9dfc61d95f test: detect no external signer connected (Sjors Provoost)
0a4ee93529 wallet: use PSBTError::EXTERNAL_SIGNER_NOT_FOUND (Sjors Provoost)
8ba2f9b7c8 refactor: use util::Result for GetExternalSigner() (Sjors Provoost)
Pull request description:
When attempting to sign a transaction involving an external signer, if the device isn't connected we throw an `std::runtime_error`. This prevents the (mainly GUI) code that's actually supposed to handle this case from running.
This PR returns a `PSBTError::EXTERNAL_SIGNER_NOT_FOUND` instead of throwing.
The first commit is a refactor to have `GetExternalSigner()` return a `util::Result<ExternalSigner>` so the caller can decide how to handle the error. There are two other places where call `GetExternalSigner()` which this PR doesn't change (which I think is fine there).
Before:
After (the translation already exist):
Fixes#32426
Additionally use `LogWarning` instead of `std::cerr` for both a missing signer and failure to sign.
ACKs for top commit:
achow101:
ACK 9dfc61d95f
brunoerg:
code review ACK 9dfc61d95f
Tree-SHA512: 22515f4f0b4f50cb0ef532b729e247f11a68be9c90e384942d4277087b2e76806a1cdaa57fb51d5883dacf0a428e5279674aab37cce8c0d3d7de0f96346b8233
useful to easily create transactions with same txid, different
wtxid and valid witness for testing scenarios in other places
(ex: private broadcast connections)
Additionally this commit gives each test its
own function.
The assert_submitblock helper is absorbed into
assert_template.
Review hint:
git show --color-moved=dimmed-zebra
0def84d407 test: Verify parent_desc in RPCs (Ava Chow)
2554cee988 test: Enable default wallet for wallet_descriptor.py (Ava Chow)
3fc9d9f241 wallet, rpc: Push the normalized parent descriptor (Ava Chow)
Pull request description:
Instead of prividing the descriptor string as stored in the db, use the normalized descriptor as is done for getaddressinfo's parent_desc field.
Split from #32489
ACKs for top commit:
Sjors:
re-utACK 0def84d407
rkrux:
ACK 0def84d407
w0xlt:
reACK 0def84d407
Tree-SHA512: 575c5b545d6f0aa7e135696b7a55c004e754fca4dd35dd9cf71b0b45b49a2e86e7b20570e768534d587005953bb893645379ec1ba4f98cfd26811f9c2f17de2d
The catchup loop in the outbound eviction functional test currently has
a small flaw, as the contained waiting for a `getheaders` message just
waits for any such message instead of one with the intended block hash.
The reason is that the `prev_prev_hash` variable is set incorrectly,
since the `tip_header` instance is not updated and its field `.hash` is
None. Fix that by updating `tip_header` and use the correct field -- we
want the tip header's previous hash (`.hashPrevBlock`).
when we reconsiderblock, previously only block and it's
descendants were considered as chain tip candidates/inserted into
setBlockIndexCandidates
ex: on this chain, with block 4 invalidated
1 -> 2 -> 3 -> 4 -> 5 -> 6 -> header 7
blocks 4, 5, 6, header 7 have BLOCK_FAILED_* flags set
previously:
- if we reconsiderblock header 7, the chain would have all the
BLOCK_FAILED_* flags cleared but would report chain tip as block 3.
- after restart, it reports correct chain tip block 6.
now:
- if we reconsiderblock header 7, the correct chain tip block 6 is
reported since ancestors are also considered as chain tip
candidates/inserted into setBlockIndexCandidates.
