An internal function had a non-unsafe signature but could be called
with data that would cause it to exhibit UB. Move the unsafety inside
of the function so that the function signature now enforces soundness.
Fixes#481
Clippy default settings seemed to have changed introducing a few new
warnings.
warning: variable does not need to be mutable
warning: deref on an immutable reference
warning: returning the result of a `let` binding from a block
Fix them all in a single patch because CI has to pass for each patch.
a431edb86a Create configuration conditional bench (Tobin C. Harding)
2a1c9ab4b8 Remove rand-std feature from unstable (Tobin C. Harding)
ddc108c117 Increase heading size (Tobin C. Harding)
596adff8ba Remove unneeded whitespace (Tobin C. Harding)
Pull request description:
As we did in rust-bitcoin [0] create a configuration conditional `bench`
that we can use to guard bench mark code. This has the benefit of
making our features additive i.e., we can now test with `--all-features`
with a stable toolchain (currently this fails because of our use of the
`test` crate).
Please note, this patch maintains the current behaviour of turning on
the `recovery` and `rand-std` features when benching although I was
unable to ascertain why this is needed.
[0] - https://github.com/rust-bitcoin/rust-bitcoin/pull/1092
ACKs for top commit:
sanket1729:
ACK a431edb86a.
apoelstra:
ACK a431edb86a
Tree-SHA512: 913f5fbe0da08ec649081bf237c1d31cee58dacdac251d6030afabb99d455286c6d1dbdb6b2ac892b5d3c24584933254d1cfeec8e12f531cc420bd9d455a6531
This causes panics. We can't add catch the panic, we can't change its output, we
can't detect if it'll happen, etc. Rather than dealing with confused bug reports
let's just drop this.
If users want to rerandomize their contexts they can do so manually.
There is probably a better solution to this but it is still under debate, even
upstream in the C library, what this should look like. Meanwhile we have bug
reports now.
As we did in rust-bitcoin [0] create a configuration conditional `bench`
that we can use to guard bench mark code. This has the benefit of
making our features additive i.e., we can now test with `--all-features`
with a stable toolchain (currently this fails because of our use of the
`test` crate).
[0] - https://github.com/rust-bitcoin/rust-bitcoin/pull/1092
56f18430ff Add must_use for mut self key manipulation methods (Tobin C. Harding)
5b86e38aea Put compiler attributes below rustdocs (Tobin C. Harding)
Pull request description:
We recently added a bunch of key tweaking methods that take `mut self`
and return the tweaked/negated keys. These functions are pure and as
such the returned result is expected to be used. To help downstream
users use the API correctly add `must_use` attributes with a descriptive
error string for each of the methods that takes `mut self`.
Patch 1 is preparatory cleanup.
ACKs for top commit:
apoelstra:
ACK 56f18430ff
Tree-SHA512: 95ee63d5d0a34a9915551471d2f71de1963875eda04bf4217544076be0ed2836dcdee1875432dba5e02678556af86d7487e39daac6e928083807661430ddbcd6
Currently the following command fails
`RUSTFLAGS='--cfg=fuzzing' RUSTDOCFLAGS='--cfg=fuzzing' cargo test --all --all-features`
This is because `fuzzing` is not a feature, we should be using `fuzzing`
directly not `feature = "fuzzing"`.
I have no idea how this got past CI.
We recently added a bunch of key tweaking methods that take `mut self`
and return the tweaked/negated keys. These functions are pure and as
such the returned result is expected to be used. To help downstream
users use the API correctly add `must_use` attributes with a descriptive
error string for each of the methods that takes `mut self`.
e275166652 derive Hash for RecoverableSignature (NicolaLS)
Pull request description:
It would be nice to also derive Hash for `RecoverableSignature` so data structures containing it don't have to implement it themself if they need to derive Hash
ACKs for top commit:
apoelstra:
ACK e275166652
Tree-SHA512: 418337e16e82a5e736c54d123450fdb164f4776db68952cf8095b36c501436446542821d554fa781dffa0f9067fc2464833a6c461897e655ff4449018da12ca2
`Default` was pointless, so it was replaced with internal
`from_raw_parts` method which also checks the length.
This commit also documents changes to `SerializedSignature`.
Closes#454
This de-clutters the code and prepares for the next step of adding
`IntoIterator`. The type is still re-exported so the change is neither
breaking nor inconvenient.
This also adds more datialed explanation of `SerializedSignature` and
why it's needed.
