Compare commits

..

40 Commits

Author SHA1 Message Date
copilot-swe-agent[bot]
6b57c273ef Remove incorrect assertion about result size matching input size
Co-authored-by: ximinez <6757714+ximinez@users.noreply.github.com>
2026-02-03 21:02:49 +00:00
copilot-swe-agent[bot]
62d65c2f4f Initial plan 2026-02-03 21:00:50 +00:00
Bart
44959524ca Add assert 2026-02-03 11:47:21 -05:00
Bart
cd3feb979a Fix pointer access 2026-02-03 11:19:17 -05:00
Bart
606ffae0e6 Merge branch 'develop' into bthomee/disable-cache 2026-02-03 11:11:35 -05:00
Bart
6eb6df9cd9 Remove const 2026-02-03 11:07:04 -05:00
Mayukha Vadari
9f6ec0adf7 docs: Update API changelog, add APIv2+APIv3 version documentation (#6308)
This change cleans up the `API-CHANGELOG.md` file. It moves the version-specific documentation to other files and fleshes out the changelog with all the API-related changes in each version.
2026-02-03 11:06:35 -05:00
Bart
00e7a04a9f fix: Restore config changes that broke standalone mode (#6301)
When support was added for `xrpld.cfg` in addition to `rippled.cfg` in https://github.com/XRPLF/rippled/pull/6098, as part of an effort to rename occurrences of ripple(d) to xrpl(d), the clearing and creation of the data directory were modified for what, at the time, seemed to result in an equivalent code flow. This has turned out to not be true, which is why this change restores two modifications to `Config.cpp` that currently break running the binary in standalone mode.
2026-02-03 11:06:35 -05:00
Ed Hennis
e5ce4f12ae chore: Add upper-case match for ARM64 in CompilationEnv (#6315) 2026-02-03 11:06:34 -05:00
Ayaz Salikhov
a4bccb9821 ci: Update hashes of XRPLF/actions (#6316)
This updates the hashes of all XRPLF/actions to their latest versions.
2026-02-03 11:06:34 -05:00
Ayaz Salikhov
d7d84e8033 chore: Format all cmake files without comments (#6294) 2026-02-03 11:06:34 -05:00
Ayaz Salikhov
96e645a63b chore: Add cmake-format pre-commit hook (#6279)
This change adds `cmake-format` as. a pre-commit hook. The style file closely matches that in Clio, and they will be made to be equivalent over time. For now, some files have been excluded, as those need some manual adjustments, which will be done in future changes.
2026-02-03 11:06:34 -05:00
Bart
4bdba71b05 Remove unused sweep 2026-02-03 11:06:34 -05:00
Bart
735f9f4118 Apply suggestions from code review
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-02-03 11:06:21 -05:00
Bart
aeb7934c0c Update src/libxrpl/nodestore/DatabaseNodeImp.cpp
Co-authored-by: Ed Hennis <ed@ripple.com>
2026-02-03 10:22:36 -05:00
Vito Tumas
b814a09a08 chore: Add .zed editor config directory to .gitignore (#6317)
This change adds the project configuration directory to `.gitignore` for the `zed` editor. 

