This has been fixed and we use nightly to lint so we have access to the
merged fix.
Removing the attribute uncovers a bunch of real lint warnings, fix
them while we are at it.
During this release cycle we deprecated `to_vec` in favour of
`to_bytes`, we have since reversed our position on the name.
Remove the deprecation of `to_bytes` from the three types that had it
and use `to_vec`.
Up until recently we were using wildcard re-exports for types moved to
`units` and `primitives`. We have decided against doing so in favour of
explicit re-exports.
Audit `units` and `primitives` using `git grep 'pub enum'` (and
`struct`) and explicitly re-export all types.
Remove all wildcards except for the re-exports from `opcodes`, there are
too many opcodes, explicitly re-exporting them does not aid clarity.
bbffa3db43 Remove the IO error from DecodeError (Tobin C. Harding)
713196be0d Return DeserError from encode::deserialize (Tobin C. Harding)
33566ac58c Split encode::Error into two parts (Tobin C. Harding)
b04142c745 Add encode::Error::MissingData variant (Tobin C. Harding)
5a42ef2850 Do not manually map IO error (Tobin C. Harding)
efd7f9f06c Add error constructor parse_failed_error (Tobin C. Harding)
ebfef3f114 Return generic error as Some (Tobin C. Harding)
a6254212dc Move consensus error code to submodule (Tobin C. Harding)
Pull request description:
The `consensus::deserialize` and `consensus::deserilaize_partial` functions should not return an I/O error. Doing so causes various other error types to include an `io::Error` and the `io::Error` is an annoying type to work with.
This PR is a bunch of steps, and it took me a good while with quite a bit of backtracking to get here. As such you may want to review the final state before looking at each patch.
The `consensus` errors can be further cleaned up but I'd prefer not to spend more time on this unless it has some chance of merging.
ACKs for top commit:
apoelstra:
ACK bbffa3db43802b30d23259c0372f16a877a0ef8b; successfully ran local tests
Tree-SHA512: 522fdd29638a214cb7fcee29dd3b9f5c846f041fba087a56a91b83e6d85f033cbed95f659dc4321cd4596943ff233bdd184cdfbfcc787fe89172bb93aa4ab186
The `consensus::encode::Error` contains an IO error but reading from a
buffer only ever errors for EOF. We converted all instances of EOF to
`MissingData` already so now we can split the IO error apart from the
actual encoding errors variants.
The extension traits are temporary just while we try to stabalize
`primitives`, they are not intended to be implemented by downstream.
Seal the extension traits so that downstream crates cannot implement
them.
Fix: #3231
Move the `Witness` over to `primitives` leaving behind any method that
takes or returns a `Script` or a signature.
Includes addition of a feature gate to unit test.
In preparation for moving the `Witness` to `primitives` we need a way to
construct the `Witness` when decoding. In order to maintain the current
performance and not introduce additional allocations we need to be able
to construct a `Witness` from the content buffer, this leaks the
implementation details of `Witness`.
Add a clearly marked unstable constructor to create a `Witness` from
parts. This introduces the concept of `foo__unstable` function names;
add a section to the README describing unstable functions and semver
guarantees.
Recently we deprecated `to_vec` in favour of `to_bytes` however we
continued to use `to_vec` in a few places. This wasn't noticed because
of our usage of `TBD` in the `deprecated` attribute.
Use `to_bytes` instead of `to_vec`.
We would like to move the `Witness` to `primitives` however in the
`Encodable` implementation we are currently accessing the private
`content` field.
Instead of accessing `content` we can iterate over the witness elements
and write each individually, this has the same result but does a bunch
of additional calls to `Write::write_all` (via `emit_slice`).
This patch effects performance negatively but makes no changes to the
encoding.
The `Witness::push_slice` function is called by `Witness::push` after
calling `as_ref`, hence is equivalent for all types that implement
`AsRef<[u8]>`. Also, `push_slice` is a private method on `Witness`.
In preparation for moving `Witness` over to `primitives` stop using
`push_slice` in favour of `push`.
Internal change only.
In preparation for moving the `Witness` oven to `primitives` use the
`len` function instead of accessing the `witness_elements` field.
No logic change, `Witness::len()` returns `witness_elements`.
