3bee563c810ace941e0e80ee8e410d843a05f5d8 refactor(wallet)!: remove TxOrdering::Bip69Lexicographic (nymius)
e5cb7b206619f5f3437e5ee95ed3e53885b745e6 feat(wallet): add TxOrdering::Custom (FadedCoder)
Pull request description:
<!-- You can erase any parts of this template not applicable to your Pull Request. -->
### Description
Resolves https://github.com/bitcoindevkit/bdk/issues/534.
Resumes from the work in https://github.com/bitcoindevkit/bdk/pull/556.
Add custom sorting function for inputs and outputs through `TxOrdering::Custom` and deprecates `TxOrdering::Bip69Lexicographic`.
<!-- Describe the purpose of this PR, what's being adding and/or fixed -->
### Notes to the reviewers
I tried consider all discussions in https://github.com/bitcoindevkit/bdk/issues/534 while implementing some changes to the original PR. I created a summary of the considerations I had while implementing this:
##### Why use smart pointers?
The size of enums and structs should be known at compilation time. A struct whose fields implements some kind of trait cannot be specified without using a smart pointer because the size of the implementations of the trait cannot be known beforehand.
##### Why `Arc` or `Rc` instead of `Box`?
The majority of the useful smart pointers that I know (`Arc`, `Box`, `Rc`) for this case implement `Drop` which rules out the implementation of `Copy`, making harder to manipulate a simple enum like `TxOrdering`. `Clone` can be used instead, implemented by `Arc` and `Rc`, but not implemented by `Box`.
##### Why `Arc` instead of `Rc`?
Multi threading I guess.
##### Why using a type alias like `TxVecSort`?
cargo-clippy was accusing a too complex type if using the whole type inlined in the struct inside the enum.
##### Why `Fn` and not `FnMut`?
`FnMut` is not allowed inside `Arc`. I think this is due to the `&mut self` ocupies the first parameter of the `call` method when desugared (https://rustyyato.github.io/rust/syntactic/sugar/2019/01/17/Closures-Magic-Functions.html), which doesn't respects `Arc` limitation of not having mutable references to data stored inside `Arc`:
Quoting the [docs](https://doc.rust-lang.org/std/sync/struct.Arc.html):
> you cannot generally obtain a mutable reference to something inside an `Arc`.
`FnOnce` > `FnMut` > `Fn`, where `>` stands for "is supertrait of", so, `Fn` can be used everywhere `FnMut` is expected.
##### Why not `&'a dyn FnMut`?
It needs to include a lifetime parameter in `TxOrdering`, which will force the addition of a lifetime parameter in `TxParams`, which will require the addition of a lifetime parameter in a lot of places more. **Which one is preferable?**
<!-- In this section you can include notes directed to the reviewers, like explaining why some parts
of the PR were done in a specific way -->
### Changelog notice
- Adds new `TxOrdering` variant: `TxOrdering::Custom`. A structure that stores the ordering functions to sort the inputs and outputs of a transaction.
- Deprecates `TxOrdering::Bip69Lexicographic`.
<!-- Notice the release manager should include in the release tag message changelog -->
<!-- See https://keepachangelog.com/en/1.0.0/ for examples -->
### Checklists
#### All Submissions:
* [ ] 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
* [ ] I've added docs for the new feature
Top commit has no ACKs.
Tree-SHA512: 0d3e3ea9aee3a6c9e9d5e1ae93215be84bd1bd99907a319976517819aeda768a7166860a48a8d24abb30c516e0129decb6a6aebd8f24783ea2230143e6dcd72a
BIP 69 proposed a deterministic way to sort transaction inputs and
outputs with the idea of enhancing privacy. It was later discovered
there was no such enhancement but rather a decrement in privacy due to
this sorting.
To avoid the promotion of bad practices, the
TxOrdering::Bip69Lexicographic variant which implemented this BIP for
BDK is removed with this change.
Notice that you can still produce a BIP 69 compliant transaction
defining order functions for TxOrdering::Custom.
Signed-off-by: Steve Myers <steve@notmandatory.org>
The deterministic sorting of transaction inputs and outputs proposed in
BIP 69 doesn't improve the privacy of transactions as originally
intended. In the search of not spreading bad practices but still provide
flexibility for possible use cases depending on particular order of the
inputs/outpus of a transaction, a new TxOrdering variant has been added
to allow the implementation of these orders through the definition of
comparison functions.
