- changes the code implementation to "the maximum number of consecutive unused addresses"
in esplora async and blocking extjensions.
- treat `stop_gap = 0` as `stop_gap = 1` for all purposes.
- renames `past_gap_limit` to `gap_limit_reached` to indicate we want to break once the gap
limit is reached and not go further in `full_scan`, as suggested in
https://github.com/bitcoindevkit/bdk/issues/1227#issuecomment-1859040463
- change the tests according to the new implementation.
- add notes on what `stop_gap` means and links to convergent definition in other
Bitcoin-related software.
Closes#1227
7c9ba3cfc8 chore(electrum,testenv): move `test_reorg_is_detected_in_electrsd` (志宇)
2462e90415 chore(esplora): rm custom WASM target spec test dependency (志宇)
04d0ab5a97 test(electrum): added scan and reorg tests Added scan and reorg tests to check electrum functionality using `TestEnv`. (Wei Chen)
4edf533b67 chore: extract `TestEnv` into separate crate `TestEnv` is extracted into its own crate to serve as a framework for testing other block explorer APIs. (Wei Chen)
6e648fd5af chore: update MSRV dependency for nightly docs (Wei Chen)
Pull request description:
### Description
`TestEnv` is extracted into its own crate with `electrsd` support added so that `TestEnv` can also serve `esplora` and `electrum`.
The tests in the `esplora` crate have also been updated to use `TestEnv`.
The `tx_can_become_unconfirmed_after_reorg()` test in `test_electrum` suggests that the electrum issue where a reorged tx would be stuck at confirmed does not exist anymore.
### Notes to the reviewers
The code for `tx_can_become_unconfirmed_after_reorg()` was adapted from the same test in `bitcoind_rpc/test_emitter`. This electrum version of the test requires extra review, as I am uncertain if I used the API efficiently.
#### 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 7c9ba3cfc8
Tree-SHA512: 36c6501e477abb7ae68073b6f4776f1f69f8964010ab758aa3677aae96df10f2cb632421872cacd73b37123d6db8a9dbefb5b6416e0dd524b712bf3fc56b7139
06d7dc5c3a doc(bdk): Update bdk README (vmammal)
Pull request description:
fixes#1044
### Notes
The code snippet is a compile fail because variables `txout` and `outpoint` are not initialized. Any other suggestions are welcome.
There is still some old commented code in the file that's probably not needed unless we want to provide another example snippet, say for getting new address info.
* [x] Fix broken links
* [x] Use a new unused tprv
* [ ] Try to make persistence example make sense, or maybe just do away with it
#### 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 06d7dc5c3a
Tree-SHA512: 09a671bc6bea574d7a4b42b64718380ee71a0c5d36c6b7ca1dc19a2c567de510b27ccc91fe05e7178bf31c562db66bc64f660415de5bb2f32f369b13c44ad3d2
5840ce473e fix(bdk): Remove extra taproot fields when finalizing Psbt (vmammal)
8c78a42163 test(psbt): Fixup test_psbt_multiple_internalkey_signers (vmammal)
Pull request description:
We currently allow removing `partial_sigs` from a finalized PSBT, which is relevant to non-taproot inputs, however taproot related PSBT fields were left in place despite the recommendation of BIP371 to remove them once the `final_script_witness` is constructed. This can cause confusion for parsers that encounter extra taproot metadata in an already satisfied input.
Fix this by introducing a new member to SignOptions `remove_taproot_extras`, which when true will remove extra taproot related data from a PSBT upon successful finalization. This change makes removal of all taproot extras the default but configurable.
fixes#1243
### Notes to the reviewers
If there's a better or more descriptive name for `remove_taproot_extras`, I'm open to changing it.
### Changelog notice
Fixed an [issue](https://github.com/bitcoindevkit/bdk/issues/1243) finalizing taproot inputs following BIP371
### 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:
evanlinjin:
ACK 5840ce473e
Tree-SHA512: 69f022c6f736500590e36880dd2c845321464f89af6a2c67987d1c520f70a298328363cade0f55f270c4e7169e740bd4ada752ec2c75afa02b6a5a851f9030c3
b290b29502 test(chain): change test case comments to docstring (志宇)
c151d8fd23 fix(chain): `KeychainTxOutIndex::lookahead_to_target` (志宇)
Pull request description:
### Description
This method was not used (so it was untested) and it was not working. This fixes it.
