bfcd60f5d5 test: activate all index types in feature_init.py (Martin Zumsande)
0243907fae index: Don't commit without valid m_best_block_index (Martin Zumsande)
Pull request description:
When an index thread receives an interrupt during init before it got to index anything (so `m_best_block_index == nullptr` still), it will still try to commit previous "work" before stopping the thread. That means that `BaseIndex::CommitInternal()` calls `GetLocator(nullptr)`, which returns an locator to the tip ([code](06b6369766/src/chain.cpp (L31-L32))), and saves it to the index DB.
On the next startup, this locator will be read and it will be assumed that we have successfully synced the index to the tip, when in reality we have indexed nothing.
In the case of coinstatsindex, this would lead to a shutdown of bitcoind without any indication what went wrong. For the other indexes, there would be no immediate shutdown, but the index would be corrupt.
This PR fixes this by not committing when `m_best_block_index==nullptr`, and it also adds an error log message to the silent coinstatsindex shutdown path.
This is another small bug found by `feature_init.py` - the second commit enables blockfilterindex and coinstatsindex for this test, enabling coinstatsindex without the first commit would have led to frequent failures.
ACKs for top commit:
fjahr:
reACK bfcd60f5d5
shaavan:
reACK bfcd60f5d5
Tree-SHA512: 8e2bac0fc40cde209518a9e59b597ae0a5a875a2a90898673987c91733718d40e528dada942bf552b58bc021bf46e59da2d0cc5a61045f48f9bae2b1baf6033b
f11dad22a5 test: refactor: remove unneeded bytes<->hex conversions in `byte_to_base58` (Sebastian Falbesoner)
Pull request description:
It seems like the only reason for using hex strings in this method was to have a convenient way to convert to an integer from the input data interpreted as big-endian. In Python3 we have `int.from_bytes(..., 'big')` for that purpose, hence there is no need for that anymore and we can simply operate on bytes only.
ACKs for top commit:
laanwj:
Code review ACK f11dad22a5
Tree-SHA512: 9b1563010066ca74d85139c3b9259e9a5bb49e1f141c30b6506a0445afddb2bde7fd421fdd917dc516956e66f93610e2c21d720817640daee8f57f803be76ee4
It seems like the only reason for using hex strings in this
method was to have a convenient way to convert to an integer
from the input data interpreted as big-endian.
In Python3 we have `int.from_bytes(..., 'big')` for that
purpose, hence there is no need for that anymore and we can
simply operate on bytes only.
d1fab9d5d2 test: Call ceildiv helper with integer (Martin Zumsande)
Pull request description:
On master,
`assert_fee_amount(Decimal("0.00000993"), 217, Decimal("0.00004531"))` passes
`assert_fee_amount(Decimal("0.00000993"), Decimal("217"), Decimal("0.00004531"))` fails.
the reason is that the // operator in `ceildiv(a,b) = -(-a//b)` has a different behavior for Decimals, see [doc](https://docs.python.org/3/library/decimal.html#decimal-objects).
`wallet_send.py` calls this function with Decimals, and I think this is the reason for the failure reported in the OP of #24151 (`wallet_send.py --legacy-wallet` line 332, the numbers used in the example above are from there). However, the other failures reported there cannot be explained by this, so this is just a partial fix.
ACKs for top commit:
ryanofsky:
Code review ACK d1fab9d5d2. Tracking down this problem was a good find, and code seems safer and easier to understand now
Tree-SHA512: 5bf0568cd1a0824f6b1a15a03580b6e9391b4f51112a97c1d00469d255bf6dda45c49a36fa567a5ba9b9973efe1d9cdd480db91965c9f4c2aa963629a8a32cba
a036358994 test: Repair failfast option for test runner (Martin Zumsande)
Pull request description:
Fixes#23990
After #23799, the `--failfast` option in the test runner for the functional tests stopped working, because a second outer loop was introduced, which would have needed a `break` too for the test runner to fail immediately. This also led to the errors reported in #23990.