Co-authored-by: Martin Zumsande <mzumsande@gmail.com>
4ef6253017 test: avoid unneeded (w)txid hex -> integer conversions (Sebastian Falbesoner)
472f3770ae scripted-diff: test: rename CTransaction `.getwtxid()` -> `wtxid_hex` for consistency (Sebastian Falbesoner)
81af4334e8 test: rename CTransaction `.sha256` -> `.txid_int` for consistency (Sebastian Falbesoner)
ce83924237 test: rename CTransaction `.rehash()`/`.hash` -> `.txid_hex` for consistency (Sebastian Falbesoner)
e9cdaefb0a test: introduce and use CTransaction `.wtxid_int` property (Sebastian Falbesoner)
9b3dce24a3 test: remove bare CTransaction `.rehash()`/`.calc_sha256()` calls (Sebastian Falbesoner)
a2724e3ea3 test: remove txid caching in CTransaction class (Sebastian Falbesoner)
Pull request description:
In the functional test framework, determining a (w)txid for a `CTransaction` instance is currently rather confusing and footgunny due to inconsistent naming/interfaces (see table below) and statefulness involved. This PR aims to improve that by:
* removing the (w)txid caching mechanism, in order to avoid the need to call additional rehashing functions (`.rehash()`/`.calculate_sha256()`, see first two commits and https://github.com/bitcoin/bitcoin/pull/32050#discussion_r1993286997). This change in theory decreases the performance, as the involved serialization and hashing involved might be called more often than previously, but I couldn't find a functional test where this leads to a measurable run-time increase on my machine.
* introduce consistent naming that shows the type of the returned txid, i.e. hex string vs. test-framework-internal representation [currently integers] (see remaining commits)
Summary table showing (w)txid determaination before/after this PR:
| Task | master | PR |
|:-----------------------|:-----------------------|:-------------|
| get TXID (hex string) | `.rehash()` / `.hash`[1] | `.txid_hex` |
| get TXID (integer) | `.sha256`[1] | `.txid_int` |
| get WTXID (hex string) | `.getwtxid()` | `.wtxid_hex` |
| get WTXID (integer) | `.calc_sha256(True)` | `.wtxid_int` |
Unfortunately, most renames can't be done with a scripted-diff, as the property names (`.hash`, `.sha256`) are also used for blocks and other message types. The PR is rather invasive and touches a lot of files, but I think it's worth to do it, also to make life easier for new contributors. Future tasks like e.g. doing the same overhaul for block (header) objects or getting rid of the integer representation (see https://github.com/bitcoin/bitcoin/pull/32050) become easier should become easier after this one.
[1] = returned value might be out-of-date, if rehashing function wasn't called after modification
ACKs for top commit:
maflcko:
re-ACK 4ef6253017🏈
achow101:
ACK 4ef6253017
marcofleon:
code review ACK 4ef6253017
Tree-SHA512: 4b472c31d169966b6f6878911a8404d25bf3e503b6e8ef30f36a7415d21ad4bc1265083af2d3ead6edfcd9fac9ccb0a8be57e1b0739ad431b836413070d7d583
Both are rational numbers. Client software should only use them to
display information to humans. Followup calculations should use the
underlying values such as target.
Therefore it's not necessary to test the handling of these floating
point values. Round them down to avoid spurious test failures.
Fixes#32515
Rather than determining a CTransaction's (w)txid as an integer by
converting it's hex value, it can be directly accessed via the
introduced `.{w,}txid_int` property.
e98c51fcce doc: update tor.md to mention the new -proxy=addr:port=tor (Vasil Dimov)
ca5781e23a config: allow setting -proxy per network (Vasil Dimov)
Pull request description:
`-proxy=addr:port` specifies the proxy for all networks (except I2P). Previously only the Tor proxy could have been specified separately via `-onion=addr:port`.
Make it possible to specify separately the proxy for IPv4, IPv6, Tor and CJDNS by e.g. `-proxy=addr:port=ipv6`. Or remove the proxy for a given network, e.g. `-proxy=0=cjdns`.
Resolves: https://github.com/bitcoin/bitcoin/issues/24450
ACKs for top commit:
pinheadmz:
ACK e98c51fcce
caesrcd:
reACK e98c51fcce
danielabrozzoni:
Code Review ACK e98c51fcce
1440000bytes:
ACK e98c51fcce
Tree-SHA512: 0cb590cb72b9393cc36357e8bd7861514ec4c5bc044a154e59601420b1fd6240f336ab538ed138bc769fca3d17e03725d56de382666420dc0787895d5bfec131
f16c8c67bf tests: Expand HTTP coverage to assert libevent behavior (Matthew Zipkin)
Pull request description:
These commits are cherry-picked from #32061 and part of a project to [remove libevent](https://github.com/bitcoin/bitcoin/issues/31194).
This PR only adds functional tests to `interface_http` to cover some HTTP server behaviors we inherit from libevent, in order to maintain those behaviors when we replace libevent with our own HTTP server.
