The wallet is currently created without setting any lookahead value for
the keychain. This implicitly makes it a lookahead of 0. As this is a
high-level interface we should avoid footguns and aim for a reasonable
default.
Instead of simply patching it for wallet, we alter `KeychainTxOutIndex`
to have a default lookahead value. Additionally, instead of a
per-keychain lookahead, the constructor asks for a `lookahead` value.
This avoids the footguns of having methods which allows the caller the
decrease the `lookahead` (and therefore panicing). This also simplifies
the API.
Co-authored-by: Antoine Poisot <darosior@protonmail.com>
Co-authored-by: 志宇 <hello@evanlinjin.me>
f4863c6314 fix(wallet): apply loaded changeset to indexed_graph (thunderbiscuit)
Pull request description:
### Description
This PR applies the tx_graph from the changeset when loading a wallet from persistence. This ensures among other things that the revealed keychain indices get picked up by the new wallet. A test for this has been added/modified from an old test.
### Notes to the reviewers
### Changelog notice
Fix: loading a wallet from persistence now restores keychain indices.
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
#### Bugfixes:
* [ ] This pull request breaks the existing API
* [x] I've added tests to reproduce the issue which are now passing
* [ ] I'm linking the issue being fixed by this PR
ACKs for top commit:
evanlinjin:
ACK f4863c6314
Tree-SHA512: c8502077ba25e6fb953829b020b396774aa0569f7272e7818f30ddabed9d1d8ce791729bebc92b9ec1059028399495cb79ea147cf900f25aace94045dd7290a6
e7fbc8bcf3 ci: run clippy_check with rust stable (Steve Myers)
2251b8d416 ci: pin home version to 0.5.5 for 1.63 MSRV (Steve Myers)
Pull request description:
### Description
Fixed 1.63 MSRV error by pinning `home` dependency to `0.5.5`, and `clippy` error by changing `check_clippy` job to using rust `stable` version.
### Notes to the reviewers
It's OK to use rust `stable` version for clippy because we already have a `clippy.toml` file to tell it which version of the clippy rules to check against.
### Changelog notice
None
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
ACKs for top commit:
evanlinjin:
ACK e7fbc8bcf3
Tree-SHA512: 4e9e12488d44a940ea80ecc32ae74dade13d04d5088cb2206271401a2d77b56407af36482df952354b187a52b83631dcdf53bd60d9084a910f4be278059df93b
0adff9c35f doc: Improve TxGraph & co docs (Daniela Brozzoni)
Pull request description:
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
ACKs for top commit:
notmandatory:
Re-ACK 0adff9c35f
Tree-SHA512: 83adffeddf6c8ddb7b5e57a0fa2e5cffaf75cdeea59be0a6c5e94dcd5a7f98328db8e2a620edc753e60ea3382282908c75d34783a280348cfd105a37982c762b
169385bb5b ci: change MSRV to 1.63.0 (Steve Myers)
Pull request description:
### Description
fixes#331
### Notes to the reviewers
We won't merge this PR until LDK merges lightningdevkit/rust-lightning#2681.
There are alot of clippy checks to fix at 1.63.0 so I left the clippy MSRV at 1.57.0 for now.
### Changelog notice
Changed
- MSRV is now 1.63.0 for bdk, chain, and bitcoind_rpc crates
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
ACKs for top commit:
evanlinjin:
ACK 169385bb5b
Tree-SHA512: ad69038173b4f050b57f637fc06a4153cf76929992cfea77e3f25d1e66be102bd2f83a46401a7e3245e9cc54602210c95b75a578f18c5c8b55d1e7229e92197f
0ecc0280c0 doc(bdk): Clarify the absolute_fee, fee_rate docs (Daniela Brozzoni)
Pull request description:
Fixes#1066
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
ACKs for top commit:
notmandatory:
ACK 0ecc0280c0
Tree-SHA512: 152e48b86c21d4fad711abf76dd8fdc0e8955030331005c1ba6ff0c866c52870161f91ec740838f8238c5ad1c620e06212099308a55d130699cb18e4666e3f3f
6817ca9bcb ci: pin hyper-rustls version to 0.24.0 for 1.57 MSRV (thunderbiscuit)
4ee41dbc40 build(esplora): Add async-https-rustls flag to esplora client (thunderbiscuit)
Pull request description:
### Description
The bdk_esplora crate currently doesn't expose the [`async-https-rustls` flag offered by the rust-esplora-client](ef1925e1ee/Cargo.toml (L44)) crate and instead requires users to build using the `default-tls` flag on reqwest, which uses the platform-specific openssl library when compiling. This creates complications for cross-compilation, notably for our Android builds that currently support 3 architectures (`arm64-v8a`, `armeabi-v7a`, and `x86_64`). In order to solve this we can either compile the openssl libraries for each of the platforms we want to support, or use the rustls-tls version of reqwest. The second options is much easier and requires less fiddling with the internals of the Android native development kit and cross-compilation rabbit holes.
