These are as suggested by @danielabrozzoni and @afilini
Also introduced `RpcSyncParams::force_start_time` for users who
prioritise reliability above all else.
Also improved logging.
Before this commit, the rpc backend would not notice immature utxos
(`listunspent` does not return them), making the rpc balance different
to other blockchain implementations.
Co-authored-by: Daniela Brozzoni <danielabrozzoni@protonmail.com>
The new implementation fixes the following:
* We can track more than 100 scriptPubKeys
* We can obtain more than 1000 transactions per sync
* `TransactionDetails` for already-synced transactions are updated when
new scriptPubKeys are introduced (fixing the missing balance/coins
issue of supposedly tracked scriptPubKeys)
`RpcConfig` changes:
* Introduce `RpcSyncParams`.
* Remove `RpcConfig::skip_blocks` (this is replaced by
`RpcSyncParams::start_time`).
9d85c9667f7d12902afef3ba08ea7231f6868a78 Fix the early InsufficientFunds error in the branch and bound (Alekos Filini)
Pull request description:
### Description
We were wrongly considering the sum of "effective value" (i.e. value -
fee cost) when reporting an early "insufficient funds" error in the
branch and bound coin selection.
This commit fixes essentially two issues:
- Very high fee rates could cause a panic during the i64 -> u64
conversion because we assumed the sum of effective values would never
be negative
- Since we were comparing the sum of effective values of *all* the UTXOs
(even the optional UTXOs with negative effective value) with the target
we'd like to reach, we could in some cases error and tell the user we
don't have enough funds, while in fact we do! Since we are not required
to spend any of the optional UTXOs, so we could just ignore the ones
that *cost us* money to spend and excluding them could potentially
allow us to reach the target.
There's a third issue that was present before and remains even with this
fix: when we report the "available" funds in the error, we are ignoring
UTXOs with negative effective value, so it may look like there are less
funds in the wallet than there actually are.
I don't know how to convey the right message the user: if we actually
consider them we just make the "needed" value larger and larger (which
may be confusing, because if the user asks BDK to send 10k satoshis, why
do we say that we actually need 100k?), while if we don't we could report
an incorrect "available" value.
### Notes to the reviewers
I'm opening this as a draft before adding tests because I want to gather some feedback on the available vs needed error reporting. I personally think reporting a reasonable "needed" value is more important than the "available", because in a wallet app I would expect this is the value that would be shown to the user.
### 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
* [ ] I'm linking the issue being fixed by this PR
ACKs for top commit:
danielabrozzoni:
utACK 9d85c9667f7d12902afef3ba08ea7231f6868a78
Tree-SHA512: 9a06758cba61ade73198f35b08070987d5eb065e01750ce62409f86b37cd0b0894640e9f75c8b2c26543c0da04e3f77bd397fab540e789f221661aae828db224
We were wrongly considering the sum of "effective value" (i.e. value -
fee cost) when reporting an early "insufficient funds" error in the
branch and bound coin selection.
This commit fixes essentially two issues:
- Very high fee rates could cause a panic during the i64 -> u64
conversion because we assumed the sum of effective values would never
be negative
- Since we were comparing the sum of effective values of *all* the UTXOs
(even the optional UTXOs with negative effective value) with the target
we'd like to reach, we could in some cases error and tell the user we
don't have enough funds, while in fact we do! Since we are not required
to spend any of the optional UTXOs, so we could just ignore the ones
that *cost us* money to spend and excluding them could potentially
allow us to reach the target.
There's a third issue that was present before and remains even with this
fix: when we report the "available" funds in the error, we are ignoring
UTXOs with negative effective value, so it may look like there are less
funds in the wallet than there actually are.
I don't know how to convey the right message the user: if we actually
consider them we just make the "needed" value larger and larger (which
may be confusing, because if the user asks BDK to send 10k satoshis, why
do we say that we actually need 100k?), while if we don't we could report
an incorrect "available" value.
7fdacdbad40f4e9f6726b064d8eb4d93789e9990 doc: Document that list_transactions() might return unsorted txs, show how to sort them if needed (w0xlt)
Pull request description:
This PR documents that `list_transactions()` might return unsorted transaction and shows how to sort them if needed.
Closes#518.