1. Pipelining: The server must respond to requests from a client in the order in which they were received [RFC 7230 6.3.2](https://www.rfc-editor.org/rfc/rfc7230#section-6.3.2)
2. `-rpcservertimeout` config option which sets the amount of time the server will keep an idle client connection alive
3. "Chunked" Transfer-Encoding: Allows a client to send a request in pieces, without the `Content-Length` header [RFC 7230 4.1](https://www.rfc-editor.org/rfc/rfc7230#section-4.1)
ACKs for top commit:
achow101:
ACK f16c8c67bf
vasild:
ACK f16c8c67bf
polespinasa:
ACK f16c8c67bf
fjahr:
utACK f16c8c67bf
Tree-SHA512: 405b59431b4d2bf118fde04b270865dee06ef980ab120d9cc1dce28e5d65dfd880a57055b407009d22f4de614bc3eebdb3e203bcd39e86cb14fbfd62195ed06a
Since the previous commit, CTransaction object calls to the
methods `.rehash()` and `.calc_sha256()` are effectively no-ops
if the returned value is not used, so we can just remove them.
Rather than txids (represented by the fields `.sha256` and `.hash`)
being stateful, simply compute them on-the-fly. This ensures that
the correct values are always returned and takes the burden of
rehashing from test writers, making the code shorter overall.
In a first step, the fields are kept at the same name with @property
functions as drop-in replacements, for a minimal diff. In later commits,
the names are changed to be more descriptive and indicating the return
type of the txid.
a189d63618 add release note for datacarriersize default change (Greg Sanders)
a141e1bf50 Add more OP_RETURN mempool acceptance functional tests (Peter Todd)
0b4048c733 datacarrier: deprecate startup arguments for future removal (Greg Sanders)
63091b79e7 test: remove unnecessary -datacarriersize args from tests (Greg Sanders)
9f36962b07 policy: uncap datacarrier by default (Greg Sanders)
Pull request description:
Retains the `-datacarrier*` args, marks them as deprecated, and does not require another startup argument for multiple OP_RETURN outputs.
If a user has set `-datacarriersize` the value is "budgeted" across all seen OP_RETURN output scriptPubKeys. In other words the total script bytes stays the same, but can be spread across any number of outputs. This is done to not introduce an additional argument to support multiple outputs.
I do not advise people use the option with custom arguments and it is marked as deprecated to not mislead as a promise to offer it forever. The argument itself can be removed in some future release to clean up the code and minimize footguns for users.
ACKs for top commit:
stickies-v:
re-ACK a189d63618
Sjors:
re-ACK a189d63618
polespinasa:
re-ACK a189d63618
hodlinator:
re-ACK a189d63618
ajtowns:
reACK a189d63618
mzumsande:
re-ACK a189d63618
petertodd:
ACK a189d63618
theStack:
re-ACK a189d63618
1440000bytes:
re-ACK a189d63618
willcl-ark:
ACK a189d63618
dergoegge:
ACK a189d63618
fanquake:
ACK a189d63618
murchandamus:
ACK a189d63618
darosior:
Concept ACK a189d63618.
Tree-SHA512: 3da2f1ef2f50884d4da7e50df2121bf175cb826edaa14ba7c3068a6d5b2a70beb426edc55d50338ee1d9686b9f74fdf9e10d30fb26a023a718dd82fa1e77b038
12ff4be9c7 test: ensure -rpcallowip is compatible with RFC4193 (Matthew Zipkin)
c02bd3c187 config: Explain RFC4193 and CJDNS interaction in help and init error (Matthew Zipkin)
f728b6b111 init: Configure reachable networks before we start the RPC server (Matthew Zipkin)
Pull request description:
Closes https://github.com/bitcoin/bitcoin/issues/32433
`MaybeFlipIPv6toCJDNS()` relies on `g_reachable_nets` to distinguish between CJDNS addresses and other IPv6 addresses. In particular, [RFC4193](https://www.rfc-editor.org/rfc/rfc4193#section-3.1) address or "Unique Local Address" with the L-bit unset also begins with the `fc` prefix. #32433 highlights a use case for these addresses that have nothing to do with CJDNS.
On master we don't parse init flags like `-cjdnsreachable` until *after* the HTTP server has started, causing conflicts with `-rpcallowip` because CJDNS doesn't support subnets.