65186e732a Add githooks (Tobin C. Harding)
6d76bd4a89 Add clippy to CI (Tobin C. Harding)
9f1ebb93cb Allow nonminimal_bool in unit test (Tobin C. Harding)
685444c342 Use "a".repeats() instead of manual implementation (Tobin C. Harding)
42de876e01 Allow let_and_return for feature guarded code (Tobin C. Harding)
d64132cd4b Allow missing_safety_doc (Tobin C. Harding)
2cb687fc69 Use to_le_bytes instead of mem::transmute (Tobin C. Harding)
c15b9d2699 Remove unneeded explicit reference (Tobin C. Harding)
35d59e7cc6 Remove explicit 'static lifetime (Tobin C. Harding)
1a582db160 Remove redundant import (Tobin C. Harding)
Pull request description:
The first 8 patches clear clippy warnings. Next we add a CI job to run clippy. Finally we add a `githooks` directory that includes running clippy, also adds a section to the README on how to use the githooks. This is identical to the text in the [open PR](https://github.com/rust-bitcoin/rust-bitcoin/pull/1044) on `rust-bitcoin` that adds githooks _without_ yet adding clippy.
**Note**: The new clippy CI job runs and is green :)
ACKs for top commit:
Kixunil:
ACK 65186e732a
apoelstra:
ACK 65186e732a
Tree-SHA512: f70a157896ce2a83af8cfc10f2fbacc8f68256ac96ef7dec4d190aa72324b568d2267418eb4fe99099aeda5486957c31070943d7c209973859b7b9290676ccd7
Clippy emits:
warning: returning the result of a `let` binding from a block
This is due to feature guarded code, add 'allow' attribute. Use
`cfg_attr` to restrict the allow to when it is needed. Add the already
present `unused_mut` inside the `cfg_attr` guard also.
We have a whole bunch of unsafe code that calls down to the FFI layer.
It would be nice to have clippy running on CI, these safety docs
warnings are prohibiting that. Until we can add the docs add a compiler
attribute to allow the lint.
Since we bumped the MSRV we have `to_le_bytes` available now, use it
instead of `mem::transmute`.
Remove code comment suggesting to do exactly this and clear clippy
warning.
While we are at it, use `cast` instead of `as`.
13af51926a Make key comparison non-fuzzable (Dr Maxim Orlovsky)
739660499b Implement PublicKey ordering using FFI (Dr Maxim Orlovsky)
0faf404f0e Benchmark for key ordering (Dr Maxim Orlovsky)
999d165c68 FFI for pubkey comparison ops (Dr Maxim Orlovsky)
Pull request description:
Re-base #309 for @dr-orlovsky on request by @Kixunil.
To do the rebase I just had to change instances of cfg_attr to use `v0_5_0` instead of `v0_4_1` e.g.,
```
#[cfg_attr(not(rust_secp_no_symbol_renaming), link_name = "rustsecp256k1_v0_5_0_xonly_pubkey_cmp")]
```
And drop the changes to `src/schnorrsig.rs`, all these changes are covered by the changes in `key.rs` I believe.
ACKs for top commit:
Kixunil:
ACK 13af51926a
apoelstra:
ACK 13af51926a
Tree-SHA512: 3054fcbc1707679f54466cdc91162c286394ad691e4f5c8ee18635a22b0854a4e60f1186ef3ca1532aacd8a637d0a153601ec203947e9e58dfcebf1bcb619955
12d4583638 Implement negate that consumes self (Tobin Harding)
5eb2d745b7 Rename tweak_add_assign -> add_tweak (Tobin Harding)
b9d08db8eb Replace _assign with _tweak (Tobin Harding)
Pull request description:
The various `_assign` methods (`add_assign`, `add_expr_assign`, `mul_assign`, `tweak_add_assign`) are cumbersome to use because a local variable that uses these methods changes meaning but keeps the same identifier. It would be more useful if we had methods that consumed `self` and returned the newly modified type.
We notice also that this API is for adding/multiplying tweaks not arbitraryly adding keys.
- Patch 1: Changes add/mul_assign -> add/mul_tweak for `PublicKey` and `SecretKey` (incl. re-working unit tests)
- Patch 2: Changes `tweak_add_assign` -> `add_tweak` for `KeyPair` and `XOnlyPublicKey`
- Patch 3: Changes `negate_assign` -> `negate`
All methods changed include:
- New method consumes self and returns the tweaked key
- Original method remains with a `deprecated` attribute, however I've left a TODO in there for adding the `since` field.
