wallet: Store transactions in a separate sqlite table #33034

achow101 commented at 10:52 PM on July 21, 2025: member

The wallet uses SQLite as a key-value store even though SQLite is a powerful relational database engine. This causes us numerous headaches due to the need to serialize multiple fields together, and since record read from the database during loading can come in any order.

Notably, for transactions, if we were able to load them in the transaction order assigned by the wallet, PRs like #27865 would be a bit simpler and easier to reason about.

This PR makes it possible for us to do that by storing transactions in a separate table. This table has a column for each field in CWalletTx, which lets us use a SQL statement like SELECT * FROM transactions ORDER BY pos which guarantees us the order in which the transactions are loaded into the wallet.

The eventual goal is to use SQLite as a relational database with tables that store our keys and other metadata that can reference each other so that the wallet doesn't just use the database as a storage mechanism. But that is still a work in progress.

This PR makes use of the last client opened features introduced in #32895 to determine when the old record style needs to be upgraded to the new transactions table. This should give us sufficient upgrade-downgrade-upgrade handling to not lose any funds.

Depends on #32763, #32895, #33033

DrahtBot added the label Wallet on Jul 21, 2025

DrahtBot commented at 10:52 PM on July 21, 2025: contributor

The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

Code Coverage & Benchmarks

For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/33034.

Reviews

See the guideline for information on the review process.

Type	Reviewers
Concept ACK	w0xlt, rkrux

If your review is incorrectly listed, please copy-paste <code></code> into the comment that the bot should ignore.

Conflicts

Reviewers, this pull request conflicts with the following ones:

#bitcoin-core/gui/872 (Menu action to export a watchonly wallet by achow101)
#35381 (wallet, test: optinrbf deprecation followups by rkrux)
#35302 (Silent Payments: Sending (take 2) by Eunovo)
#34909 (wallet, refactor: modularise wallet by extracting out legacy wallet migration by rkrux)
#34502 (wallet: remove most asserts of WALLET_FLAG_DESCRIPTORS flag by rkrux)
#33033 (wallet, sqlite: Encapsulate SQLite statements in a RAII class by achow101)
#32895 (wallet: Prepare for future upgrades by recording versions of last client to open and decrypt by achow101)
#32763 (wallet: Replace CWalletTx::mapValue and vOrderForm with explicit class members by achow101)
#32489 (wallet: Add exportwatchonlywallet RPC to export a watchonly version of a wallet by achow101)
#27865 (wallet: Track no-longer-spendable TXOs separately by achow101)
#25722 (refactor: Use util::Result class for wallet loading by ryanofsky)

If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

LLM Linter (✨ experimental)

Possible typos and grammar issues:

// InMempool needs to be written as InActive(abandoned=false) -> Inactive [“InActive” appears to be a misspelling of the transaction state name.]

2026-05-29 19:15:13

DrahtBot added the label CI failed on Jul 22, 2025

DrahtBot commented at 12:40 AM on July 22, 2025: contributor

🚧 At least one of the CI tasks failed. Task lint: https://github.com/bitcoin/bitcoin/runs/46429203587 LLM reason (✨ experimental): The CI failure is caused by Python lint errors due to unresolved 'self' references in test files.

<details><summary>Hints</summary>

Try to run the tests locally, according to the documentation. However, a CI failure may still happen due to a number of reasons, for example:

Possibly due to a silent merge conflict (the changes in this pull request being incompatible with the current code in the target branch). If so, make sure to rebase on the latest commit of the target branch.
A sanitizer issue, which can only be found by compiling with the sanitizer and running the affected test.
An intermittent issue.

Leave a comment here, if you need help tracking down a confusing failure.

</details>

achow101 force-pushed on Jul 22, 2025

DrahtBot removed the label CI failed on Jul 22, 2025

achow101 force-pushed on Jul 23, 2025

w0xlt commented at 9:15 PM on July 23, 2025: contributor

Concept ACK

achow101 force-pushed on Jul 24, 2025

DrahtBot added the label Needs rebase on Jul 29, 2025

achow101 force-pushed on Jul 29, 2025

DrahtBot removed the label Needs rebase on Jul 29, 2025

DrahtBot added the label Needs rebase on Aug 8, 2025

achow101 force-pushed on Aug 8, 2025

DrahtBot removed the label Needs rebase on Aug 8, 2025

DrahtBot added the label Needs rebase on Aug 13, 2025

achow101 force-pushed on Aug 14, 2025

DrahtBot removed the label Needs rebase on Aug 14, 2025

DrahtBot added the label Needs rebase on Aug 16, 2025

achow101 force-pushed on Aug 16, 2025

achow101 force-pushed on Aug 19, 2025

DrahtBot removed the label Needs rebase on Aug 19, 2025

DrahtBot added the label Needs rebase on Sep 23, 2025

achow101 force-pushed on Sep 23, 2025

DrahtBot removed the label Needs rebase on Sep 24, 2025

rkrux commented at 8:14 AM on October 9, 2025: contributor

Concept ACK a25557a

I have gone through the PR description and have skimmed over the commit messages as of now. At the outset, I find myself agreeing with the intent here. I am not aware of a strong enough reason to keep using SQLite as a KVS for all purposes, using its relational database properties for wallet transactions seem like a good start. I did notice some serialisation specifics of transaction properties in a couple previous PRs that I didn't fully understand why were required; it could be a lot cleaner and explicit in implementation if using SQLite as an RDBMS can avoid the need for doing those specific serialisations.