Signed-off-by: Steve Myers <steve@notmandatory.org>
af75817d4bedbb5e148812d1073fd0729a23358b ref(tx_graph): Change last_seen to `HashMap<Txid, u64>` (valued mammal)
6204d2c766f968af6b63c664dda61fa8fc897e85 feat(tx_graph): Add method `txs_with_no_anchor_or_last_seen` (valued mammal)
496601b8b148afd2199a530d40edeafcb7967d46 test(tx_graph): Add test for `list_canonical_txs` (valued mammal)
c4057297a96ccbd49d984b7139994b801c2120dc wallet: delete method `insert_anchor` (valued mammal)
b34790c6b6d612661a4595bf10910294af862323 ref(tx_graph)!: Rename `list_chain_txs` to `list_canonical_txs` (valued mammal)
2ce4bb4dfc4f7b779a815eca45a3f6cee3d6c4e0 test(indexed_tx_graph): Add test_get_chain_position (valued mammal)
36f58870cb6eb24fe8c50ba4cf3ede910dd11fe8 test(wallet): Add test_insert_tx_balance_and_utxos (valued mammal)
bbc19c3536b25c78be8b5f3fe0cd9810aa679742 fix(tx_graph)!: Change tx_last_seen to `Option<u64>` (valued mammal)
324eeb3eb4e5231c6a81e6d197df788ad08b23a8 fix(wallet)!: Rework `Wallet::insert_tx` to no longer insert anchors (valued mammal)
Pull request description:
The PR changes the type of last_seen to `Option<u64>` for `txs` member of `TxGraph`.
This fixes an issue where unbroadcast and otherwise non-canonical transactions were returned from methods `list_chain_txs` and `Wallet::transactions` because every new tx inserted had a last_seen of 0 making it appear unconfirmed.
fixes#1446fixes#1396
### Notes to the reviewers
### Changelog notice
Changed
- Member `last_seen_unconfirmed` of `TxNode` is changed to `Option<u64>`
- Renamed `TxGraph` method `list_chain_txs` to `list_canonical_txs`
- Changed `Wallet::insert_tx` to take a single `tx: Transaction` as parameter
Added
- Add method `txs_with_no_anchor_or_last_seen` for `TxGraph`
- Add method `unbroadcast_transactions` for `Wallet`
### 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
* [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:
Re ACK af75817d4bedbb5e148812d1073fd0729a23358b
Tree-SHA512: e664b3b49e2f547873923f15dffbbc7fa032b6240e5b856b180e9e26123ca141864d10448912dc4a31bbb200c75bef4251a910a4330dac17ee6841b564612d13
Also fixup `test_list_owned_txouts` to check that the right
outputs, utxos, and balance are returned at different local
chain heights.
This fixes an issue where unbroadcast and otherwise non-canonical
transactions were returned from methods `list_chain_txs` and
`Wallet::transactions` because every tx inserted had a last_seen
of 0 making it appear unconfirmed.
Note this commit changes the way `Balance` is represented due to
new logic in `try_get_chain_position` that no longer considers
txs with non-canonical anchors. Before this change, a tx anchored
to a block that is reorged out had a permanent effect on the
pending balance, and now only txs with a last_seen time or an
anchor confirmed in the best chain will return a `ChainPosition`.
8f5b172e59b4d5e199b98777855b0d9b46742f66 test(wallet): verify wallet panics in dev mode if using keychains with duplicate spks (Steve Myers)
46c6f18cc3ff71efda2b8547aa32854f9b88762e refactor(chain): calculate DescriptorId as sha256 hash of spk at index 0 (Steve Myers)
Pull request description:
### Description
Rename `DescriptorId` to `KeychainId` and `descriptor_id()` to `keychain_id()`.
Calculate keychain ids as the hash of the spk derived from its descriptor as index 0.
Added docs to `Wallet` and `KeychainTxOutIndex::insert_descriptor()` explaining that it's the users responsibility not to use wildcard and non-wildcard descriptors that can derive the same script pubkey. I also recommended for `Wallet` that legacy non-wildcard descriptors be added to a temporary `Wallet` and swept into a modern wildcard descriptor.
### Notes to the reviewers
fixes#1483
### Changelog notice
changed
- Renamed DescriptorId to KeychainId, DescriptorExt::descriptor_id() to keychain_id().
