920599da94 Add test for previous commit (Martin Habovstiak)
a7f3458c27 Fix bug in `ScriptBuf::extend` for short iterators (Martin Habovstiak)
Pull request description:
`ScriptBuf::extend` contained an optimization for short scripts that was
supposed to preallocate the buffer and then fill it. By mistake it
attempted to fill it from already-exhausted iterator instead of the
temporary array it drained the items into. This obviously produced
garbage (empty) values.
This was not caught by tests because the optimization is only active for
scripts with known maximum length and the test used `Instructions` which
doesn't know the maximum length.
ACKs for top commit:
sanket1729:
ACK 920599da94 . Tested that the bug is correctly fixed and tested in the new test
tcharding:
ACK 920599da94
Tree-SHA512: a80f5f262a840d8e77efd42d63c511224380ee3efa6c31855233e81c90332ac15db228e8d552d039d729d7b642e03c3939c8b6a92d3279001377515acb83abea
8ce928b8e7 Add testing section to readme (Tobin C. Harding)
2e79a0bdc4 Introduce mutation testing (Tobin C. Harding)
Pull request description:
Introduce mutation testing by way of mutagen [0] (see #1484 for context).
- Conditionally add the dev-dependency `mutagen` (using `do_mutate` flag)
This flag is not very well named but `mutagen` and `mutate` are already taken?
- Mutate all methods of the `U256` struct that do not require additional unit tests.
Uses `cfg(all(test, do_mutate), mutate)` - I cannot workout why we need to check on `test` as well i.e., I don't understand why we cannot use `cfg(do_mutate, mutate)`?
With this applied test can be run as usual with a stable toolchain. To run mutagen we use `RUSTFLAGS='--cfg=do_mutate' cargo +nightly mutagen` (doing so runs 29 mutants).
[0] https://github.com/llogiq/mutagen
ACKs for top commit:
Kixunil:
ACK 8ce928b8e7
apoelstra:
ACK 8ce928b8e7
Tree-SHA512: 024ba5d2dc983f7cd0444e09ba13280771157204999d2a44502e07a1d6436f571b75003c7cb543c943f17949b848d4070eda4e194bda774a3e41443ff79af0af
We now have a few different test harnesses in use, add a section to the
readme about each
- normal unit/integration tests
- benchmarks
- kani
- mutagen
e428486002 Add `from_bytes(Vec<u8>)` to `ScriptBuf` (Martin Habovstiak)
Pull request description:
This is useful when one already has bytes allocated in a vec that can be reused.
The change also documents that the mirror method `into_bytes()` doesn't allocate.
ACKs for top commit:
tcharding:
ACK e428486002
sanket1729:
ACK e428486002
Tree-SHA512: 257b9c2cae24fd5006b6ec0e4db4a8baac177e67308758fb6167ef90e79ce75d5553433557d0e887a3e18a8fc63036f5a2acd061528a4467f76f25b6f4852400
This fixes several API bugs:
* Use `TryFrom` instead of `From` for fallible conversions
* Move byte conversion methods from `impl_array_newtype` to
`impl_bytes_newtype`
* Add missing trait impls like `AsRef`, `Borrow`, their mutable versions
and infallible conversions from arrays
Closes#1336
`ScriptBuf::extend` contained an optimization for short scripts that was
supposed to preallocate the buffer and then fill it. By mistake it
attempted to fill it from already-exhausted iterator instead of the
temporary array it drained the items into. This obviously produced
garbage (empty) values.
This was not caught by tests because the optimization is only active for
scripts with known maximum length and the test used `Instructions` which
doesn't know the maximum lenght.
In order to get better test coverage we should not enable the secp26k1
feature "rand-std" in dev-dependencies but instead feature gate tests
that depend on this feature.
Currently we enable "secp256k1/rand-std" in the "rand" feature, this is
incorrect because it means "rand" implies "std" which it does not.
Add a "rand-std" feature that turns on "seck256k1/rand-std" and make the
"rand" feature turn on "seck256k1/rand".
Introduce mutation testing by way of mutagen [0]
- Conditionally add the dev-dependency `mutagen` (using `do_mutate`
flag)
This flag is not very well named but `mutagen` and `mutate` are already
taken?
