == 'success'
+        with:
+          name: corpus
+          path: ./corpus
+
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v6
+        if: failure() && steps.build.outcome == 'success'
+        with:
+          name: artifacts
+          path: ./artifacts

Makefile

@@ -12,21 +12,32 @@ test.o: CPPFLAGS += $(TEST_CPPFLAGS)
 test.o: CXXFLAGS += -Wno-multichar
 
 %.o: %.cpp	simplecpp.h
-	$(CXX) $(CPPFLAGS) $(CXXFLAGS) -c $<
+	$(CXX) $(CPPFLAGS) $(CXXFLAGS) -c $< $(LIB_FUZZING_ENGINE)
+
+fuzz_no.o: fuzz.cpp
+	$(CXX) $(CPPFLAGS) $(CXXFLAGS) -DNO_FUZZ -c -o $@ $^
 
 testrunner:	test.o	simplecpp.o
-	$(CXX) $(LDFLAGS) simplecpp.o test.o -o testrunner
+	$(CXX) $(LDFLAGS) -o $@ $^
 
 test:	testrunner	simplecpp
 	./testrunner
 	python3 run-tests.py
 	python3 -m pytest integration_test.py -vv
 
+fuzz:	fuzz.o simplecpp.o
+	# TODO: use -stdlib=libc++ -lc++
+	# make fuzz CXX=clang++ CXXOPTS="-O2 -fno-omit-frame-pointer -g -gline-tables-only -DFUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION -fsanitize=address,undefined -fsanitize-address-use-after-scope -fno-sanitize=integer -fno-sanitize-recover=undefined" LIB_FUZZING_ENGINE="-fsanitize=fuzzer"
+	$(CXX) $(LDFLAGS) $(CXXFLAGS) -o $@ $^ $(LIB_FUZZING_ENGINE)
+
+no-fuzz:	fuzz_no.o simplecpp.o
+	$(CXX) $(LDFLAGS) $(CXXFLAGS) -o $@ $^
+
 selfcheck:	simplecpp
 	CXX=$(CXX) ./selfcheck.sh
 
 simplecpp:	main.o simplecpp.o
-	$(CXX) $(LDFLAGS) main.o simplecpp.o -o simplecpp
+	$(CXX) $(LDFLAGS) -o $@ $^
 
 clean:
-	rm -f testrunner simplecpp *.o
+	rm -f testrunner fuzz no-fuzz simplecpp *.o

fuzz.cpp

@@ -0,0 +1,70 @@
+/*
+ * simplecpp - A simple and high-fidelity C/C++ preprocessor library
+ * Copyright (C) 2016-2024 simplecpp team
+ */
+
+#include "simplecpp.h"
+
+#include <cstdint>
+
+#ifdef NO_FUZZ
+#include <cstdlib>
+#include <fstream>
+#include <sstream>
+#include <string>
+#endif
+
+/*
+  0 - store in corpus
+ -1 - omit from corpus
+*/
+static int doProcess(const uint8_t *data, size_t dataSize)
+{
+    simplecpp::OutputList outputList;
+    std::vector<std::string> files;
+    simplecpp::TokenList rawtokens(data, dataSize, files, "test.cpp", &outputList);
+
+    simplecpp::TokenList outputTokens(files);
+    simplecpp::FileDataCache filedata;
+    const simplecpp::DUI dui;
+    std::list<simplecpp::MacroUsage> macroUsage;
+    std::list<simplecpp::IfCond> ifCond;
+    simplecpp::preprocess(outputTokens, rawtokens, files, filedata, dui, &outputList, &macroUsage, &ifCond);
+
+    simplecpp::cleanup(filedata);
+
+    return 0;
+}
+
+#ifndef NO_FUZZ
+extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t dataSize);
+
+int LLVMFuzzerTestOneInput(const uint8_t *data, size_t dataSize)
+{
+    return doProcess(data, dataSize);
+}
+#else
+int main(int argc, char * argv[])
+{
+    if (argc < 2 || argc > 3)
+        return EXIT_FAILURE;
+
+    std::ifstream f(argv[1]);
+    if (!f.is_open())
+        return EXIT_FAILURE;
+
+    std::ostringstream oss;
+    oss << f.rdbuf();
+
+    if (!f.good())
+        return EXIT_FAILURE;
+
+    const int cnt = (argc == 3) ? std::stoi(argv[2]) : 1;
+
+    const std::string code = oss.str();
+    for (int i = 0; i < cnt; ++i)
+        doProcess(reinterpret_cast<const uint8_t*>(code.data()), code.size());
+
+    return EXIT_SUCCESS;
+}
+#endif