ACKs for top commit:
danielabrozzoni:
re-ACK 7fdacdbad40f4e9f6726b064d8eb4d93789e9990
Tree-SHA512: 83bec98e1903d6dc6b8933e8994cb9d04aad059cee8a7b8e1e3a322cf52511364b36d0cd6be1c8cb1fd82c67f8be5a262bbd2c76e30b24eb4097c30f38aa8b10
Before this commit `fee_amount` and `amount_needed` were passed as independent
parameters. From the perspective of coin selection algorithms, they are always
used jointly for the same purpose, to create a coin selection with a total
effective value greater than it's summed values.
This commit removes the abstraction that the use of the two parameter
introduced by consolidating both into a single parameter, `target_amount`, who
carries their values added up.
419dc248b667db05295cd4c68347c4ef51f51023 test: Document `test_bump_fee_add_input_change_dust` (Daniela Brozzoni)
632dabaa07ef9c58926facf0af5190f62bb65d12 test: Check tx feerate with longer signatures (Daniela Brozzoni)
2756411ef7cf0415baf2f2401e2d5a78481d0aa1 test: Reproduce #660 conditions (Daniela Brozzoni)
50af51da5a5c906d8bf660d35a4f922ceb996068 test: Fix P2WPKH_FAKE_WITNESS_SIZE (Daniela Brozzoni)
ae919061e2b341ae74c90f0133ba392e835cb4e1 Take into account the segwit tx header when... ...selecting coins (Daniela Brozzoni)
7ac87b8f99fc0897753ce57d48ea24adf495a633 TXIN_BASE_WEIGHT shouldn't include the script len (Daniela Brozzoni)
ac051d7ae9512883e11a89ab002ad2d2c3c55c19 Calculate fee amount after output addition (Daniela Brozzoni)
00d426b88546a346820c102386cd1bfff82ca8f6 test: Check that the feerate is never below... ...the requested one in assert_fee_rate (Daniela Brozzoni)
42fde6d4575b4aea121286f884f712b1c1cf64be test: Check fee_amount in assert_fee_rate (Daniela Brozzoni)
Pull request description:
### Description
This PR mainly fixes two bugs:
1. TXIN_BASE_WEIGHT wrongly included the `script_len` (Fixes#160)
2. We wouldn't take into account the segwit header in the fee calculation, which could have resulted in a transaction with a lower feerate than the requested one
3. In tests we used to push 108 bytes on the witness as a fake signature, but we should have pushed 106 instead
I also add a test to reproduce the conditions of #660, to check if it's solved. Turns out it's been solved already in #630, but if you're curious about what the bug was, here it is: https://github.com/bitcoindevkit/bdk/issues/660#issuecomment-1196436776
### 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:
afilini:
ACK 419dc248b667db05295cd4c68347c4ef51f51023
Tree-SHA512: c7b55342eac440a3607a16b94560cb9c08c4805c853432adfda8e21c5177f85d5a8afe0e7e61140e92c8f10934332459c6234fc5f1509ea699d97b1d04f030c6
a713a5a0629c9a643708a4b0d8d6ac3a87a13195 Better customize signing in taproot transactions (Daniela Brozzoni)
Pull request description:
Fixes#616
### 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
* [x] I've updated `CHANGELOG.md`
ACKs for top commit:
afilini:
ACK a713a5a0629c9a643708a4b0d8d6ac3a87a13195
Tree-SHA512: 1100d43cb394b429450fc34f49dd815a024701987c0e6dd163865bd5c4c6f7102127b1ea6e10ced5fdb319874be97baeeb0deea66b4138410871a1d68b4def10
We would previously always try to sign with the taproot internal
key, and try to sign all the script leaves hashes.
Instead, add the `sign_with_tap_internal_key` and `TapLeaveOptions`
parameters, to be able to specify if we should sign with the internal
key, and exactly which leaves we should sign.
Fixes#616
Issue #660 has been fixed by 32ae95f463f62c42c6d6aec62c1832a30298fce4,
when we moved the change calculation inside the coin selection.
This commit just adds a test to make sure that the problem is fixed.
We would previously push 108 bytes on a P2WPKH witness
to simulate signature + pubkey. This was wrong: we should push
106 bytes instead.