DrahtBot added the label Needs rebase on Dec 2, 2025

achow101 force-pushed on Dec 10, 2025

DrahtBot removed the label Needs rebase on Dec 10, 2025

DrahtBot added the label CI failed on Dec 10, 2025

DrahtBot commented at 8:15 PM on December 10, 2025: contributor

🚧 At least one of the CI tasks failed. Task test max 6 ancestor commits: https://github.com/bitcoin/bitcoin/actions/runs/20110042688/job/57704599161 LLM reason (✨ experimental): CMake build failed with exit code 2 during project compilation (gmake reported an error).

<details><summary>Hints</summary>

Try to run the tests locally, according to the documentation. However, a CI failure may still happen due to a number of reasons, for example:

Possibly due to a silent merge conflict (the changes in this pull request being incompatible with the current code in the target branch). If so, make sure to rebase on the latest commit of the target branch.
A sanitizer issue, which can only be found by compiling with the sanitizer and running the affected test.
An intermittent issue.

Leave a comment here, if you need help tracking down a confusing failure.

</details>

achow101 force-pushed on Dec 10, 2025

DrahtBot removed the label CI failed on Dec 11, 2025

DrahtBot added the label Needs rebase on Dec 17, 2025

achow101 force-pushed on Dec 22, 2025

DrahtBot removed the label Needs rebase on Dec 22, 2025

DrahtBot added the label CI failed on Dec 23, 2025

achow101 force-pushed on Jan 3, 2026

DrahtBot removed the label CI failed on Jan 3, 2026

DrahtBot added the label Needs rebase on Jan 19, 2026

achow101 force-pushed on Jan 19, 2026

DrahtBot added the label CI failed on Jan 19, 2026

DrahtBot commented at 7:49 PM on January 19, 2026: contributor

🚧 At least one of the CI tasks failed. Task test max 6 ancestor commits: https://github.com/bitcoin/bitcoin/actions/runs/21148471093/job/60819287525 LLM reason (✨ experimental): Build failed: the CI cmake/make run exited with non-zero status (gmake: all target) after linking bitcoinkernel, indicating a generic build error.

<details><summary>Hints</summary>

Try to run the tests locally, according to the documentation. However, a CI failure may still happen due to a number of reasons, for example:

Possibly due to a silent merge conflict (the changes in this pull request being incompatible with the current code in the target branch). If so, make sure to rebase on the latest commit of the target branch.
A sanitizer issue, which can only be found by compiling with the sanitizer and running the affected test.
An intermittent issue.

Leave a comment here, if you need help tracking down a confusing failure.

</details>

DrahtBot removed the label Needs rebase on Jan 19, 2026

achow101 force-pushed on Jan 19, 2026

achow101 force-pushed on Jan 26, 2026

DrahtBot added the label Needs rebase on Feb 4, 2026

achow101 force-pushed on Feb 4, 2026

DrahtBot removed the label Needs rebase on Feb 4, 2026

DrahtBot added the label Needs rebase on Mar 2, 2026

achow101 force-pushed on Mar 5, 2026

DrahtBot removed the label Needs rebase on Mar 5, 2026

DrahtBot added the label Needs rebase on Mar 11, 2026

achow101 force-pushed on Mar 14, 2026

DrahtBot removed the label Needs rebase on Mar 14, 2026

achow101 force-pushed on Apr 1, 2026

achow101 force-pushed on Apr 2, 2026

DrahtBot added the label Needs rebase on Apr 19, 2026

achow101 force-pushed on Apr 20, 2026

DrahtBot removed the label Needs rebase on Apr 20, 2026

DrahtBot removed the label CI failed on Apr 20, 2026

sedited referenced this in commit db98e357d3 on May 2, 2026

DrahtBot added the label Needs rebase on May 2, 2026

achow101 force-pushed on May 19, 2026

DrahtBot removed the label Needs rebase on May 19, 2026

DrahtBot added the label Needs rebase on May 26, 2026

achow101 force-pushed on May 26, 2026

DrahtBot removed the label Needs rebase on May 26, 2026

DrahtBot added the label CI failed on May 26, 2026

DrahtBot removed the label CI failed on May 27, 2026

DrahtBot added the label Needs rebase on May 28, 2026

DrahtBot commented at 10:44 PM on May 28, 2026: contributor

🐙 This pull request conflicts with the target branch and needs rebase.

wallet: Pass replaces_txid to CommitTransaction outside of mapValue

Instead of updating mapValue with the "replaces_txid" value and passing
the updated mapValue to CommitTransaction, pass the replaces_txid
directly to CommitTransaction which updates mapValue as necessary.

This is prepration for removing mapValue.

b3dce17d02

wallet: Pass comment and comment_to to CommitTransaction

Instead of passing these by setting them in mapValue, pass them directly
to CommitTransaction.