This PR ensures that `NET_CJDNS` is only present in the reachable networks list if set by `-cjdnsreachable` *before* `-rpcallowip` is checked. If it is set all `fc` addresses are assumed to be CJDNS, can not have subnets, and can't be set for `-rpcallowip`.
I also noted this specific parameter interaction in the init help as well as the error message if configured incorrectly.
This can be tested locally:
`bitcoind -regtest -rpcallowip=fc00:dead:beef::/64 -rpcuser=u -rpcpassword=p`
On master this will just throw an error that doesn't even mention IPv6 at all.
On the branch, this will succeed and can be tested by adding the ULA to a local interface.
On linux: `sudo ip -6 addr add fc00:dead:beef::1/64 dev lo`
On macos: `sudo ifconfig lo0 inet6 fc00:dead:beef::1/128 add`
then: `curl -v -g -6 --interface fc00:dead:beef::1 u:p@[::1]:18443 --data '{"method":"getblockcount"}'`
If the `rpcallowip` option is removed, the RPC request will fail to authorize.
Finally, adding `-cjdnsreachable` to the start up command will throw an error and specify the incompatibility:
> RFC4193 is allowed only if -cjdnsreachable=0.
ACKs for top commit:
achow101:
ACK 12ff4be9c7
tapcrafter:
tACK 12ff4be9c7
ryanofsky:
Code review ACK 12ff4be9c7
willcl-ark:
ACK 12ff4be9c7
Tree-SHA512: a4dd70ca2bb9f6ec2c0a9463fd73985d1ed80552c674a9067ac9a86662d1c018cc275ba757cebb2993c5f3971ecf4778b95d35fe7a7178fb41b1d18b601c9960
7cfbb8575e test: wallet: cover wallet passphrase with a null char (brunoerg)
Pull request description:
This PR adds test coverage for the `walletpassphrase`/`walletpassphrasechange` RPC when the passphrase is incorrect due to a null character.
For reference: https://github.com/bitcoin/bitcoin/pull/27068 introduced the usage of `SecureString` to allow null characters.
ACKs for top commit:
maflcko:
lgtm ACK 7cfbb8575e
achow101:
ACK 7cfbb8575e
w0xlt:
Code review ACK 7cfbb8575e
BrandonOdiwuor:
Code Review ACK 7cfbb8575e
theStack:
ACK 7cfbb8575e
pablomartin4btc:
cr ACK 7cfbb8575e
Tree-SHA512: ecdb48662ceb6c55c4b301ca7f537c3159ece7b66ee40ea977583ffb74bd3d06e334ab3a5639a9cde3aa6443129f412f9aea0ee5a8b73b31dba0728d0890b7f1
6ee32aaaca test: signet tool genpsbt and solvepsbt commands (Sjors Provoost)
0a99d99fe4 signet: miner skips PSBT step for OP_TRUE (Sjors Provoost)
cdfb70e5a6 signet: split decode_psbt miner helper (Sjors Provoost)
Pull request description:
[BIP325](https://github.com/bitcoin/bips/blob/master/bip-0325.mediawiki) mentions the following rule:
> In the special case where an empty solution is valid (ie scriptSig and scriptWitness are both empty) this additional commitment can optionally be left out. This special case is to allow non-signet-aware block generation code to be used to test a custom signet chain where the challenge is trivially true.
Such a signet can be created using e.g. `-signetchallenge=51` (`OP_TRUE`). However `contrib/signet/miner` won't omit the commitment.
This PR improves the miner by skipping the PSBT for known trivial scripts (just `OP_TRUE` and trivial pushes for now). This prevents it from appending the 4 byte signet header to the witness commitment, as allowed by the above rule.
---
Previously the script would fail with `PSBT signing failed`, making it difficult to mine. This is no longer the case.
ACKs for top commit:
achow101:
ACK 6ee32aaaca
theStack:
re-ACK 6ee32aaaca
danielabrozzoni:
ACK 6ee32aaaca
Tree-SHA512: e47fbf471f2909286a6c1c073799ea388b9c19551afcce96cf9af45cc48d25c02f1e48e08861a88b604361e2c107a759d5baf393da8a37360de419f31651758a
86e1111239 test: verify node skips loading legacy wallets during startup (furszy)
9f94de5bb5 wallet: init, don't error out when loading legacy wallets (furszy)
Pull request description:
Instead of failing during initialization and shutting down the app when encountering a legacy wallet, skip loading the wallet and notify the user accordingly.