This provides a straightforward fix for that.
There is also #23995 which is a larger refactor, but that hasn't been updated in a while to fix the failfast issue.
ACKs for top commit:
pg156:
Tested ACK a036358994. I agree adding the `all_passed` flag to break out of the outer loop when needed makes sense. The "failfast" option works after this change.
Tree-SHA512: 3e2f775e36c13d180d32a05cd1cfe0883274e8615cdbbd4e069a9899e9b9ea1091066cf085e93f1c5326bd8ecc6ff524e0dad7c638f60dfdb169fefcdb26ee52
Warning: Replacing fs::system_complete calls with fs::absolute calls
in this commit may cause minor changes in behaviour because fs::absolute
no longer strips trailing slashes; however these changes are believed to
be safe.
Co-authored-by: Russell Yanofsky <russ@yanofsky.org>
Co-authored-by: Hennadii Stepanov <32963518+hebasto@users.noreply.github.com>
438e6f4c33 test: speedup interface_rest.py by whitelisting peers (immediate tx relay) (Sebastian Falbesoner)
11b9684dfd test: use MiniWallet for rest_interface.py (Sebastian Falbesoner)
Pull request description:
This PR enables one more of the non-wallet functional tests (interface_rest.py) to be run even with the Bitcoin Core wallet disabled by using the MiniWallet instead, as proposed in #20078.
Note that the original test sent funds from one node to another and checked node's balances, but the state of a node's wallet is not relevant to any of the REST endpoints, i.e. the replacement is quite stright-forward. In an additional commit, the test is further sped up by using the good ol' immediate tx relay trick (parameter `-whitelist=noban@127.0.0.1`).
ACKs for top commit:
brunoerg:
ACK 438e6f4c33
Tree-SHA512: eac351c5fb7e043c36c193d51545f20f563be9aaa04f3429a2bfb452ae4aa72294d2552800d6cac55c9a3ec2b4f30bcda2abcd74736dec3ed75e7d83c5af437f
3ee6d0788e test: add more wallet conflicts assertions (S3RK)
3b98bf9c43 Revert "Add to spends only transcations from me" (S3RK)
Pull request description:
This reverts commit d04566415e from #22929.
This commit was based on invalid assumption that `mapTxSpends` should contain only outgoing txs and broke wallet conflicts feature.
ACKs for top commit:
achow101:
ACK 3ee6d0788e
Tree-SHA512: bf5a77ced6bac57d5eb85771d9189c53e1edc295d179ed5a1bdce18e365794a9101b4cecf35387b27f67260db3b47f7214e7876e490494529b748cceeb95632d
By whitelisting the peers via -whitelist, the inventory is transmissioned
immediately rather than on average every 5 seconds, speeding up the test by at
least a factor of two:
before:
$ time ./interface_rest.py
...
0m14.82s real 0m01.44s user 0m01.19s system
with this commit:
$ time ./interface_rest.py
...
0m05.67s real 0m01.07s user 0m01.35s system
This is needed for the next commit.
Also, it doesn't really test anything novel. wait_for_debug_log is
inherently racy, so will randomly terminate at the exact point or later.
So the randomization is already sufficiently covered by the existing
test.
a380922891 Release notes for getdeploymentinfo rpc (Anthony Towns)
240cad09ba rpc: getdeploymentinfo: include signalling info (Anthony Towns)
376c0c6dae rpc: getdeploymentinfo: include block hash/height (Anthony Towns)
a7469bcd35 rpc: getdeploymentinfo: change stats to always refer to current period (Anthony Towns)
7f15c1841b rpc: getdeploymentinfo: allow specifying a blockhash other than tip (Anthony Towns)
fd826130a0 rpc: move softfork info from getblockchaininfo to getdeploymentinfo (Anthony Towns)
Pull request description:
The aim of this PR is to improve the ability to monitor soft fork status. It first moves the softfork section from getblockchaininfo into a new RPC named getdeploymentinfo, which is then also able to query the status of forks at an arbitrary block rather than only at the tip. In addition, bip9 status is changed to indicate the status of the given block, rather than just for the next block, and an additional field is included to indicate whether each block in the signalling period signaled.