- Mutate all methods of the `U256` struct that do not require additional
unit tests.
Uses `cfg(all(test, do_mutate), mutate)` - I cannot workout why we need
to check on `test` as well i.e., I don't understand why we cannot use
`cfg(do_mutate, mutate)`?
With this applied test can be run as usual with a stable toolchain. To
run mutagen we use `RUSTFLAGS='--cfg=do_mutate' cargo +nightly mutagen`.
[0] https://github.com/llogiq/mutagen
This is useful when one already has bytes allocated in a vec that can be
reused.
The change also documents that the mirror method `into_bytes()` doesn't
allocate.
`bitcoin-internals` contains a more performant implementation of hex
encoding than what `bitcoin_hashes` uses internally. This switches the
implementations for formatting trait implementations as a step towards
moving over completely.
The public macros are also changed to delegate to inner type which is
technically a breaking change but we will break the API anyway and the
consuers should only call the macro on the actual hash newtypes where
the inner types already have the appropriate implementations.
Apart from removing reliance on internal hex from public API this
reduces duplicated code generated and compiled. E.g. if you created 10
hash newtypes of SHA256 the formatting implementation would be
instantiated 11 times despite being the same.
To do all this some other changes were required to the hex
infrastructure. Mainly modifying `put_bytes` to accept iterator (so that
`iter().rev()` can be used) and adding a new `DisplayArray` type. The
iterator idea was invented by Tobin C. Harding, this commit just adds a
bound check and generalizes over `u8` and `&u8` returning iterators.
While it may seem that `DisplayByteSlice` would suffice it'd create and
initialize a large array even for small arrays wasting performance.
Knowing the exact length `DisplayArray` fixes this.
Another part of refactoring is changing from returning `impl Display` to
return `impl LowerHex + UpperHex`. This makes selecting casing less
annoying since the consumer no longer needs to import `Case` without
cluttering the API with convenience methods.
bae264d0c2 Add `tapscript_leaf_hash()` to `Script` (Jiri Jakes)
Pull request description:
Adds convenience method to `Script` for computing leaf hash of tapscript. Closes#1482.
The little test case is taken from `bip341_tests.json`.
ACKs for top commit:
Kixunil:
ACK bae264d0c2
apoelstra:
ACK bae264d0c2
Tree-SHA512: fb7a3a552017208decd56ca7d27eab1987a3a92aae5b8620896b3a02986c2fc13043c200ccfbadf9cfdd2d74af38b0bc25936338f55b7d318c1296acc88bf22a
6acf9ac8b8 Patch hashes and update the code (Martin Habovstiak)
Pull request description:
This patches `bitcoin_hashes` to use the version in the repository and fixes the code after removal of `Deref`.
ACKs for top commit:
tcharding:
ACK 6acf9ac8b8
apoelstra:
ACK 6acf9ac8b8
Tree-SHA512: b779fa79309f1d648020146b58e96346b67e9f76e29551cbd50251ea6bb7bcfc4c5d42f49cc7ad5660dcd0a5f6306efe96dfcd9530e4b32c62edf4af7076d830
67ca3463c0 Mention Script::is_v1_p2tr above Witness::tapscript (Casey Rodarmor)
Pull request description:
It seems useful to document that this check is also provided.
ACKs for top commit:
tcharding:
ACK 67ca3463c0
Kixunil:
ACK 67ca3463c0
apoelstra:
ACK 67ca3463c0
Tree-SHA512: b36ff89cd7bb4ffe48e1bc4fcbfe35116492f5b0f9fbce271abd83f266fdcb25f7faa459acc35f944c4bdfa2626c00c194400a11f32ad84a323a07bfa5ec1b0a
This patches `bitcoin_hashes` to use the version in the repository and
fixes the code after removal of `Deref`.
This also turns off `AS_DEPENDENCY` check with the intention to refactor
it later.
02c1cd6291 add some documentation clarifying the locktime ordering shenanigans in #1330 (Andrew Poelstra)
Pull request description:
Updates the CHANGELOG and also the doccomment on `Transaction`.