The max satisfaction size for a P2WPKH is 112 WU:
elements in witness (1 byte, 1WU) + OP_PUSH (1 byte, 1WU) +
pk (33 bytes, 33 WU) + OP_PUSH (1 byte, 1WU) + signature and sighash
(72 bytes, 72 WU) + scriptsig len (1 byte, 4WU)
We should push on the witness pk + signature and sighash. This is 105
WU. Since we push just once instead of twice, we add 1WU for the OP_PUSH
we are omitting.
...selecting coins
We take into account the larger segwit tx header for every
transaction, not just the segwit ones. The reason for this is that
we prefer to overestimate the fees for the transaction than
underestimating them - the former might create txs with a slightly
higher feerate than the requested one, while the latter might
create txs with a slightly lower one - or worse, invalid (<1 sat/vbyte)!
We would before calculate the TXIN_BASE_WEIGHT as prev_txid (32 bytes) +
prev_vout (4 bytes) + sequence (4 bytes) + script_sig_len (1 bytes), but
that's wrong: the script_sig_len shouldn't be included, as miniscript
already includes it in the `max_satisfaction_size` calculation.
Fixes#160
We would previously calculate the fee amount in two steps:
1. Add the weight of the empty transaction
2. Add the weight of each output
That's unnecessary: you can just use the weight of the transaction
*after* the output addition. This is clearer, but also avoids a
rare bug: if there are many outputs, adding them would cause the
"number of outputs" transaction parameter lenght to increase, and we
wouldn't notice it.
This might still happen when adding the drain output - this
commit also adds a comment as a reminder.
235011feef8a6faadc08b814e199e5d5ced2f3a0 Add assertions in the FeeRate constructor (Alekos Filini)
Pull request description:
### Description
Disallow negative, NaN, infinite or subnormal fee rate values.
### Notes to the reviewers
This commit is technically an API break because it makes the `FeeRate::from_sat_per_vb` function non-const. I think it's worth it compared to the risk of having completely nonsensical fee rates (that can break the coin selection in interesting ways).
EDIT: it's also a breaking change because our code can now panic in scenarios where it didn't before. Again, I think it's worth it.
### 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
* [ ] I'm linking the issue being fixed by this PR
ACKs for top commit:
danielabrozzoni:
re-ACK 235011feef8a6faadc08b814e199e5d5ced2f3a0
Tree-SHA512: c9432956162fadfd255edf20b825635a487adb29c88d791e18f170da79a2aac6f8e745b5e5be09be3c211697d0b1f4bddc1da75c181e8f9fc4fddf566a7a3e5c
The former way to compute and create change was inside `create_tx`, just after
performing coin selection.
It blocked the opportunity to have an "ensemble" algorithm to decide between
multiple coin selection algorithms based on a metric, like Waste.
Now, change isn't created inside `coin_select` but the change amount and the
possibility to create change is decided inside the `coin_select` method. In
this way, change is associated with the coin selection algorithm that generated
it, and a method to decide between them can be implemented.
Previously, electrum-based blockchain implementations only synced for
`scriptPubKey`s that are already cached in `Database`.
This PR introduces a feedback mechanism, that uses `stop_gap` and the
difference between "current index" and "last active index" to determine
whether we need to cache more `scriptPubKeys`.
The `WalletSync::wallet_setup` trait now may return an
`Error::MissingCachedScripts` error which contains the number of extra
`scriptPubKey`s to cache, in order to satisfy `stop_gap` for the next call.
`Wallet::sync` now calls `WalletSync` in a loop, cacheing inbetween
subsequent calls (if needed).
2c02a44586c67d1ec9720f17a3748f28c1d18643 Test: No address reuse for single descriptor (志宇)
Pull request description:
### Description
Just a simple new test.
This test is to ensure there are no regressions when we later change
internal logic of `Wallet`. A single descriptor wallet should always get
a new address with `AddressIndex::New` even if we alternate grabbing
internal/external keychains.