5fab6b178f Rename iter len unit test (Tobin C. Harding)
f6a74ef4af Refactor the serde Witness unit tests (Tobin C. Harding)
9860453b5b Improve Witness consensus encode unit test (Tobin C. Harding)
7e2899d310 Improve Witness::push unit test (Tobin C. Harding)
fe967279e5 Improve witness unit tests for single empty element (Tobin C. Harding)
Pull request description:
In preparation for moving the `Witness` type over to `primitives` refactor and improve all the unit tests that will be moved, do not touch the ones that will stay behind.
The first five patches are from #3406, the last is just a re-name of the test function I tried to refactor in ac6fe3a881
ACKs for top commit:
apoelstra:
ACK 5fab6b178f successfully ran local tests
Tree-SHA512: bc00f81e3c5cc92ae58dd2fc876d368a487ae6c08cc0735d7227c3a89287e321dbfb5b571b951d0616af0ec7cf9a0ea2d0e724645b1c419933a212ece80a0fbf
Rust convention is to not use `test_` prefix on unit tests. Also this
unit test is testing that the `ExactSizedIterator` trait is implemented
and working.
Re-name unit test to `exact_sized_iterator`.
In preparation for moving unit tests to `primitives` give the serde
tests some love by doing:
- Split them up to do one thing only
- Round trip arbitrary witness
- Use better names
Make an effort to clean up the encoding unit test, by doing:
- Remove element accessor assertions (tested already above)
- Add roundtrip encoding assertion
The current unit test is incorrect, the indices field of a witness with
a single element starts at 1 because 0 is encode as a single
byte (compact encoded integer).
Fix the debug test and add a test that pushes an empty slice.
At some stage we named the compact encoding `VarInt` (which makes sense
because the compact size encoding is a variable length integer encoding).
However it turns out the term "varint" is used in Core for a different
encoding so this may lead to confusion.
While we fix this naming thing observe also that the `VarInt` type is
unnecessarily complicated, all we need to be able to do is encode and
decode integers in compact form as specified by Core. We can do this
simply by extending our `WriteExt` and `ReadExt` traits.
Add `emit_compact_size` and `read_compact_size` to emit and read compact
endcodings respectively.
Includes addition of `internals::compact_size::encoded_size_const`.
Patch originally written by Steven, Tobin cherry-picked and did a bunch
of impovements after the varint vs compact_size thing (#1016).
ref: https://en.bitcoin.it/wiki/Protocol_documentation#Variable_length_integer
Co-developed-by: Tobin C. Harding <me@tobin.cc>
We would like to add a `emit_varint` function, however doing so requires
that we can get access to the length of a slice when we are encoding it
so we can use `emit_slice` to implement `emit_varint`. It would be
easier to do so if `emit_slice` returned the length of the slice.
In preparation for adding `emit_varint` (and removing the `VarInt` type)
return the encoded length of a slice from `WriteExt::emit_slice`.
(Patch originally written by Steven, cherry-pick and patch description
written by Tobin.)
Co-developed-by: Tobin C. Harding <me@tobin.cc>
d65de7c7de Introduce and use new compact_size module (Tobin C. Harding)
Pull request description:
We would like to move the witness module to `primitives` but there is a bunch of usage of `VarInt`.
Introduce a module that does the encoding and decoding instead, note that while the functionality is internal decoding returns an error which may one day end up in the public API. So put the module in `primitives` and make it public.
Adds the module to `primitives`, adds a public `MAX_ENCODABLE_SIZE` variable that is commented with an issue link.
https://github.com/rust-bitcoin/rust-bitcoin/issues/3264
ACKs for top commit:
apoelstra:
ACK d65de7c7de successfully ran local tests
Kixunil:
ACK d65de7c7de
Tree-SHA512: d9483c29b2b324e27460564a23f4639dde4037e6e773f4356216b02ebdea893a6361c342002b8e93a54de47b71ac69369431554f8cd0a2522fc451bf8493c81c
We would like to move the witness module to `primitives` but there is
a bunch of usage of `VarInt`.
Introduce a module that does the encoding and decoding instead, this
code is internal so put it in `internals`.
Note we add an unused public `MAX_ENCODABLE_SIZE` variable that is
commented with an issue link. Done like this because its quite
important that we see to it and it makes it clear that we are not and
we know about it.
https://github.com/rust-bitcoin/rust-bitcoin/issues/3264
The `Witness` struct is non-trivial, in particular it is not immediately
obvious where and when the compact size encode value for each witness
element is stored.