Before we merge this I want to make sure I understand the tradeoffs between the `native-tls` and the `rustls-tls` and confirm that there are not potential issues there, but from what I understand they should provide the same functionality/security, and because these are already available/exposed in reqwest and rust-esplora-client, I think this should be a fairly straightforward additional feature we offer.
### Changelog notice
```txt
Added:
- New async-https-rustls feature flag for the bdk_esplora crate, allowing to compile rust-esplora-client using rustls-tls instead of the default native-tls.
```
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
ACKs for top commit:
notmandatory:
ACK 6817ca9bcb
realeinherjar:
ACK 6817ca9bcb
danielabrozzoni:
ACK 6817ca9bcb
Tree-SHA512: 1d417da7cf85e157d71f56442a06e817e8741822d7bff9089f7fbb70ff8b4854f1f52befbc348b849e9c98cae848b792d426cd5bf551e7a9089b15467d28efdd
43aed386bc esplora: also test the gap limit bounds in the async client (Antoine Poinsot)
cb713e5b8c esplora: also fix the gap limit check in the async client (Antoine Poinsot)
2c4e90a76f esplora: scan gap limit bounds testing (Antoine Poinsot)
18bd329617 esplora: fix incorrect gap limit check in blocking client (Antoine Poinsot)
Pull request description:
The gap limit was checked such as if the last_index was not None but the last_active_index was, we'd consider having reached it. But the last_index is never None for this check. This effectively made it so the gap limit was always 1: if the first address isn't used last_active_index will be None and we'd return immediately.
Fix this by avoiding error-prone Option comparisons and correctly checking we've reached the gap limit before breaking out of the loop.
ACKs for top commit:
danielabrozzoni:
ACK 43aed386bc
evanlinjin:
ACK 43aed386bc
Tree-SHA512: c6a172e0627380add56aec79e7fe36c0358e092b59f0bea8885d3524e65c10336291943efe6aeb5cc83f90c4f2146ed63215e56e08583d703b6ab57284487f03
The gap limit was checked such as if the last_index was not None but the
last_active_index was, we'd consider having reached it. But the
last_index is never None for this check. This effectively made it so the
gap limit was always 1: if the first address isn't used
last_active_index will be None and we'd return immediately.
Fix this by avoiding error-prone Option comparisons and correctly
checking we've reached the gap limit before breaking out of the loop.
278210bb89 docs(bdk): clarify `insert_txout` docs (志宇)
6fb45d8a73 test(bdk): add `test_list_output` (志宇)
e803ee9010 feat(bdk): add `Wallet::list_output` method (志宇)
82632897aa refactor(bdk)!: rename `LocalUtxo` to `LocalOutput` (志宇)
Pull request description:
Fixes#1184
### Description
Introduce `Wallet::list_output` method that lists all outputs (both spent and unspent) in a consistent history.
### Changelog notice
- Rename `LocalUtxo` to `LocalOutput`.
- Add `Wallet::list_output` method.
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
#### New Features:
* [x] I've added tests for the new feature
* [x] I've added docs for the new feature
ACKs for top commit:
notmandatory:
re-ACK 278210bb89
danielabrozzoni:
ACK 278210bb89
Tree-SHA512: 151af0e05e55d9c682271ef0c7a82e189531db963f65aa62c2ba0507f203dde39ab7561521c56e72c26830828e94ff96b7bd7e3f46082b23f79c5e0f89470136
02fa340896 chore: remove bdk dependency on log and dev dependency on env_logger (Steve Myers)
Pull request description:
### Description
As suggested by @TheBlueMatt we shouldn't use the `log` crate because even though it is a rust-lang project is may depend on some other crates that aren't as well maintained. Since we were only use `log` in a few places and not it any other bdk workspace projects except the `bdk` package I removed it.
### Notes to the reviewers
For consistency I also removed the `env_logger` from the dev dependencies since it was only being used in a few places in a couple examples and println! is adequate for examples.
### Changelog notice
None
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
Top commit has no ACKs.