This allows users to access migration functionalities without needing to manually remove the wallet from settings.json or resort to using the bitcoin-wallet utility.
This means that GUI users will be able to use the migration button, and bitcoin-cli users will be able to call the migratewallet RPC directly after init.
ACKs for top commit:
achow101:
ACK 86e1111239
w0xlt:
ACK 86e1111239
Tree-SHA512: 85d594a503ee7a833a23754b71b6ba4869ca34ed802c9ac0cd7b2fa56978f5fcad84ee4bd3acdcc61cf8e7f08f0789336febc5d76beae1eebf7bd51462512b78
getaddressinfo, listunspent, listtransactions, listsinceblock, and
gettransaction all include parent_desc(s). Make sure that these are
consistent with each other, as well as being in normalized form.
Datacarrier output script sizes and output counts are now
uncapped by default.
To avoid introducing another startup argument, we modify the
OP_RETURN accounting to "budget" the spk sizes.
If a user has set a custom default, this results in that
budget being spent over the sum of all OP_RETURN outputs'
scripts in the transaction, no longer capping the number
of OP_RETURN outputs themselves. This should allow a
superset of current behavior while respecting the passed
argument in terms of total arbitrary data storage.
Co-authored-by: Anthony Towns <aj@erisian.com.au>
f3a444c45f gui: Disallow loading legacy wallets (Ava Chow)
09955172f3 wallet, rpc: Give warning in listwalletdir for legacy wallets (Ava Chow)
Pull request description:
A new field `warnings` is added for each wallet in `listwalletdir`. If a legacy wallet is detected, the warning will contain a message that the wallet is a legacy wallet and will need to be migrated before it can be loaded.
In the GUI, the "Open Wallet" menu is changed to show legacy wallets greyed out with "(needs migration)" appended to their name to indicate to the user that the legacy wallet will need to be migrated.
ACKs for top commit:
maflcko:
lgtm ACK f3a444c45f
adyshimony:
Test ACK [f3a444c](f3a444c45f)
furszy:
Code review ACK f3a444c45f
w0xlt:
Code Review ACK f3a444c45f
Tree-SHA512: 496caec0ca37845487bd709e592240315eb23461fbd697e68a7fde8e4d9b74b48aab1212c88dbbcc8a107a896b824c2e1f69691068641812ae903f873fa2f22b
b1ea542ae6 test: test MAX_SCRIPT_SIZE for block validity (Greg Sanders)
Pull request description:
I don't believe there are direct tests for this.
ACKs for top commit:
achow101:
ACK b1ea542ae6
TheCharlatan:
ACK b1ea542ae6
theStack:
ACK b1ea542ae6
Tree-SHA512: 1d7d3eab9c54977844bf2ca1aa403b070aae0f818db2fb5cae367d1c4d12f1e403b6fdec224af769a2ebb648cbca8bfd0d7df5db2a89fccf256c9c244484eba2
84aa484d45 test: fix transaction_graph_test reorg test (Greg Sanders)
eaf44f3767 test: check chainlimits respects on reorg (Greg Sanders)
47894367b5 functional test: add MAX_DISCONNECTED_TX_POOL_BYTES coverage (Greg Sanders)
Pull request description:
`DisconnectedBlockTransactions::LimitMemoryUsage()` has unit test coverage, but the default value end to end doesn't have coverage.
This test adds exercised coverage of memory limiting of the disconnect pool, and some basic behavior sanity checks.
Another test added is making sure chainlimits are being respected on reorg, and the expected transactions pruned.
Lastly, fix the existing test case which was using a deficient test via directly inducing reorgs with `invalidateblock`
ACKs for top commit:
maflcko:
re-ACK 84aa484d45🚋
TheCharlatan:
ACK 84aa484d45
Tree-SHA512: f5cdb9647fadc8eb30352ce38de44064103825e5358787dfccd6416fa8faf6ceea42552fe2250b37d56271a6c3898b3912e1c028652da122f5c99304aafddb64
Currently, the migration process creates a brand-new descriptor wallet with no
connection to the user's legacy wallet when the legacy wallet lacks key material
and contains only watch-only scripts. This behavior is not aligned with user
expectations. If the legacy wallet contains only watch-only scripts, the migration
process should only generate a watch-only wallet instead.