Make an effort to improve the docs on `Witness` in relation to the
compact size encoded length of each witness element.
In preparation to move script types to `primitives` we replace impl
block with extension traits by replacing the temporary modules with
`define_extension_trait`.
ac4db6369d witness: Add Witness::witness_script inspector (Steven Roose)
6cc6c8621a witness: Add Witness::taproot_annex (Steven Roose)
b0848022eb witness: Add Witness::taproot_control_block (Steven Roose)
ef336e1387 witness: Improve Witness::tapscript (Steven Roose)
e48a2e4225 script: Add Script::redeem_script inspector (Steven Roose)
Pull request description:
Bundled these because they are very similar. Got a bunch of larger changes coming up based on these. I've been using these for a while for TXHASH work.
ACKs for top commit:
apoelstra:
ACK ac4db6369d but will need to wait for next release. I think we should merge these as-is although they will be much clearer after we do script tagging.
tcharding:
ACK ac4db6369d
Tree-SHA512: e1590d1bdc8b91aeba137453f0cdaa7e1ae6df3c8e9e1e0f087ed9be1a6beaf2286818379247d26c5dd27d07c12c10433db1c9b9a71667ab4d8d37c7deff1373
c717f7f424 Improve docs on private Witness fields (Tobin C. Harding)
Pull request description:
The `Witness` type is a reasonable complex data structure, make an effort to clarify its structure in the docs on the private fields.
Private docs only.
(Original idea pulled out of #2133.)
ACKs for top commit:
Kixunil:
ACK c717f7f424
apoelstra:
ACK c717f7f424 much clearer, thanks!
Tree-SHA512: 9d54b7eeefec97e584fb5f275049dbac0473c949fae8ab05c6961d6fc424c17a058af7037c2220ef1446af294d78c68bfee741cfeca1b18ecc402935d8069dab
d099b9c195 Remove wildcard from prelude import (Jamil Lambert, PhD)
Pull request description:
This patch replaces `prelude::*` wildcard imports with the types actually used. In a couple of cases `DisplayHex` was previously imported by the wildcard but was only used in the test module, an additional import was added to the test module instead of at the top where it causes an unused import warning.
Close: #2875
ACKs for top commit:
Kixunil:
ACK d099b9c195
tcharding:
ACK d099b9c195
Tree-SHA512: d59dfac0961d2649d509039a11c1b5574d81d05fef567a624cf15be2f587de796ea960ba5a08bef788199331c2f790fb06f7b393182538c7d8b1891ded119efc
The `Witness` type is a reasonable complex data structure, make an
effort to clarify its structure in the docs on the private fields.
Private docs only.
Wildcards have been replaced with what is actually used.
In a couple of cases an additional use statement was added to the test
module to import `DisplayHex` which is only used in test, but
previously imported with the wildcard at the top.
We should pass `Copy` types by value not by reference.
Currently this is not done in secp, but lets do it here in bitcoin.
Pass by value:
- `SerializedSignature`
- bitcoin sigs
- secp sigs
- secp `Message`
the `blockdata` directory is code organisation thing, all the
types/modules are re-exported from other places. In preparation for, and
to make easier, the `primitives` crate smashing work - remove all
explicit usage of `blockdata`.
Note that the few instances remain as they seem required e.g.,
`pub(in crate::blockdata::script)`
Refactor only, no logic changes.
eda61ddfef Deprecate to_vec in favour of to_bytes (Tobin C. Harding)
Pull request description:
Currently we have to method names for the same thing "copy this object into a vector". The library is easier to use if we are uniform and just use one.
Elect to use `to_bytes`, for context see discussion in PR #2585.
ACKs for top commit:
apoelstra:
ACK eda61ddfef Nice. IMO we should start deprecating stuff for two releases rather than one, so that people have a year to update.
Tree-SHA512: 0aadd1258a07bfa53806f19a3c41af8d3b1132aa42e7a2015a59c58c4309d7a9b50b86d076c181ce5870ba5acd989feec32669352ecf857ae6fd982873482c34
Tobin C. Harding
merge-script
Fmt Bot
Jamil Lambert, PhD
Steven Roose
Shing Him Ng
Martin Habovstiak