Tree-SHA512: beb9e4d465112090093590c45e3e5e2286a99a819312512eb3e5b40d0eade9740314bb8e45a2ad3fa0a4c86e32711d2ef8f966842815874b9315fd0b63bc7283
00ec19ef2d ci: fix MSRV pinning for rustls 0.21.9 (Steve Myers)
77f9977c02 feat(wallet): Add infallible Wallet get_address(), get_internal_address functions (Steve Myers)
9e7d99e3bf refactor(bdk)!: add context specific error types, remove top level error mod (Steve Myers)
Pull request description:
### Description
To remove some places where there were `.expect("TODO")` I added a new `CreateTxError` type which is returned from `TxBuilder::finish()`. I also updated related tests and doc tests.
Fixes https://github.com/bitcoindevkit/bdk/issues/996#issuecomment-1621036206
Also added fallible `Wallet::try_get_address()` and `try_get_internal_address()` to return `Result` with a possible `D:WriteError` when a PersistBackend is used. This should fix#996.
I removed catch-all bdk::Error and replaced usages with new types and updated related functions, fixes#994.
### Notes to the reviewers
~~I didn't add all possible bdk::Error types that `Wallet::create_tx()` and `TxBuilder::finish()` functions might throw. It's probably not too much more work but will take a bit more research so I want to make sure this is the right general approach first.~~
I added `anyhow` to the dev-dependencies so I could remove some `.expect()` lines from the docs tests and make the examples closer to what an end user should do. I also used the `anyhow!()` macro to replace a few places that were using the `bdk::Error::Generic` in example code.
I also moved the module level error.rs file to wallet/error.rs so no one would be tempted to make any new catch all errors and to make it clear that all the errors in it are wallet module related.
### Changelog notice
Changed
- Updated bdk module to use new context specific error types
- wallet: MiniscriptPsbtError, CreateTxError, BuildFeeBumpError error enums
- coin_selection: module Error enum
- Renamed fallible Wallet address functions to try_get_address() and try_get_internal_address()
Removed
- Removed catch-all top level bdk::Error enum
- Removed impl_error macro
Added
- Added infallible Wallet get_address(), get_internal_address functions
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
#### Bugfixes:
* [x] This pull request breaks the existing API
* [ ] I've added tests to reproduce the issue which are now passing
* [ ] I'm linking the issue being fixed by this PR
Top commit has no ACKs.
Tree-SHA512: a87c0856d71f9c945d12b6de6d368f49bd62d73886ac46ac83d00ddb81f2c38c5233ba053e40c76dea73ee7bfc19dac510eec5d7c9026ae50a2dc7308ac4786f
27a63abd1e chore: typos fixed (Einherjar)
Pull request description:
### Description
Fixes the typos and remove unused speculos dockerfiles that was done in #1165.
Moving these changes into this PR to be merged first.
Then, we can rebase #1165 and make it only related to CI and Nix.
(Maybe do a big squash 😄)
## Note to Reviewers
About the speculos stuff, we are not using them, removed in #793,
more specifically in 3f5a78ae3b.
### Changelog notice
- Fix typos in codebase and docs
- Remove unused CI tests with hardware signer Dockerfiles
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
#### New Features:
* [ ] I've added tests for the new feature
* [ ] I've added docs for the new feature
#### Bugfixes:
* [ ] This pull request breaks the existing API
* [ ] I've added tests to reproduce the issue which are now passing
* [ ] I'm linking the issue being fixed by this PR
ACKs for top commit:
danielabrozzoni:
ACK 27a63abd1e
Tree-SHA512: a01101d0741e2b0e1d1254b5cae670c5a936bb0b89332c102feb57d58d2b9ae995ed4436068b0dc5fae73dbe22431c3143d6e04cdc12eab991bd726cfd2fbe25
f1b112e8f9 docs(bitcoind_rpc): update docs for `Emitter::new` (志宇)
9a250baf62 chore: make clippy happy (志宇)
79b84bed0e feat(bdk): changeset's `Append` impl checks that network is consistent (志宇)
06a956ad20 feat!: change `load_from_persistence` to return an option (志宇)
c3265e2514 test(bdk): add tests for wallet constructor methods (志宇)
96f1d94e2c test(file_store): add construction method tests (志宇)
1886dc4fe7 chore(examples): use `Wallet::new_or_load` method where appropriate (志宇)
24994a3ed4 feat(file_store)!: have separate methods for creating and opening Store (志宇)
d294e2e318 feat(wallet)!: add `new_or_load` methods (志宇)
7c6cbc4d9f chore(file_store): rm empty test file (志宇)
6cf3963c6c feat(bdk)!: have separate methods for creating and loading `Wallet` (志宇)
7d5f31f6cc feat(chain, file_store): add `is_empty` method to `PersistBackend` trait (志宇)
5998a22819 feat!: `LocalChain` with hardwired genesis checkpoint (志宇)
Pull request description:
closes#1079closes#1107
### Description
Many methods of `TxGraph` require a `chain_tip: BlockId` input to use against a `ChainOracle` implementation. This is used to ask the `ChainOracle` implementation whether a certain block exists in the chain identified by the `chain_tip`. This guarantees that the `TxGraph` methods will return a consistent history of transactions.