I thought of adding this during work on #647
### 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:
danielabrozzoni:
tACK 2c02a44586c67d1ec9720f17a3748f28c1d18643
rajarshimaitra:
tACK 2c02a44586c67d1ec9720f17a3748f28c1d18643
Tree-SHA512: d065ae0979dc3ef7c26d6dfc19c88498e4bf17cc908e4f5677dcbf62ee59162e666cb00eb87b96d4c2557310960e3677eec7b6d907a5a4860cb7d2d74dba07b0
9d2024434eb0d542133d06db14020968e713fd9b Fix: Run README.md example on the CI (meryacine)
Pull request description:
### Description
Seems like `doc(include = "../README.md")` doesn't include the readme file as doc for the dummy struct. This might be due to a difference in Rust edition used back then or something.
Fixes#637
### 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:
danielabrozzoni:
tACK 9d2024434eb0d542133d06db14020968e713fd9b
Tree-SHA512: 5842f7cdc34d76045596a248ec80bbcf86591ec9abe32d92af8322672e7a5d08d3b4baf1a000b1556542b449271dc8c438e6269eaf0204bee815c67fcf1218a8
`get_checksum_bytes` returns a descriptor checksum as `[u8; 8]` instead
of `String`, potentially improving performance and memory usage.
In addition to this, since descriptors only use charaters that fit
within a UTF-8 8-bit code unit, there is no need to use the `char` type
(which is 4 bytes). This can also potentially bring in some performance
and memory-usage benefits.
Seems like `doc(include = "../README.md")` doesn't include the readme file as docs for the dummy struct. This might be due to a difference in Rust edition used back then or something
92b9597f8b8dc3694508062b5e7c5f23acbc3a4f Rename `set_current_height` to `current_height` (Alekos Filini)
Pull request description:
### Description
Usually we don't have any prefix except for methods that can *add* to a list or replace the list entirely (e.g. `add_recipients` vs `set_recipients`)
I missed this during review of #611
### 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:
danielabrozzoni:
utACK 92b9597f8b8dc3694508062b5e7c5f23acbc3a4f - I'm sorry I didn't notice it!
Tree-SHA512: 3391068b2761bcd04d740ef41f9e772039fca7bc0e0736afcbc582ec74b6c91eb155d9e09dd7a07462eec29e32ac86e41ba339d9a550af3f754164cab6bdbf61
This bug seems to be Electrum-specific. The fix is to check the
proposed changes against the current state of the database. Ensure
newly suggested indexes are not smaller than indexes already in
database.
Changes:
* Check index updates before they are applied to database during
Electrum Blockchain sync (Thank you @rajarshimaitra for providing
an elegant solution).
Tests added:
* bdk_blockchain_tests!::test_sync_address_index_should_not_decrement
* bdk_blockchain_tests!::test_sync_address_index_should_increment
These tests ensure there will be no unexpected address reuse when
grabbing a new address via `Wallet::get_address` with `AddressIndex::New`.
Other changes:
* Tweak `rpc.rs` so that clippy is happy.
This test is to ensure there are no regressions when we later change
internal logic of `Wallet`. A single descriptor wallet should always get
a new address with `AddressIndex::New` even if we alternate grabbing
internal/external keychains.
e3a17f67d90f11d1d1a27a98ec97674c8cd3d2f7 add try_finalize to SignOptions (KaFai Choi)
c2e4ba8cbd77f1a41d269b621b0001d042c1ea57 add remove_partial_sigs to SignOptions (KaFai Choi)
Pull request description:
<!-- You can erase any parts of this template not applicable to your Pull Request. -->
### Description
This PR is to add 2 keys(`try_finalize` and `remove_partial_sigs`) in `SignOptions`. See this issue for detail https://github.com/bitcoindevkit/bdk/issues/612
### Notes to the reviewers
~I found the negative naming of these 2 new keys `do_not_finalize` and `do_not_remove_partial_sigs` are a bit confusing(like most negative named paremeter/variable). Should we actually change it back to positive naming(`do_finalize` and `do_remove_partial_sigs`)?~
### 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
* [x] I've updated `CHANGELOG.md`
#### 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:
notmandatory:
ReACK e3a17f67d90f11d1d1a27a98ec97674c8cd3d2f7
Tree-SHA512: 781b31d3ecf0bcd605206c0481fd5de3125f1c8ff18a463dbf4c821e5557847f7d70a3fe8618e100fb89f4f6899655ac0efa3593f77f915ad5bcb7e558bb2a7a
Create blockchain::GetBlockHash trait
with a method to get block hash given
a block height. Then, implement this
trait for all backends (Electrum, RPC
, Esplora, CBF). Referenced in issue 603.