ACKs for top commit:
tcharding:
ACK 02c1cd6291
Kixunil:
ACK 02c1cd6291
sanket1729:
ACK 02c1cd6291
Tree-SHA512: e2d23a90fb1e53758449fe49a3db7ae1497a260ce7efcade4b50265fa70840db273609019590d9d0a69e1272607a6bcf37924b805b4f09909487eb0c3b91a3cd
b78ba730f2 hashes: Run clippy in ci (Tobin C. Harding)
5e67f7a7cb Remove the unnecessary explicit reference (Tobin C. Harding)
Pull request description:
Currently we only run the linter in `bitcoin/contrib/test.sh`, we should do the same in the `hashes` ci script.
- Patch 1: Fix current clippy issues in `hashes` crate
- Patch 2: Run clippy in CI for `hashes` crate
ACKs for top commit:
apoelstra:
ACK b78ba730f2
Kixunil:
ACK b78ba730f2
Tree-SHA512: f9fedcd8c1a06c715396cf6c7b752e2c9e32dbfde48c8b4bcb9ac5e701abe109ddeadc2e7466f6926f7c3ab74fa26e68b70473b4a5b5009cb4644d634707d4ea
8e428562cb Implemented unsized `Script` (Martin Habovstiak)
Pull request description:
This renames `Script` to `ScriptBuf` and adds unsized `Script` modeled
after `PathBuf`/`Path`. The change cleans up the API a bit, especially
all functions that previously accepted `&Script` now accept truly
borrowed version. Some functions that perviously accepted `&[u8]` can
now accept `&Script` because constructing it is no loger costly.
This is a more idiomatic alternative to #884 with two unavoidable lines of `unsafe` copied from `std`.
Closes#522Closes#949
(For 949, we allow users to use whichever they like but still use `ScriptBuf` in `Transaction`.)
ACKs for top commit:
apoelstra:
ACK 8e428562cb
tcharding:
ACK 8e428562cb
sanket1729:
ACK 8e428562cb. Let's get this in
Tree-SHA512: 2ff7f0b3abd1999261388b7c5075aa1caa17bdaf4538b443bd098c31565d9dc8423ae3f9a9b3cd2f0aee6a01591992dbe8c9592f9cf14ec0f7cc395f696b9a66
This renames `Script` to `ScriptBuf` and adds unsized `Script` modeled
after `PathBuf`/`Path`. The change cleans up the API a bit, especially
all functions that previously accepted `&Script` now accept truly
borrowed version. Some functions that perviously accepted `&[u8]` can
now accept `&Script` because constructing it is no loger costly.
1b15a13e5a run cargo clippy and fmt (Andrew Poelstra)
f2a5596899 examples: clean up taproot PSBT example locktime handling (Andrew Poelstra)
821842e1a1 drop Ord on absolute::LockTime; add Ord to Transaction (Andrew Poelstra)
5b7d801ee6 remove PackedLockTime type (Andrew Poelstra)
4dee116b8a delete PackedLockTime by aliasing it to LockTime (Andrew Poelstra)
fa81568fb6 locktime: add `FromHexStr` impl for `LockTime` (Andrew Poelstra)
74ff4946e4 locktime: unify serde impls (Andrew Poelstra)
Pull request description:
This is potentially a controversial PR, but hear me out. My argument is that right now `absolute::LockTime` and `PackedLockTime` are basically identical types; they can be converted between each other with `From` and they have exactly the same semantics except that `PackedLockTime` has `Ord` on it. This makes it very confusing to tell which one should be used, for example in PSBT2 where there are now extra locktime-related fields.
The motivation for having `PackedLockTime` independent of `LockTime` are:
* `PackedLockTime` is theoretically more efficient because you don't need to unpack the field, don't need to store a enum discriminate, etc. I don't buy this. If you are trying to save individual bytes in your transaction-parsing there are lots of places you'd look before messing around with locktimes, so we shouldn't privilege this specific thing.
* `PackedLockTIme` has an `Ord` impl, and removing that will have a cascading effect on transactions, blocks, etc., preventing them from being held in `BTreeMaps` etc. **My proposal**, implemented here, is to just manually impl `Ord` on `Transaction` and don't impl it on `LockTime`.
I recall some argument that we need to be able to sort miniscripts, and miniscripts might have locktimes in them, but I think this is wrong -- miniscripts always have explicitly either a `Height` or a `Time`, and there is no problem ordering these.