ACKs for top commit:
laanwj:
Code review and lightly tested ACK a380922891
Sjors:
tACK a380922891
fjahr:
tACK a380922891
Tree-SHA512: 7417d733b47629f229c5128586569909250481a3e94356c52fe67a03fd42cd81745246e384b98c4115fb61587714c879e4bc3e5f5c74407d9f8f6773472a33cb
6498ba151b transaction decoding infer output descriptors (Gregory Sanders)
Pull request description:
Following discussion in #16725 this is complementary data to expose. All outputs are inferred.
ACKs for top commit:
achow101:
ACK 6498ba151b
meshcollider:
utACK 6498ba151b
Tree-SHA512: 36664117ddbe46d5fdde7ed6541ef2c9d8dfb7a3636b97f363bf1c325096fe00d9d2acea2d1917ea19fdb82f1ea296c12e440c5c703d6a9bfc1a02fba028bcd8
449dffc610 test, bugfix: fix intermittent failure in getrawtransaction_tests (Jon Atack)
Pull request description:
Easiest to review with `colorMoved = dimmed-zebra` and `colorMovedWs = allow-indentation-change`.
Resolves#23991
Top commit has no ACKs.
Tree-SHA512: a50c597857e45c52dae1b6fc5988ea077c09f2b32ffe3605f0c64bd780f5dd2635755cb26c4981233bf925c5ea9f7310163349a549239dc85657558f253ffb7b
75656adfd2 test: add functional test for `-maxtipage` parameter (Sebastian Falbesoner)
Pull request description:
This PR adds a missing test for the `-maxtipage` parameter which controls what is the allowed maximum tip age for leaving IBD:
792d0d8d51/src/init.cpp (L540)
Relevant code path in the `CChainState::IsInitialBlockDownload` method:
792d0d8d51/src/validation.cpp (L1479-L1480)
The test is pretty simple and should be self-explanatory.
ACKs for top commit:
MarcoFalke:
review ACK 75656adfd2
Tree-SHA512: 0a10dca13cb18c29e64fc8412f4c8f2bcaff1bab8645bd85266c242ba88ce036a150c03cbbe9810c3bb44649810af0aa9cb3584dbae886a7bdb16b72150d08de
fac8165443 Remove unused checkFinalTx (MarcoFalke)
fa272eab44 wallet: Avoid dropping confirmed coins (MarcoFalke)
888841ea8d interfaces: Remove unused is_final (MarcoFalke)
dddd05e7a3 qt: Treat unconfirmed txs as unconfirmed (MarcoFalke)
Pull request description:
The wallet has several issues:
## Unconfirmed txs in the GUI
The GUI clumsily attempts to guess if unconfirmed txs are locked until a future time. This is currently based on the locktime only, not nSequence, thus wrong. Fix this by removing the clumsy code and treat all unconfirmed txs as unconfirmed. The GUI already prints whether a tx is in the mempool, in which case the user knows that the tx wasn't locked until a future time. If the tx is not in the mempool, it might be better to report the exact reject reason from the mempool instead of using incorrect heuristics.
## Confirmed txs in the wallet
The wallet drops coins that it incorrectly assumes to be locked until a future time, even if they are already confirmed in the chain. This is because the wallet is using the wrong time (adjusted network time) instead of MTP, due to the `-1` default argument of `CheckFinalTx`.
The issues are fixed in separate commits and there is even a test.
ACKs for top commit:
achow101:
ACK fac8165443
prayank23:
reACK fac8165443
glozow:
code review ACK fac8165443, I understand now how this fixes both issues.