However, the `ChainOracle` trait's `get_chain_tip` method returns an option of `BlockId`. It becomes unclear what to do when `get_chain_tip` returns `None`.
This PR changes the `ChainOracle::get_chain_tip` method to always return a `BlockId` (no `Option`). `LocalChain` now hardwires the genesis block in order to implement `ChainOracle`.
`bdk::Wallet` and `bdk_file_store::Store` are changed to have separate constructor methods for initializing a fresh instance and recovering a previous instance from persistence.
### Notes to the reviewers
### Changelog notice
- Changed `ChainOracle::get_chain_tip` method to return a `BlockId` instead of an `Option` of a `BlockId`.
- Refactored `LocalChain` so that the genesis `BlockId` is hardwired. This way, the `ChainOracle::get_chain_tip` implementation can always return a tip.
- Add `is_empty` method to `PersistBackend`. This returns true when there is no data in the persistence.
- Changed `Wallet::new` to initialize a fresh wallet only.
- Added `Wallet::load` to restore an instance of a wallet.
- Replaced `Store::new` with separate methods to create/open the database file.
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
#### New Features:
* [x] I've added tests for the new feature
* [x] I've added docs for the new feature
Top commit has no ACKs.
Tree-SHA512: 31b75fb53cc451f1fce7e409f1112c43973db7e8b5b31640e01e5b52089683b60320565427d6ea0478ff4c8680dbdb9272fdab08aef69d30f257da52e731e1a3
`PersistBackend::is_empty` is removed. Instead, `load_from_persistence`
returns an option of the changeset. `None` means persistence is empty.
This is a better API than a separate method. We can now differentiate
between a persisted single changeset and nothing persisted.
`Store::aggregate_changeset` is refactored to return a `Result` instead
of a tuple. A new error type (`AggregateChangesetsError`) is introduced
to include the partially-aggregated changeset in the error. This is a
more idiomatic API.
These methods try to load wallet from persistence and initializes the
wallet instead if non-existant.
An internal helper method `create_signers` is added to reuse code.
Documentation is also improved.
`Wallet::new` now creates a new wallet. `Wallet::load` loads an existing
wallet. The network type is now recoverable from persistence. Error
types have been simplified.
Fixes#1118.
Adds `dependabot.yml` to `.github/` to check for `"github-action"`
updates on a `"weekly"` basis.
This does not touch Rust code or Cargo workflows.
It will create PRs and we would need to approve them
(they would be subject to the same merge policy)
to instantiate the proposed dependabots into `master`.
991cb77b6f fix(chain): filter coinbase tx not in best chain (Wei Chen)
Pull request description:
### Description
Fixes#1144.
Coinbase transactions cannot exist in the mempool and be unconfirmed. `TxGraph::try_get_chain_position` should always return `None` for coinbase transactions not anchored in best chain.
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
#### Bugfixes:
* [ ] This pull request breaks the existing API
* [x] I've added tests to reproduce the issue which are now passing
* [x] I'm linking the issue being fixed by this PR
ACKs for top commit:
notmandatory:
ACK 991cb77b6f
danielabrozzoni:
ACK 991cb77b6f
Tree-SHA512: 9e06d8404708eee050c96807a876a470303f4983666c82c56c17d97c2d4b72784e75271279fd393c53a6a967a352aea1ef2762da71ac4bb58f7a0c2f05354948
0112c67b60 chore: rename `ConfirmationTimeAnchor` to `ConfirmationTimeHeightAnchor` (Wei Chen)
Pull request description:
### Description
Closes#1187.
An `Anchor` implementation that records both height and time should have both attributes included in the name.
### Checklists
#### All Submissions:
* [x] I've signed all my commits
* [x] I followed the [contribution guidelines](https://github.com/bitcoindevkit/bdk/blob/master/CONTRIBUTING.md)
* [x] I ran `cargo fmt` and `cargo clippy` before committing
ACKs for top commit:
notmandatory:
ACK 0112c67b60
Tree-SHA512: 024cbc83c8aca36baeaf2ce36979d62f235ffea7702e7ac8d4e7669cbc1730f7e1469ba78bf3da6c5a14abedbf1a9e832bdd66fdaa154ad2bef29cb187e1c504