### 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
* [ ] 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:
LLFourn:
ACK 8f5b172e59b4d5e199b98777855b0d9b46742f66
oleonardolima:
Concept ACK 8f5b172e59b4d5e199b98777855b0d9b46742f66
Tree-SHA512: 07defa208d9cfcd61bc6e31ded06a287c392c51bcc4949f601ecfac635c3443e7d08c62d92618ed894dc5ef13cdcf784771a6bf8904a5397110bedb1563f52d4
cf7aca84d113e639441350651112e9e701364497 Remove usage of blockdata:: from bitcoin paths (Tobin C. Harding)
Pull request description:
### Description
In `rust-bitcoin` the `blockdata` module is a code organisation thing, it should never have been public. One day those guys would like to remove it, so as not to be a PITA for `bdk` when they do lets remove all usage of `blockdata::` now.
### Changelog notice
Internal change only, no externally visible changes.
### 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: 4c5e54a2ac3f71835c25ce97ad0acd1859e5ae8d45ebfde087e19e9494754e0aa70a47ca5f3d6a3b509f27e28ef9d4a5269573c686250f43834d09378155681c
In `rust-bitcoin` the `blockdata` module is a code organisation thing,
it should never have been public. One day those guys would like to
remove it, so as not to be a PITA for `bdk` when they do lets remove all
usage of `blockdata::` now.
Internal change only, no externally visible changes.
438cd4682d14b5f4c6c9c653c9f05ccaaecb815d refactor(wallet)!: change WeightedUtxo to use Weight type (Leonardo Lima)
Pull request description:
fixes#1466
depends on #1448
<!-- You can erase any parts of this template not applicable to your Pull Request. -->
### Description
This PR is a follow-up on top of #1448, and should be rebased and merged after it, it uses the rust-bitcoin `Weight` type instead of the current `usize` usage.
NOTE: ~~It also adds a new `MiniscriptError` variant, and remove the `.unwrap()` usage.~~ As suggested I'll address this on another issue #1485, trying to reproduce the error first.
<!-- Describe the purpose of this PR, what's being adding and/or fixed -->
### Notes to the reviewers
It should be ready to review after #1448 gets merged.
<!-- In this section you can include notes directed to the reviewers, like explaining why some parts
of the PR were done in a specific way -->
### Changelog notice
- Change `WeightedUtxo` `satisfaction_weight` has been changed to use `Weight` type.
- Change `TxBuilder` methods `add_foreign_utxo` and `add_foreign_utxo_with_sequence` to expect `Weight` as `satisfaction_weight` type.
<!-- Notice the release manager should include in the release tag message changelog -->
<!-- See https://keepachangelog.com/en/1.0.0/ for examples -->
### 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:
storopoli:
Anyways, ACK 438cd4682d14b5f4c6c9c653c9f05ccaaecb815d
notmandatory:
ACK 438cd4682d14b5f4c6c9c653c9f05ccaaecb815d
Tree-SHA512: 1998fe659833da890ce07aa746572ae24d035e636732c1a11b7828ffed48e753adb4108f42d00b7cd05e6f45831a7a9840faa26f94058fc13760497837af002f
55a17293a455435c868f60f0c9f06ba80f2f0e4c ref(signer): Use `Psbt::sighash_ecdsa` for computing sighashes (valued mammal)
f2a2dae84cd04df4301f91745efa137975eeb8e4 refactor(signer): Remove trait ComputeSighash (valued mammal)
Pull request description:
This PR does some cleanup of the `bdk_wallet` signer module most notably by removing the internal trait `ComputeSighash` and replacing old code for computing the sighash (for legacy and segwit context) with a single method [`Psbt::sighash_ecdsa`](https://docs.rs/bitcoin/0.31.2/bitcoin/psbt/struct.Psbt.html#method.sighash_ecdsa). The logic for computing the taproot sighash is unchanged and extracted to a new helper function `compute_tap_sighash`.
- [x] Unimplement `ComputeSighash`
- [x] Try de-duplicating code by using `Psbt::sighash_ecdsa`. see https://github.com/bitcoindevkit/bdk/pull/1023#discussion_r1263140218
- Not done in this PR: Consider removing unused `SignerError` variants
fixes#1038
### Notes to the reviewers
### Changelog notice
### 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: 56af3c9c463513ca3bae5480aa5b90d78de119c3c09c824a7220eb6832d5f403b172afc8168228918ea1adabb4bf8fca858790adfebf84fc334b4fc1cc99d3cd
- Change param `hash` to `&Message` in `sign_psbt_ecdsa`
- Remove unused methods `compute_legacy_sighash`,
and `compute_segwitv0_sighash`.