Tree-SHA512: 210afb855f4c6d903fee49eba6b1a9735d699cf0168b669eabb38178e53b3a522258b7cc669f52489c6cd3e38bf358afde12eef3ba2e2f2ffaeb06b8f652ccd0
3866272c45 tests: Test specifying input weights (Andrew Chow)
6fa762a372 rpc, wallet: Allow users to specify input weights (Andrew Chow)
808068e90e wallet: Allow user specified input size to override (Andrew Chow)
4060c50d7e wallet: add input weights to CCoinControl (Andrew Chow)
Pull request description:
When funding a transaction with external inputs, instead of providing solving data, a user may want to just provide the maximum signed size of that input. This is particularly useful in cases where the input is nonstandard as our dummy signer is unable to handle those inputs.
The input weight can be provided to any input regardless of whether it belongs to the wallet and the provided weight will always be used regardless of any calculated input weight. This allows the user to override the calculated input weight which may overestimate in some circumstances due to missing information (e.g. if the private key is not known, a maximum size signature will be used, but the actual signer may be doing additional work which reduces the size of the signature).
For `send` and `walletcreatefundedpsbt`, the input weight is specified in a `weight` field in an input object. For `fundrawtransaction`, a new `input_weights` field is added to the `options` object. This is an array of objects consisting of a txid, vout, and weight.
Closes#23187
ACKs for top commit:
instagibbs:
reACK 3866272c45
glozow:
reACK 3866272 via range-diff
t-bast:
ACK 3866272c45
Tree-SHA512: 2c8b471ee537c62a51389b7c4e86b5ac1c3a223b444195042be8117b3c83e29c0619463610b950cbbd1648d3ed01ecc5bb0b3c4f39640680da9157763b9b9f9f
c5b36b1c1b Mempool Update Cut-Through Optimization (Jeremy Rubin)
c49daf9885 [TESTS] Increase limitancestorcount in tournament RPC test to showcase improved algorithm (Jeremy Rubin)
Pull request description:
Often when we're updating mempool entries we update entries that we ultimately end up removing the updated entries shortly thereafter. This patch makes it so that we filter for such entries a bit earlier in processing, which yields a mild improvement for these cases, and is negligible overhead otherwise.
There's potential for a better -- but more sophisticated -- algorithm that can be used taking advantage of epochs, but I figured it is better to do something that is simple and works first and upgrade it later as the other epoch mempool work proceeds as it makes the patches for the epoch algorithm simpler to understand, so you can consider this as preparatory work. It could either go in now if it is not controversial, or we could wait until the other patch is ready to go.
ACKs for top commit:
instagibbs:
reACK c5b36b1
sipa:
utACK c5b36b1c1b
mzumsande:
Code Review ACK c5b36b1c1b
Tree-SHA512: 78b16864f77a637d8a68a65e23c019a9757d8b2243486728ef601d212ae482f6084cf8e69d810958c356f1803178046e4697207ba40d6d10529ca57de647fae6
Added tests to rpc_fundrawtransaction, wallet_send, and rpc_psbt that
test that external inputs can be spent when input weight is provided.
Also tested that the input weight overrides any calculated weight.
Additionally, rpc_psbt's external inputs test is cleaned up a bit to be
more similar to rpc_fundrawtransaction's and avoid potential pitfalls
due to non-deterministic coin selection behavior.
dc5d6b0d47 fs: Make compatible with boost 1.78 (Andrew Chow)
Pull request description:
Boost 1.78 removed `operator+` in a way that breaks our usage of it in a subclass. A [proposed workaround](https://github.com/boostorg/filesystem/issues/223#issuecomment-1000230207) for this is to cast the argument to `boost::filesystem::path`, and this is backwards compatible with older versions of boost.
Additionally, it appears that `fs::canonical` no longer removes trailing slashes. This was causing a test to fail. The solution is to explicitly remove the trailing separator in the one place that `fs::canonical` is used.
Lastly, `fs::create_directories` now has an error message saying `create_directories` instead of `create_directory`. This caused wallet_multiwallet.py to fail. The error message check has been updated to be able accept either string.