The old implementation used `.next_store_index` which returned the keychain's last index stored in `.inner` (which include lookahead spks). This is WRONG because `.replenish_lookahead` needs the difference from last revealed.
### Changelog notice
Fix `KeychainTxOutIndex::lookahead_to_target`
### 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] I've added tests to reproduce the issue which are now passing
ACKs for top commit:
notmandatory:
ACK b290b29502
Tree-SHA512: af50c6af18b6b57494cfa37f89b0236674fa331091d791e858f67b7d0b3a1e4e11e7422029bd6a2dc1c795914cdf6d592a14b42a62ca7c7c475ba6ed37182539
We currently allow removing `partial_sigs` from a finalized Psbt,
which is relevant to non-taproot inputs, however taproot related Psbt
fields were left in place despite the recommendation of BIP371 to remove
them once the `final_script_witness` is constructed. This can cause
confusion for parsers that encounter extra taproot metadata in an
already satisfied input.
Fix this by introducing a new member to SignOptions
`remove_taproot_extras`, which when true will remove extra taproot
related data from a Psbt upon successful finalization. This change
makes removal of all taproot extras the default but configurable.
test(wallet): Add test
`test_taproot_remove_tapfields_after_finalize_sign_option`
that checks various fields have been cleared for taproot
Psbt `Input`s and `Output`s according to BIP371.
to verify the signature of the input and ensure the right internal
key is used to sign. This fixes a shortcoming of a previous
version of the test that relied on the result of `finalize_psbt`
but would have erroneously allowed signing with the wrong key.
5489f905a4 feat(chain): add `map_anchors` for `TxGraph` and `ChangeSet` (Antonio Yang)
022d5a21cf test(chain) use `Anchor` generic on `init_graph` (Antonio Yang)
Pull request description:
### Description
Fix#1295
### 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:
evanlinjin:
ACK 5489f905a4
LLFourn:
ACK 5489f905a4
Tree-SHA512: c8327f2e7035a46208eb32c6da1f9f0bc3e8625168450c5b0b39f695268e42b0b9053b6eb97805b116328195d77af7ca9edb1f12206c50513fbe295dded542e7
13ab5a835d chore(chain): Improve TxGraph::ChangeSet docs (LLFourn)
dbbd514242 fix(chain)!: rm duplicate `is_empty` method in tx graph changeset (志宇)
ae00e1ee7b fix(chain): tx_graph::ChangeSet::is_empty (LLFourn)
Pull request description:
🙈
### Changelog notice
- Fix bug in `tx_graph::ChangeSet::is_empty` where is returns true even when it wasn't empty
### 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
ACKs for top commit:
LLFourn:
Self-ACK: 13ab5a835d
evanlinjin:
ACK 13ab5a835d
Tree-SHA512: b9f1f17fd2ed0f8e2337a8033e1cbd3e9f15b1ad4b32da3f0eb73a30913d6798e7a08d6b297d93bd08c2e1c388226e97648650ac636846b2c7aa95c3bcefbcfd
552f11cb5f feat(esplora): include previous `TxOut`s for fee calculation The previous `TxOut` for transactions received from an external wallet are added as floating `TxOut`s to `TxGraph` to allow for fee calculation. (Wei Chen)
Pull request description:
### Description
Partially implements #1265.
The previous `TxOut` for transactions received from an external wallet are added as floating `TxOut`s to `TxGraph` to allow for fee calculation.
### Notes to the reviewers
Currently only the `esplora` portion of #1265 has been implemented.
The `electrum` portion will potentially be done in a new PR, as discussed on the 1/30/24 Lib call.
### Checklists
#### To Do:
* [ ] Implement `electrum` portion of #1265.
#### 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:
evanlinjin:
re-ACK 552f11cb5f
danielabrozzoni:
ACK 552f11cb5f
Tree-SHA512: 752a24ebd0b9ad7952c1b093ecb251473e346c77b860c1a80c73418130189227405a0f0d7652967cf8c7b89994e8c37df96cd52b52b6daff9cc8c88b5194069a
9bb39a3a3f Avoid a wildcard match in tx construction (Steven Roose)
9e098a5b6d tx_builder: Support setting explicit nSequence for foreign inputs (Steven Roose)
Pull request description:
Fixes https://github.com/bitcoindevkit/bdk/issues/1315.