5d00f8238886a993ef21056e5b3e216a4aae6951 test that BDK won't add unconf inputs when fee bumping (Daniela Brozzoni)
98748906f6799041341227de33bec20e8c6ef4b0 test: fix populate_test_db conf calculation (Daniela Brozzoni)
1d9fdd01faf3a0f17c57381a7c54d136e9d69ffe Remove wrong TODO comment in build_fee_bump (Daniela Brozzoni)
Pull request description:
### Description
Closes#144
### Notes to reviewers
#144 is describing a bug that doesn't seem to happen in BDK master anymore (BDK not respecting BIP125 rule 2). This PR just adds a test to check that the bug is 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
#### 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:
afilini:
ACK 5d00f8238886a993ef21056e5b3e216a4aae6951
Tree-SHA512: 95833f3566f9716762884d65f3f656346482e45525a3e92efa86710b9f574fdd9af7d235f1f425e4298d6ff380db9af60d1d2008ccde2588d971757db2d136b8
populate_test_db would previously give back a transaction with N + 1
confirmations when you asked for N.
This commit also fixes test_spend_coinbase, which would improperly
ask for a transaction with 0 confirmations instead of 1.
The proposed solution is bad for privacy as well.
Let's call the initial change output, which is normally shrink when you
fee bump, change#1, and the extra output aforementioned change#2 (as,
in this case, it's going to be a change output as well). If you add change#2
you might not revel change#1, but you're still revealing change#2.
You're not improving your privacy, and you're wasting money in fees.
e85aa247cb85601e96f4d82b7a996f709559223f Avoid using immature coinbase inputs (Daniela Brozzoni)
0e0d5a0e957fbf602023011d9114d8bcb8effb67 populate_test_db accepts a `coinbase` param (Daniela Brozzoni)
Pull request description:
### Description
With this PR we start considering how many confirmations a coinbase has. If it's not mature yet, we don't use it for building transactions.
Fixes#413
### Notes to the reviewers
This PR is based on #611, review that one before reviewing this 😄
007c5a78335a3e9f6c9c28a077793c2ba34bbb4e adds a coinbase parameter to `populate_test_db`, to specify if you want the db to be populated with immature coins. This is useful for `test_spend_coinbase`, but that's probably going to be the only use case.
I don't think it's a big deal to have a test function take an almost_always_useless parameter - it's not an exposed API, anyways. But, if you can come up with a different way of implementing `test_spend_coinbase` that doesn't require 007c5a78335a3e9f6c9c28a077793c2ba34bbb4e, even better! I looked for it for a while, but other than duplicating the whole `populate_test_db` code, which made the test way harder to comprehend, I didn't find any other way.
### 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:
afilini:
ACK e85aa24
Tree-SHA512: 30f470c33f9ffe928500a58f821f8ce445c653766459465eb005031ac523c6f143856fc9ca68a8e1f23a485c6543a9565bd889f9557c92bf5322e81291212a5f
612da165f8cfbc2b0aee0a37b2cdc44d6da017c8 `Blockchain` stop_gap testing improvements (志宇)
8a5f89e129d421a41af02ea85383d5b82f5ff665 Fix hang when `ElectrumBlockchainConfig::stop_gap == 0` (志宇)
Pull request description:
* Ensure `chunk_size` is > 0 during wallet sync.
* Slight refactoring for better readability.
* Add test: `test_electrum_blockchain_factory_sync_with_stop_gaps`
<!-- You can erase any parts of this template not applicable to your Pull Request. -->
### Description
`Wallet::sync` hangs indefinitely when syncing with Electrum with `stop_gap` set as 0.
The culprit is having `chunk_size` set as `stop_gap`. A zero value results in syncing not being able to progress.
Fixes#651
### 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:
afilini:
ACK 612da165f8cfbc2b0aee0a37b2cdc44d6da017c8
Tree-SHA512: 56f1bff788855facc21856209922594cff9f639c5c58ecd180a0493322a75a564b72ded330ab0b6d6c90007ce859d2b8a5d2870d619bae5ddf9a3d64837f3753