Closes#1455
ACKs for top commit:
sanket1729:
code review ACK 1b15a13e5a
tcharding:
ACK 1b15a13e5a
Tree-SHA512: d9776f14560c3822980e8fff8978bf8ca753035152f957a84af25d063c66e4e9df3d5cf98af38d6cb59cc52ba407282f722925b1ca670ae623ac91add3149d2f
Currently we are using the `--all` flag in `cargo` commands in the
`hashes` CI script. This flag (the deprecated version of `--workspace`)
causes cargo to run the command for the whole workspace, this is not
what we want because we run test individually for each crate using a ci
script per crate.
The effect of this patch is to reduce re-runs of tests i.e., reduce
machine usage during CI runs with no reduction of coverage - PROFIT!
b7a84d0c68 hashes: Do not implement Deref (Tobin C. Harding)
Pull request description:
Currently we implement `Deref` for hashes. From the docs [0]
> Deref should only be implemented for smart pointers to avoid confusion
Furthermore because we implement `Deref` as well as implement `internals::hex::display::DisplayHex` for slices hashes get coerced into slices and `to_lower_hex_string` can be called on them, this is incorrect because `DisplayHex` does not account for hashes that display backwards so we end up with the wrong string.
This is an API breaking change, and I have not built any other crates in our stack to check if anything breaks.
[0] https://doc.rust-lang.org/std/ops/trait.Deref.html
ACKs for top commit:
apoelstra:
ACK b7a84d0c68
sanket1729:
ACK b7a84d0c68
Tree-SHA512: 573169095dd9f9032e3f685e3b0af3b569f1cb5368e1b2f37940504fd887592f46488abcfbe84107baf9c5453c7db47bc342a6bc273ff98cb0570ffacb549c21
This still has the line
let lock_time = absolute::LockTime::from_height(psbt.unsigned_tx.lock_time.to_consensus_u32() + lock_time_delta).unwrap();
I'm unsure whether this "adding height to a locktime" concept is a
meaningful thing or just the sort of thing that shows up in example
code. Maybe we should have first-class support for it.
Note that the line, as written, depends on the fact that the original
locktime was a small blockheight. A proper function for this would
handle the exceptional case gracefully.
Currently we implement `Deref` for hashes. From the docs [0]
> Deref should only be implemented for smart pointers to avoid confusion
Furthermore because we implement `Deref` as well as implement
`internals::hex::display::DisplayHex` for slices hashes get coerced into
slices and `to_lower_hex_string` can be called on them, this is
incorrect because `DisplayHex` does not account for hashes that display
backwards so we end up with the wrong string.
[0] https://doc.rust-lang.org/std/ops/trait.Deref.html
This can be replicated by deleting the `type PackedLockTime = LockTime'
line, and then running
find . -type f | xargs sed -i 's/PackedLockTime/LockTime/g
at the root of the repo.
e00dfa9806 impl FromHexStr for structs with single u32 member (connormullett)
Pull request description:
Closes: #1112
- Adds new trait `FromStrHex` with 2 methods: `from_hex_str` and `from_hex_str_no_prefix`
- Impl new trait on each tuple struct with single u32 member. eg `Time(u32)`
As stated in the issue, grep through codebase with `\(u32\)` and `\(pub u32\)` to see all implementations and verify none were missed.
NonStandardSighashType is an error type and should never be constructed from a hex string. Therefore, it has been omitted from this change.
Tests are somewhat redundant, but cover 4 cases each. 2 happy paths, 1 for each function. 1 case for malformed/invalid hex input, and 1 for calling no_prefix without a prefix
ACKs for top commit:
Kixunil:
ACK e00dfa9806
apoelstra:
ACK e00dfa9806
Tree-SHA512: 221faef7fc1fa8fdb4cba79cfae317a0b63984937c345c6ca2287123a078f38911cdc07db7589a88b7bc6fbecf389e9bcff47952728410510ffcfc1857e0f91f
sanket1729
Jiri Jakes
Tobin C. Harding
Andrew Poelstra
Martin Habovstiak
DanGould
Calvin Kim
Byron Hambly
Daniela Brozzoni
Casey Rodarmor