This is preparation for removing mapValue.

bc43765841

wallet: Drop mapValue from CommitTransaction

The values previously passed in mapValue are now parameters to
CommitTransaction so there is no need for mapValue to be passed.

77ea833c50

wallet: Make CWalletTx "from" and "message" member variables

Instead of storing "from" and "message" inside of mapValue, store these
explicitly as members of CWalletTx.

bd2873722c

wallet: Make CWalletTx "comment" and "to" member variables

Instead of storing "comment" and "to" inside of mapValue, store these
expliclty as members of CWalletTx.

7367002f72

wallet: Make CWalletTx "replaces_txid" and "replaced_by_txid" member variables

Instead of storing "replaces_txid" and "replaced_by_txid" as strings inside of
mapValue, store these expliclty as members of CWalletTx.

e4b8adc5e8

wallet: Drop mapValue from CWalletTx

It doesn't make sense to be storing relevant metadata variables inside
of a string map in CWalletTx. All of the fields have been pulled out
into separate members, so there is no need for mapValue to stick around.

2aa356a62a

wallet: Drop vOrderForm from CommitTransaction

This parameter is only used to pass in the "Messages" from the GUI.
Instead of making it opaque by putting those into vOrderForm, use a
specific dedicated parameter for providing the messages.

5f91350ee3

wallet: Replace CWalletTx's vOrderForm with specific fields

vOrderForm contained 2 kinds of strings: BIP 21 messages, and BIP 70
Payment Requests. Instead of having both inside of a single vOrderForm
field that is opaque, split them into separate std::vector<std::string>
to contain this metadata.

de2cb8c2db

walletdb: Decouple last client record from CLIENT_VERSION

Instead of tying the last client record with the node version, introduce
a separate wallet client version that will be increased as necessary
when new automatic upgrades are introduced.

dde8b25c22

wallet: Introduce LastClientFeatures flags and LAST_OPENED_FEATURES record

LastClientFeatures are feature flags indicating what automatic upgrade
features supported by the last client that opened the wallet.

The LAST_OPENED_FEATURES record stores these flags and must be
set to match the exact features supported by the client that opens a
wallet.

c7f66e0d77

wallet: Record the supported features of the last client to decrypt a wallet 42377223cc

wallet: Set last opened and decrypted features during migration

Set these flags to avoid the automatic upgrade after migrating.

1b3812707b

sqlite: Add SQLiteStatement RAII class

This class will be used to encapsulate a sqlite3_stmt

86b180a73e

sqlite: Make Column template function 291070dad4

sqlite: Refactor ReadPragmaInteger to use SQLiteStatement 8efdfd7ab8

sqlite: Use SQLiteStatement in PRAGMA integrity_check dae205dc1c

sqlite: Use SQLiteStatement in check_main_stmt 6a1ae3c708

sqlite: Have SQLiteCursor store SQLiteStatement fb70de7311

sqlite: Replace remaining sqlite3_stmt usage with SQLiteStatement 8835b8de3a

sqlite: Refactor common writing code from WriteKey and ExecStatement

WriteKey and ExecStatement use the same code for the actual execution of
the statement. This is refactored into a separate function, also called
ExecStatement, and the original ExecStatement renamed to
ExecEraseStatement as it is only used by the erase functions.

7836fa7107

sqlite: Inline BindBlobToStatement and SpanFromBlob a87b95c6d9

sqlite: Construct SQLiteStatements when needed

Instead of constructing all SQLiteStatements when SQLiteBatch is
constructed, construct them once when they are needed before each read,
write, or erase operation. Once constructed, the statement will persist
for the lifetime of the SQLiteBatch to be reused across multiple
statements.

9183e10e2b

util: Add span constructor to (W)Txid 06e1eb94ea

sqlite: Make SQLiteStatement::Bind a template function

We will need to bind data types other than blob

1d70ab83fa

sqlite: Make SQLiteDatabase::Column an std::optional

To handle columns containing NULL values, Column needs to return some
value representing NULL, so make it a std::optional.

92d5b0611b

sqlite: Add additional blob types to SQLiteStatement::Column

Not all blob data types fit in ColumnBlob, so we need additional
template type requirements to match those.

f45af415a9

wallet: Add Un/Serialize to TxState structs 45f333890d

wallet: Add new serialization format functions for TxState 0ee6abb579

sqlite: Create a 'transactions' table if it does not exist 7a9ef4e84e

sqlite: Add functions and statements for transactions table b55269a394

sqlite: Iterate the transactions table with SQLiteCursor 70230035c0

walletdb: Add functions to modify transactions table ee889415a4

wallet: Also write to the new transactions table 7751a225bb

wallet: Perform automatic upgrade to using the transactions table 39b8b16e3e

walletdb: Load from transactions table and use original tx for upgrade

When loading a wallet, always load from the transactions table, except
when loading a legacy wallet for migration.

The original 'tx' records are only used to upgrade to the transactions
table if an upgraded is necessary.

aa00219a57

achow101 force-pushed on May 29, 2026