Close: #415
ACKs for top commit:
apoelstra:
ACK 12d4583638
Tree-SHA512: 026e8722892f3a0f18956281e4d2356d2789ef535a7ab71a375758201b180663d068397cde2dca5f60858ab7158069e53d7096326bfbd5a364269b0be680940c
3ca7f499e0 Add fixed-width-serde integration tests (Tobin Harding)
bf9f556225 Add rustdocs describing fixed width serde (Tobin Harding)
c28808c5a4 Improve rustdocs for KeyPair (Tobin Harding)
6842383161 Use fixed width serde impls for keys (Tobin Harding)
Pull request description:
Currently we serialize keys using the `BytesVisitor`, this causes the serialized data to contain additional metadata encoding the length (an extra 8 bytes) when serialized with [bincode.](https://docs.rs/bincode/latest/bincode/index.html). This extra data is unnecessary since we know in advance the length of these two types.
We do not control the data output by serialization of our types because it depends on which crate is used to do the serialization. This PR improves the situation for serialization using the `bincode` crate, and this PR introduces mentions of `bincode` in the rustdocs, is this acceptable? See below for a table that describes binary serialization by other crates.
Implement a sequence based visitor that encodes the keys as fixed width data for:
- `SecretKey`
- `PublicKey`
- `KeyPair`
- `XOnlyPublicKey`
Fixes: #295
**Question**: PR only does keys, do we want to do signatures as well?
ACKs for top commit:
apoelstra:
ACK 3ca7f499e0
Tree-SHA512: 77babce74fa9f0981bb3b869c4e77a68a4d1ec28d22d2c3be4305e27ef01d4828dac210e20b968cbbe5de8a0563cd985d7969bccf75cfe627a34a116fed1a5df
Feature guard the custom implementations of `Ord` and `PartialOrd` on
`cfg(not(fuzzing))`. When fuzzing, auto-derive implementations.
Co-authored-by: Tobin C. Harding <me@tobin.cc>
The method `negate_assign` (on pub/sec key) is cumbersome to use because
a local variable that uses these methods changes meaning but keeps the
same identifier. It would be more useful if we had methods that consumed
`self` and returned a new key.
Add method `negate` that consumes self and returns the negated key.
Deprecated the `negate_assign` methods.
We now have a method `add_tweak` on the `SecretKey` and `PublicKey`. We
can add similar methods that consumes self and return the tweaked key
for the `KeyPair` and `XOnlyPublicKey` types.
The justification for doing so is that a local variable that calls
`tweak_add_assign` changes in meaning but the identifier remains the
same, this leads to cumbersome renaming of the local variable.
The tweaking done to the `KeyPair` is actually done via the xonly public
key not the public key. To reflect this call the method
`add_xonly_tweak`, this is similar to how it is named in secp
`secp256k1_keypair_xonly_tweak_add`.
The key methods `add_assign`, `add_expr_assign`, and `mul_assign` are
cumbersome to use because a local variable that uses these methods
changes meaning but keeps the same identifier. It would be more useful
if we had methods that consumed `self` and returned a new key.
Observe also that these to methods are for adding/multiplying a key by a
tweak, rename the methods appropriately.
Add methods `add_tweak`, `add_expr_tweak`, and `mul_tweak` to the
`SecretKey` and `PublicKey` type. Deprecate `add_assign`,
`add_expr_assign`, and `mul_assign`.
The current display code for `Error` is a little unusual. We typically
just implement `Display` and if a `str` is needed use `format!`.
Improve the `Error` type by doing
- Remove the `as_str` function and implement `Display` directly.
- Remove the 'secp: ' prefix of all the error messages.
- Use a newly defined macro `write_err` that writes the error if `std`
feature is not enabled so that no-std builds do not loose error info.
Note: The `write_err` macro is currently being introduced in
`rust-bitcoin` also. Elect to just duplicate it here and not share it
between the crates.
This adds `Scalar` newtype to better represent values accepted by
tweaking functions. This type is always 32-bytes and guarantees being
within curve order.
We recently added fixed width serialization for some types however
serialization is only fixed width when data is serialized with the
`bincode` crate.
Add rustdocs describing fixed width serde to `SecretKey`, `PublicKey`,
and `XOnlyPublicKey` (`KeyPair` is already done).
elsirion
Andrew Poelstra
Tobin C. Harding
NicolaLS
Martin Habovstiak
Tim Ruffing
Dr Maxim Orlovsky