- Match on `self.ctx` when signing for `SignerWrapper<PrivateKey>`
since we'd be lacking context that should normally occur during
sync with a chain source. The logic for inserting a graph
anchor from a `ConfirmationTime` is moved to the wallet common
test module in order to simulate receiving new txs and
confirming them.
4bddb0de6262fb4014d51baf8c9453eb45a3d0ef feat(wallet): add back TxBuilder finish() and sort_tx() with thread_rng() (Steve Myers)
45c0cae0a461232bf746375083e2005c5df5f913 fix(bdk): remove rand dependency (rustaceanrob)
Pull request description:
### Description
WIP towards removing `rand` fixes#871
The `rand` dependency was imported explicitly, but `rand` is also implicitly used through the `rand-std` feature flag on `bitcoin`.
### Notes to he reviewers
**Updated:**
`rand` was used primarily in two parts of `bdk`. Particularly in signing and in building a transaction.
Signing:
- Used implicitly in [`sign_schnorr`](https://docs.rs/bitcoin/latest/bitcoin/key/struct.Secp256k1.html#method.sign_schnorr), but nowhere else within `signer`.
Transaction ordering:
- Used to shuffle the inputs and outputs of a transaction, the default
- Used in the single random draw __as a fallback__ to branch and bound during coin selection. Branch and bound is the default coin selection option.
See conversation for proposed solutions.
### Changelog notice
- Remove the `rand` dependency from `bdk`
### 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
#### Bugfixes:
* [x] 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:
ValuedMammal:
ACK 4bddb0de6262fb4014d51baf8c9453eb45a3d0ef
notmandatory:
ACK 4bddb0de6262fb4014d51baf8c9453eb45a3d0ef
Tree-SHA512: 662d9bcb1e02f8195d73df16789b8c2aba8ccd7b37ba713ebb0bfd19c66163acbcb6f266b64f88347cbb1f96b88c8a150581012cbf818d1dc8b4437b3e53fc62
996605f2bf9440dd42647123a127c038253f0247 fix(wallet)!: Simplify `SignOptions` and improve finalization logic (valued mammal)
Pull request description:
Rather than comingle various `SignOptions` with the finalization step, we simply clear all fields when finalizing as per the PSBT spec in BIPs 174 and 371 which is more in line with user expectations.
### Notes to the reviewers
I chose to re-implement some parts of [`finalize_input`](https://docs.rs/miniscript/latest/src/miniscript/psbt/finalizer.rs.html#434) since it's fairly straightforward, see also https://github.com/bitcoindevkit/bdk/issues/1461#issuecomment-2171983426. I had to refactor some wallet tests but didn't go out of my way to add additional tests.
closes#1461
### Changelog notice
- Removed field `remove_partial_sigs` from `SignOptions`
- Removed field `remove_taproot_extras` from `SignOptions`
### 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
* [x] I'm linking the issue being fixed by this PR
ACKs for top commit:
evanlinjin:
re-ACK 996605f2bf9440dd42647123a127c038253f0247
Tree-SHA512: 63e78e75c22031424e87fcc26cd6b0015c626cd57c02680256bad9d1783cef71f4048b2d7ce5d0425cd4239351e37dd0e2a626dda7e8417af7fc52cb0afe6933
Rather than comingle various `SignOptions` with the finalization
step, we simply clear all fields when finalizing as per the PSBT
spec in BIPs 174 and 371 which is more in line with user
expectations.
e21affdbbb9308bd6be38218ae5a26d9c800efe2 Bump bdk version to 1.0.0-alpha.13 (Steve Myers)
Pull request description:
### Description
Bump bdk version to 1.0.0-alpha.13
bdk_chain to 0.16.0
bdk_bitcoind_rpc to 0.12.0
bdk_electrum to 0.15.0
bdk_esplora to 0.15.0
bdk_file_store to 0.13.0
bdk_sqlite keep at 0.2.0
bdk_testenv to 0.6.0
bdk_hwi to 0.3.0
fixes#1471
Top commit has no ACKs.