Fixes#23846
ACKs for top commit:
ryanofsky:
Code review ACK dc5d6b0d47
vincenzopalazzo:
ACK dc5d6b0d47
Tree-SHA512: d4d8e7b49b8dfbf0ced9bfe9a2b3827841227fc755fc799f19159076b0ccf882432cc8b6ad93cdeda98fb58b942b9ba50a9e0a6b4f6b1e0097e80f1074ae5682
bd52684508 test: rest /tx with an invalid/unknown txid (brunoerg)
Pull request description:
This PR adds test coverage to the endpoint `/tx` (rest) passing an invalid and an unknown txid to test its return.
Invalid -> should return status code 400 (bad request)
Unknown -> should return status code 404 (not found)
ACKs for top commit:
kallewoof:
ACK bd52684508
Tree-SHA512: a7fbb63f30d06fc0855133a36e8317c7930ba13aa2b4a2dd1fc35079d59eacace72e1ffe7ae1b3e067066fe51792415940d72d923e83a659a0d5965e4110b32a
On a period boundary, getdeploymentinfo (and previously getblockchaininfo)
would report the status and statistics for the next block rather than
the current block. Change this to always report the status/statistics
of the current block, but add status-next to report the status for the
next block.
aa8a65e4a8 test: use MiniWallet for mempool_accept.py (Sebastian Falbesoner)
b24f6c6855 test: MiniWallet: support default `from_node` for creating txs (Sebastian Falbesoner)
f30041c914 test: create txs with current `nVersion` (2) by default (Sebastian Falbesoner)
2f79786822 test: refactor: add constant for sequence number `SEQUENCE_FINAL` (Sebastian Falbesoner)
Pull request description:
This PR enables one more of the non-wallet functional tests (mempool_accept.py) to be run even with the Bitcoin Core wallet disabled by using the MiniWallet instead, as proposed in #20078.
It also includes some other minor changes that came up while working on the replacement:
* [commit 1/4] replace magic number 0xffffffff for a tx's nSequence with a new constant `SEQUENCE_FINAL`
* [commit 2/4] create `CTransaction` instances with the current nVersion=2 by default, in order to use BIP68 for tests
* [commit 3/4] support default `from_node` parameter for creating txs (this is a stripped down version of PR #24025)
ACKs for top commit:
MarcoFalke:
re-ACK aa8a65e4a8📊
Tree-SHA512: 34cd085ea4147ad5bd3f3321c84204064ceb95f382664c7fe29062c1bbc79d9d9465c5e46d35e11c416f2f3cd46030c90a09b518c829c73ae40d060be5e4c9cb
If no `from_node` parameter is passed explicitely to the
`create_self_transfer` method, the test node passed in the course
of creating the MiniWallet instance is used. This seems to
be the main use-case in most of the current functional
tests, i.e. in many instances the calls can be shortened.
c9374af102 test: set ban after mocking time (brunoerg)
Pull request description:
Fixes#23988
Set ban after mocking time to avoid intermittent failures related to the assertion of ban_duration and time_remaining.
See: https://cirrus-ci.com/task/6754020390862848?logs=ci#L4652
ACKs for top commit:
mzumsande:
Tested ACK c9374af102
vincenzopalazzo:
ACK c9374af102
Tree-SHA512: fac3ac91a045bb46334d7c568f6a53a3b0a45b306914a54ea13bcc845734eaaad1ff295ff3ab158037fd9d08df77344058331336110b8f7888832b16b0589be5
0754e9c01b test: run feature_pruning.py without wallet compiled (Sebastian Falbesoner)
Pull request description:
Only one small part of the pruning test (sub-test `wallet_test`) is wallet-related, hence we can run all other parts without wallet compiled.
ACKs for top commit:
MarcoFalke:
cr ACK 0754e9c01b
Tree-SHA512: 856856903d21d64953ed0102cc2a96f55975c4b7d8e93e57b82c266024967160df64df2b6068be089efc05e883e8d6d12e7327053420d4c640b9d8cc5bcb1c58