ACKs for top commit:
evanlinjin:
ACK 9bb39a3a3f
danielabrozzoni:
utACK 9bb39a3a3f
Tree-SHA512: 42c96a58a762fa8737402ebd0132ce20ce0359c996cee9feeecb0b84e6fb73305be1aec9429fb97ba932486f0b35ac5caed7b43656456c5fb053e55330a12d47
1c15cb2f91 ref(example_cli): Add new struct Init (vmammal)
89a7ddca7f ref(esplora): `Box` a large `esplora_client::Error` (vmammal)
097d818d4c ref(wallet): `Wallet::preselect_utxos` now accepts a `&TxParams` (vmammal)
f11d663b7e ref(psbt): refactor body of `get_utxo_for` to address `clippy::manual_map` (vmammal)
4679ca1df7 ref(example_cli): add typedefs to reduce type complexity (vmammal)
64a90192d9 refactor: remove old clippy allow attributes (vmammal)
Pull request description:
closes#1127
There are several instances in the code where we allow clippy lints that would otherwise be flagged during regular checks. It would be preferable to minimize the number of "clippy allow" attributes either by fixing the affected areas or setting a specific configuration in `clippy.toml`. In cases where we have to allow a particular lint, it should be documented why the lint doesn't apply.
For context see https://github.com/bitcoindevkit/bdk/issues/1127#issuecomment-1784256647 as well as the commit message details.
One area I'm unsure of is whether `Box`ing a large error in 4fc2216 is the right approach. Logically it makes sense to avoid allocating a needlessly heavy `Result`, but I haven't studied the implications or tradeoffs of such a change.
### 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 1c15cb2f91
Tree-SHA512: 5fa3796a33678651414e7aad7ef8309b4cbe2a9ab00dce094964b40784edb2f46a44067785d95ea26f4cd88d593420485be94c9b09ac589f632453fbd8c94d85
These lints either resolved themselves, or the code has changed such that
they no longer apply, hence they can be removed with no further changes.
`clippy::derivable_impls`
`clippy::needless_collect`
`clippy::almost_swapped`
c5afbaa95d ci: update zstd-sys version to work with MSRV 1.63 (Steve Myers)
929b5ddb0c refactor(esplora): better variable naming and docs (志宇)
216648bcfd chore(esplora): Clarify consistency guarantees (LLFourn)
63fa710319 fix(esplora): reuse returned height instead of zipping (志宇)
6f824cf325 test(esplora): introduce test cases for `update_local_chain` (志宇)
f05e8502e6 feat(esplora): greatly simplify `update_local_chain` (志宇)
Pull request description:
Fixes#1199
### Description
After a second look at the `update_local_chain` implementations, it was clear that they were over complicated. This PR simplifies the `EsploraExt::update_local_chain` method(s) of the `bdk_esplora` crate and adds a whole bunch of tests.
### Notes to the reviewers
The description of #1199 is very brief, however @danielabrozzoni mentioned about potentially-problematic logic with `ASSUME_FINAL_DEPTH`. The logic was indeed convoluted and it did not need to be that way. This PR removes the need for `ASSUME_FINAL_DEPTH`.
### Changelog notice
Fixed
- Simplified `EsploraExt::update_local_chain` logic.
### 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~ (there are now lots of tests though)
* [x] I'm linking the issue being fixed by this PR
ACKs for top commit:
evanlinjin:
ACK c5afbaa95d
notmandatory:
utACK c5afbaa95d
Tree-SHA512: f588493a4643f0f68d3181f27adf91793d4e336be6e853a26289e0916ed83169e1067260b75e627f190842f691ec095e668c0b799ca80e7da2849dd28de32754
5299db34cb fix(wallet): filter duplicates before coin selection (志宇)
d9501187ef test(wallet): fix tests helpers to generate unique utxos (志宇)
Pull request description:
Fixes#1240
### Description
We now filter out duplicate coins before calling `CoinSelectionAlgorithm::coin_select`. If a UTXO exists in both `required_utxos` and `optional_utxos`, only the copy in `required_utxos` will be kept.
Test helper methods are also updated to not create duplicate UTXOs.
### Changelog notice
Fixed
* Filter out duplicate UTXOs before calling `CoinSelectionAlgorithm::coin_select`. If a UTXO exists in both `required_utxos` and `optional_utxos`, only the copy in `required_utxos` will be kept.
### 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:
danielabrozzoni:
utACK 5299db34cb
Tree-SHA512: 9cb5517b7f74f89c06172efc344766b16b3216a25b1ebdd6eb84767a9e103124cead9eb0a7f3b5feb562fbb925517a9bf0404399de74b4e898982a5b3795aa04