Tree-SHA512: 987adf5084dc84261fb3767d8b1f8ebe3dd94a8a4eb30b8529b70c055e4f122cb48063d205e90831169c2d7b2a1509aef1966857bd6b67e78cfb0d52144822d9
bdk_chain to 0.16.0
bdk_bitcoind_rpc to 0.12.0
bdk_electrum to 0.15.0
bdk_esplora to 0.15.0
bdk_file_store to 0.13.0
bdk_sqlite keep at 0.2.0
bdk_testenv to 0.6.0
bdk_hwi to 0.3.0
a0bf45bef1b53f8aec273f7fe89915f3d41974c0 docs: remove PersistBackend from docs, Cargo.toml and wallet README.md (Steve Myers)
feb27df180938759670afd2c9cd15529a22c4772 feat(chain)!: add `take` convenience method to `Append` trait (志宇)
1eca568be5bb0c234fa1cea2eb7f9424f60974eb feat!: rm `persist` submodule (志宇)
Pull request description:
### Description
@LLFourn suggested these changes which greatly simplifies the amount of code we have to maintain and removes the `async-trait` dependency. We remove `PersistBackend`, `PersistBackendAsync`, `StageExt` and `StageExtAsync` completely. Instead, we introduce `Wallet::take_staged(&mut self) -> Option<ChangeSet>`.
The original intention to keep a staging area (`StageExt`, `StageExtAsync`) is to enforce:
1. The caller will not persist an empty changeset.
2. The caller does not take the staged changeset if the database (`PersistBackend`) fails.
We achieve `1.` by returning `None` if the staged changeset is empty.
`2.` is not too important. The caller can try figure out what to do with the changeset if persisting to db fails.
### Notes to the reviewers
I added a `take` convenience method to the `Append` trait. I thought it would be handy for the caller to do a staging area with this.
### Changelog notice
* Remove `persist` submodule from `bdk_chain`.
* Change `Wallet` to outsource it's persistence logic by introducing `Wallet::take_staged`.
* Add `take` convenience method to `Append` trait.
### 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~
* [x] I've added docs for the new feature
ACKs for top commit:
notmandatory:
ACK a0bf45bef1b53f8aec273f7fe89915f3d41974c0
evanlinjin:
self-ACK a0bf45bef1b53f8aec273f7fe89915f3d41974c0
Tree-SHA512: 38939ab446c84d9baecd4cd36a7b66add5a121212ad5e06ade04a60f7903133dd9a20219e230ab8a40404c47e07b946ccd43085572d71c3a2a80240a2223a500
Remove `PersistBackend`, `PersistBackendAsync`, `StageExt` and
`StageExtAsync`. Remove `async` feature flag and dependency. Update
examples and wallet.
20341a3ca1ab85ad77e022b5028136a635c3f42c fix: typo on `SignedAmount` instead of `Amount` (Leonardo Lima)
Pull request description:
<!-- You can erase any parts of this template not applicable to your Pull Request. -->
### Description
It fixes the typo on the `expect()` message introduced on #1426, noticed at https://github.com/bitcoindevkit/bdk/pull/1426#discussion_r1626761825
<!-- Describe the purpose of this PR, what's being adding and/or fixed -->
### 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:
storopoli:
ACK 20341a3ca1ab85ad77e022b5028136a635c3f42c
notmandatory:
ACK 20341a3ca1ab85ad77e022b5028136a635c3f42c
Tree-SHA512: 62deb7308b2a6891eeb4598ebdf3c5ee1541fc52fd454bca2816b819bd7f6ab25c18e10c4166c80c4e553bcb3ce2207323376d260484a956837ac857854cbd6b
ec36c7eccacc8347568b77ed1dfc0c50a2524906 feat(example): use changeset staging with rpc polling example (志宇)
19328d4999c19557778b7b108f88f42b1258957e feat(wallet)!: change persist API to use `StageExt` and `StageExtAsync` (志宇)
2e40b0118cc88539f38420e347eb4d562b1be0b1 feat(chain): reintroduce a way to stage changesets before persisting (志宇)
36e82ec6869c5c5c5669d4d049636225f9a8e986 chore(chain): relax `miniscript` feature flag scope (志宇)
9e97ac03307f3194e043ef430e4bde2e5c25254c refactor(persist): update file_store, sqlite, wallet to use bdk_chain::persist (Steve Myers)
54b0c11cbe1d08eb955e50f0ac719a0b19e3032a feat(persist): add PersistAsync trait and StagedPersistAsync struct (Steve Myers)
aa640ab2770bea19259eedea6882b202a5845f35 refactor(persist): rename PersistBackend to Persist, move to chain crate (Steve Myers)
Pull request description:
### Description
Sorry to submit another refactor PR for the persist related stuff, but I think it's worth revisiting. My primary motivations are:
1. remove `db` from `Wallet` so users have the ability to use `async` storage crates, for example using `sqlx`. I updated docs and examples to let users know they are responsible for persisting changes.
2. remove the `anyhow` dependency everywhere (except as a dev test dependency). It really doesn't belong in a lib and by removing persistence from `Wallet` it isn't needed.
