Add arbitrary_witness and parse_int fuzz targets
#5029
Conversation
|
Heads up that I no longer run honggfuzz fuzz tests in my local CI because honggfuzz has been broken so long. Nice that they have a new version a couple days ago, but it's too much of a PITA for me to restore deleted code to bring it back just to test it. So thanks for updating the dep (and alerting me to the new version) but at some point we need to just cut over to cargo-fuzz. |
|
In dad80b5: You add the |
This version was released on 23 Sept 2025
|
Fixed the commit. I'll also start working on moving over to cargo-fuzz |
| let deserialized: Result<Witness, _> = deserialize(serialized.as_slice()); | ||
| assert!(deserialized.is_ok(), "Deserialization error: {:?}", deserialized.err().unwrap()); |
There was a problem hiding this comment.
Just a question, not requesting any change; what is the purpose of these two lines instead of just calling unwrap() on line 16?
There was a problem hiding this comment.
No real purpose, I just had it like that so i could quickly find Deserialization error: in the output in a previous test and just kept using it. I can change to just use .unwrap() on the deserialize output here and in the other places where it's used in a follow up so it's less verbose
There was a problem hiding this comment.
actually i'll just remove this entirely since it unwraps and checks equality in the next line
There was a problem hiding this comment.
Oh, yeah, unwrap would probably be better here. I'm just so used to fuzztests ending in assertions.
3f20d54 Remove unnecessary check (Shing Him Ng) 8e7f2d2 Move serialization round trip check before object mutation (Shing Him Ng) Pull request description: I thought I tested all of the targets I updated before I published them, but this one slipped through the cracks. Updated to move the serialization before object mutation [like before](https://github.com/rust-bitcoin/rust-bitcoin/blob/66c4ea20ff30cfa3e20f3073bd3beca3117177fd/fuzz/fuzz_targets/bitcoin/deserialize_transaction.rs#L10-L11), and I ran `cycle.sh` to make sure that there weren't any glaringly obvious bugs like the one that I pushed. Also remove some unnecessary `assert`s mentioned in #5029 ACKs for top commit: apoelstra: ACK 3f20d54; successfully ran local tests; but how come our CI was fine with these? Tree-SHA512: cdf79a0bb73a6b3d236bd567ce46c56e7eca94947c93ce7ef63146dc1256c520ec17b3d9462570753d64930381a383bf48112cd409ca0f287a30ae6da90e9cf5
3 participants
No description provided.