af76664b12 test: Test migration of a solvable script with no privkeys (Ava Chow)
17f01b0795 test: Test migration of taproot output scripts (Ava Chow)
1eb9a2a39f test: Test migration of miniscript in legacy wallets (Ava Chow)
e8c3efc7d8 wallet migration: Determine Solvables with CanProvide (Ava Chow)
fa1b7cd6e2 migration: Skip descriptors which do not parse (Ava Chow)
440ea1ab63 legacy spkm: use IsMine() to extract watched output scripts (Ava Chow)
b777e84cd7 legacy spkm: Move CanProvide to LegacyDataSPKM (Ava Chow)
b1ab927bbf tests: Test migration of additional P2WSH scripts (Ava Chow)
c39b3cfcd1 test: Extra verification that migratewallet migrates (Ava Chow)
Pull request description:
The legacy wallet `IsMine()` is essentially a black box that would tell us whether the wallet is watching an output script. In order to migrate legacy wallets to descriptor wallets, we need to be able to compute all of the output scripts that a legacy wallet would watch. The original approach for this was to understand `IsMine()` and write a function which would be its inverse. This was partially done in the original migration code, and attempted to be completed in #30328. However, further analysis of `IsMine()` has continued to reveal additional edge cases which make writing an inverse function increasingly difficult to verify correctness.
This PR instead changes migration to utilize `IsMine()` to produce the output scripts by first computing a superset of all of the output scripts that `IsMine()` would watch and testing each script against `IsMine()` to filter for the ones that actually are watched. The superset is constructed by computing all possible output scripts for the keys and scripts in the wallet - for keys, every key could be a P2PK, P2PKH, P2WPKH, and P2SH-P2WPKH; for scripts, every script could be an output script, the redeemScript of a P2SH, the witnessScript of a P2WSH, and the witnessScript of a P2SH-P2WSH.
Additionally, the legacy wallet can contain scripts that are redeemScripts and witnessScripts, while not watching for any output script utilizing that script. These are known as solvable scripts and are migrated to a separate "solvables" wallet. The previous approach to identifying these solvables was similar to identifying output scripts - finding known solvable conditions and computing the scripts. However, this also can miss scripts, so the solvables are now identified in a manner similar to the output scripts but using the function `CanProvide()`. Using the same superset as before, all output scripts which are `ISMINE_NO` are put through `CanProvide()` which will perform a dummy signing and then a key lookup to determine whether the legacy wallet could provide any solving data for the output script. The scripts that pass will have their descriptors inferred and the script included in the solvables wallet.
The main downside of this approach is that `IsMine()` and `CanProvide()` can no longer be deleted. They will need to be refactored to be migration only code instead in #28710.
Lastly, I've added 2 test cases for the edge cases that prompted this change of approach. In particular, miniscript witnessScripts and `rawtr()` output scripts are solvable and signable in a legacy wallet, although never `ISMINE_SPENDABLE`.
ACKs for top commit:
sipa:
Code review ACK af76664b12; I did not review the tests in detail.
brunoerg:
code review ACK af76664b12
rkrux:
ACK af76664b12
Tree-SHA512: 7f58a90de6f38fe9801fb6c2a520627072c8d66358652ad0872ff59deb678a82664b99babcfd874288bebcb1487d099a77821f03ae063c2b4cbf2d316e77d141
LegacySPKM would determine whether it could provide any script data to a
transaction through the use of the CanProvide function. Instead of
partially reversing signing logic to figure out the output scripts of
solvable things, we use the same candidate set approach in
GetScriptPubKeys() and instead filter the candidate set first for
things that are ISMINE_NO, and second with CanProvide(). This should
give a more accurate solvables wallet.
InferDescriptors can sometimes make descriptors which are actually
invalid and cannot be parsed. Detect and skip such descriptors by doing
a Parse() check before adding the descriptor to the wallet.
Instead of (partially) trying to reverse IsMine() to get the output
scripts that a LegacySPKM would track, we can preserve it in migration
only code and utilize it to get an accurate set of output scripts.
This is accomplished by computing a set of output script candidates from
map(Crypted)Keys, mapScripts, and setWatchOnly. This candidate set is an
upper bound on the scripts tracked by the wallet. Then IsMine() is used
to filter to the exact output scripts that LegacySPKM would track.
By changing GetScriptPubKeys() this way, we can avoid complexities in
reversing IsMine() and get a more complete set of output scripts.
f6a6d91205 test: add check for getting SigningProvider for a CPubKey (Sebastian Falbesoner)
62a95f5af9 test: refactor: move `CreateDescriptor` helper to wallet test util module (Sebastian Falbesoner)
493656763f desc spkm: Return SigningProvider only if we have the privkey (Ava Chow)
Pull request description:
If we know about a pubkey that's in our descriptor, but we don't have the private key, don't return a SigningProvider for that pubkey.
This is specifically an issue for Taproot outputs that use the H point as the resulting PSBTs may end up containing irrelevant information because the H point was detected as a pubkey each unrelated descriptor knew about.
Split from #29675
ACKs for top commit:
fjahr:
ACK f6a6d91205
theStack:
re-ACK f6a6d91205
furszy:
utACK f6a6d91205. Only reviewed the actual change in detail, not the test commit.
Tree-SHA512: 30a196e611a0c5d9ebe5baf6d896caaa6af66f1615463dbb0c31e52604d53cf342922bb9967b3c697b47083d76b0485c77a5f545bd6381247c8bc44321c70f97
69e95c2b4f tests: Test cleanup of mkeys from wallets without privkeys (Andrew Chow)
2b9279b50a wallet: Remove unused encryption keys from watchonly wallets (Andrew Chow)
813a16a463 wallet: Add HasCryptedKeys (Andrew Chow)
Pull request description:
An earlier version allowed users to create watchonly wallets (wallets without private keys) that were "encrypted". Such wallets would have a stored encryption keys, but nothing would actually be encrypted with them. This can cause unexpected behavior such as https://github.com/bitcoin-core/gui/issues/772.
We can detect such wallets as they will have the disable private keys flag set, no encrypted keys, and encryption keys. For such wallets, we can remove those encryption keys thereby avoiding any issues that may result from this unexpected situation.
ACKs for top commit:
sipa:
utACK 69e95c2b4f.
laanwj:
Code review re-ACK 69e95c2b4f
furszy:
Code review ACK 69e95c2b4f
Tree-SHA512: 901932cd709c57e66c598f011f0105a243b5a8b539db2ef3fcf370dca4cf35ae09bc1110e8fca8353be470f159468855a4dd96b99bc9c1112adc86ccc50e1b9d
Non-HD keys in legacy wallets without a HD seed ID were being migrated
to separate pk(), pkh(), sh(wpkh()), and wpkh() descriptors for each key.
These could be more compactly represented as combo() descriptors, so
migration should make combo() for them.
It is possible that existing non-HD wallets that were migrated, or
wallets that started blank and had private keys imported into them have
run into this issue. However, as the 4 descriptors produce the same output
scripts as the single combo(), so any previously migrated wallets are
not missing any output scripts. The only observable difference should be
performance related, and the wallet size on disk.
If we know about a pubkey that's in our descriptor, but we don't have
the private key, don't return a SigningProvider for that pubkey.
This is specifically an issue for Taproot outputs that use the H point
as the resulting PSBTs may end up containing irrelevant information
because the H point was detected as a pubkey each unrelated descriptor
knew about.