3. remove the `bdk_persist` crate and revert back to the original design with generic error types. I kept the `Debug` and `Display` constrains on persist errors so they could still be used with the `anyhow!` macro.
### Notes to the reviewers
I also replaced/renamed old `Persist` with `StagedPersist` struct inspired by #1453, it is only used in examples. The `Wallet` handles it's own staging.
### Changelog notice
Changed
- Removed `db` from `Wallet`, users are now responsible for persisting changes, see docs and examples.
- Removed the `bdk_persist` crate and moved logic back to `bdk_chain::persist` module.
- Renamed `PersistBackend` trait to `Persist`
- Replaced `Persist` struct with `StagedPersist`
### 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:
ValuedMammal:
ACK ec36c7eccacc8347568b77ed1dfc0c50a2524906
Tree-SHA512: 380b94ae42411ea174720b7c185493c640425f551742f25576a6259a51c1037b441a238c6043f4fdfbf1490aa15f948a34139f1850d0c18d285110f6a9f36018
Still enable the `persist` submodule without `miniscript` feature flag.
Only disable `CombinedChangeSet`.
Also stop `cargo clippy` from complaining about unused imports when
`miniscript` is disabled.
8dd174479f9719309663ed979a5b4b86aca0a6e9 refactor(chain): compute txid once for `KeychainTxOutIndex::index_tx` (志宇)
639d735ca0ae54d8b2c3bc28241032154b94d45e refactor(chain): change field names to be more sane (志宇)
5a02f40122f1bfa06c80bac93f68f5799225d133 docs(chain): fix docs (志宇)
c77e12bae7f465ec7fb08b8be16d99793c757cf0 refactor(chain): `KeychainTxOutIndex` use `HashMap` for fields (志宇)
4d3846abf4f59b4a97bb825281655a6b67275603 chore(chain): s/replenish_lookahead/replenish_inner_index/ (LLFourn)
8779afdb0bf4e9b1004f47f86a770d25938d206d chore(chain): document insert_descriptor invariants better (LLFourn)
69f2a695f7dc25478489080598fea0813ea7d93d refactor(chain): improve replenish lookeahd internals (LLFourn)
5a584d0fd8c138757a10c7af93ec9e09523317e1 chore(chain): Fix Indexed and KeychainIndexed documentaion (Lloyd Fournier)
b8ba5a02066fad7ab2ce276ba071385cd1dbbe3a chore(chain): Improve documentation of keychain::ChangeSet (LLFourn)
101a09a97fa5e8d675c13396b9a800665b1b6c22 chore(chain): Standardise KeychainTxOutIndex return types (LLFourn)
bce070b1d662db7ac120e1d236fdda51842ad738 chore(chain): add type IndexSpk, fix clippy type complexity warning (Steve Myers)
4d2442c37f5c1bd822795271a79676d1ffbe7916 chore(chain): misc docs and insert_descriptor fixes (LLFourn)
bc2a8be97919f0d09b61438527bda24796bcec94 refactor(keychain): Fix KeychainTxOutIndex range queries (LLFourn)
3b2ff0cc953204c9925ace8e2f0bbef409c63ad5 Write failing test for keychain range querying (LLFourn)
Pull request description:
Fixes#1459
This reverts part of the changes in #1203. There the `SpkTxOutIndex<(K,u32)>` was changed to `SpkTxOutIndex<(DescriptorId, u32>)`. This led to a complicated translation logic in `KeychainTxOutIndex` (where the API is based on `K`) to transform calls to it to calls to the underlying `SpkTxOutIndex` (which now indexes by `DescriptorId`). The translation layer was broken when it came to translating range queries from the `KeychainTxOutIndex`. My solution was just to revert this part of the change and remove the need for a translation layer (almost) altogether. A thin translation layer remains to ensure that un-revealed spks are filtered out before being returned from the `KeychainTxOutIndex` methods.
I feel like this PR could be extended to include a bunch of ergonomics improvements that are easier to implement now. But I think that's the point of https://github.com/bitcoindevkit/bdk/pull/1451 so I held off and should probably go and scope creep that one instead.
### 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
* [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:
evanlinjin:
ACK 8dd174479f9719309663ed979a5b4b86aca0a6e9
Tree-SHA512: 283e6b6d4218902298e2e848fe847a6c85e27af4eee3e4337e3dad6eacf9beaa08ac99b1dce7b6fb199ca53931e543ea365728a81c41567a2e510cce77b12ac0
