In readme, there is an inconsistency between the running commands about using Graphfuzz.
After running gfuzz gen cpp schema.yaml ./output, the file schema.json will be put into ./output with fuzz_exec.cpp and fuzz_write.cpp together. Therefore, when running ./fuzz_exec, we should add --graphfuzz_schema=./output/schema.json.
However, considering the running log below is loading schema.json rather than ./output/schema.json, it is better to directly change gfuzz gen cpp schema.yaml ./output to gfuzz gen cpp schema.yaml . to maintain consistency without modifying the original runnign log.
By the way, looks like it should be -lprotobuf to link protobuf and it is unnecessary to retain -lmylib.
In readme, there is an inconsistency between the running commands about using Graphfuzz. After running
gfuzz gen cpp schema.yaml ./output, the fileschema.jsonwill be put into./outputwithfuzz_exec.cppandfuzz_write.cpptogether. Therefore, when running./fuzz_exec, we should add--graphfuzz_schema=./output/schema.json. However, considering the running log below is loadingschema.jsonrather than./output/schema.json, it is better to directly changegfuzz gen cpp schema.yaml ./outputtogfuzz gen cpp schema.yaml .to maintain consistency without modifying the original runnign log. By the way, looks like it should be-lprotobufto link protobuf and it is unnecessary to retain-lmylib.