As per the [documentation](https://zed.dev/docs/remote-development?highlight=.zed#zed-settings), the project configuration files are stored in the `.zed` directory at the project root dir.
2026-02-03 10:13:10 -05:00
Mayukha Vadari
6d369e0f02 docs: Update API changelog, add APIv2+APIv3 version documentation (#6308)
This change cleans up the `API-CHANGELOG.md` file. It moves the version-specific documentation to other files and fleshes out the changelog with all the API-related changes in each version.
2026-02-03 02:12:26 +00:00
Bart
b182430178 fix: Restore config changes that broke standalone mode (#6301)
When support was added for `xrpld.cfg` in addition to `rippled.cfg` in https://github.com/XRPLF/rippled/pull/6098, as part of an effort to rename occurrences of ripple(d) to xrpl(d), the clearing and creation of the data directory were modified for what, at the time, seemed to result in an equivalent code flow. This has turned out to not be true, which is why this change restores two modifications to `Config.cpp` that currently break running the binary in standalone mode.
2026-02-03 01:15:56 +00:00
Ed Hennis
fe31cdc9f6 chore: Add upper-case match for ARM64 in CompilationEnv (#6315) 2026-02-02 23:57:10 +00:00
Ayaz Salikhov
ff4520cc45 ci: Update hashes of XRPLF/actions (#6316)
This updates the hashes of all XRPLF/actions to their latest versions.
2026-02-02 19:37:06 +00:00
Bart
9639b79155 Copilot feedback 2026-02-02 10:59:45 -05:00
Bart
eabd485927 Merge branch 'develop' into bthomee/disable-cache 2026-01-29 09:28:48 +00:00
Bart
f3ea3e9646 Merge branch 'develop' into bthomee/disable-cache 2025-11-22 11:06:58 -05:00
Bart Thomee
4306d9ccc3 Restore freshening caches of tree node cache 2025-09-17 12:17:35 -04:00
Bart Thomee
1a4d9732ca Merge branch 'develop' into bthomee/disable-cache 2025-09-17 11:43:06 -04:00
Bart
aad6edb6b1 Merge branch 'develop' into bthomee/disable-cache 2025-08-13 08:03:40 -04:00
Bart
a4a1c4eecf Merge branch 'develop' into bthomee/disable-cache 2025-07-03 15:43:50 -04:00
Bart
fca6a8768f Merge branch 'develop' into bthomee/disable-cache 2025-06-02 12:02:43 -04:00
Bart
d96c4164b9 Merge branch 'develop' into bthomee/disable-cache 2025-05-22 09:18:07 -04:00
Bart Thomee
965fc75e8a Reserve vector size 2025-05-20 10:07:12 -04:00
Bart Thomee
2fa1c711d3 Removed unused config values 2025-05-20 09:50:13 -04:00
Bart Thomee
4650e7d2c6 Removed unused caches from SHAMapStoreImp 2025-05-20 09:49:55 -04:00
Bart Thomee
a213127852 Remove cache from SHAMapStoreImp 2025-05-19 16:59:43 -04:00
Bart Thomee
6e7537dada Remove cache from DatabaseNodeImp 2025-05-19 16:51:32 -04:00
Bart Thomee
0777f7c64b Merge branch 'develop' into bthomee/disable-cache 2025-05-19 16:37:11 -04:00
Bart Thomee
39bfcaf95c Merge branch 'develop' into bthomee/disable-cache 2025-05-17 18:26:07 -04:00
Bart Thomee
61c9a19868 Merge branch 'develop' into bthomee/disable-cache 2025-05-07 11:02:43 -04:00
Bart Thomee
d01851bc5a Only disable the database cache 2025-04-01 13:24:18 -04:00
Bart Thomee
d1703842e7 Fully disable cache 2025-04-01 11:41:20 -04:00
Bart Thomee
8d31b1739d TEST: Disable tagged cache to measure performance 2025-03-28 13:21:19 -04:00
25 changed files with 268 additions and 454 deletions

View File

@@ -11,7 +11,7 @@ on:
jobs:
# Call the workflow in the XRPLF/actions repo that runs the pre-commit hooks.
run-hooks:
uses: XRPLF/actions/.github/workflows/pre-commit.yml@282890f46d6921249d5659dd38babcb0bd8aef48
uses: XRPLF/actions/.github/workflows/pre-commit.yml@320be44621ca2a080f05aeb15817c44b84518108
with:
runs_on: ubuntu-latest
container: '{ "image": "ghcr.io/xrplf/ci/tools-rippled-pre-commit:sha-ab4d1f0" }'

View File

@@ -36,7 +36,7 @@ jobs:
uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
- name: Get number of processors
uses: XRPLF/actions/get-nproc@2ece4ec6ab7de266859a6f053571425b2bd684b6
uses: XRPLF/actions/get-nproc@cf0433aa74563aead044a1e395610c96d65a37cf
id: nproc
with:
subtract: ${{ env.NPROC_SUBTRACT }}

View File

@@ -101,13 +101,13 @@ jobs:
steps:
- name: Cleanup workspace (macOS and Windows)
if: ${{ runner.os == 'macOS' || runner.os == 'Windows' }}
uses: XRPLF/actions/cleanup-workspace@2ece4ec6ab7de266859a6f053571425b2bd684b6
uses: XRPLF/actions/cleanup-workspace@cf0433aa74563aead044a1e395610c96d65a37cf
- name: Checkout repository
uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
- name: Prepare runner
uses: XRPLF/actions/prepare-runner@f05cab7b8541eee6473aa42beb9d2fe35608a190
uses: XRPLF/actions/prepare-runner@2cbf481018d930656e9276fcc20dc0e3a0be5b6d
with:
enable_ccache: ${{ inputs.ccache_enabled }}
@@ -119,7 +119,7 @@ jobs:
uses: ./.github/actions/print-env
- name: Get number of processors
uses: XRPLF/actions/get-nproc@2ece4ec6ab7de266859a6f053571425b2bd684b6
uses: XRPLF/actions/get-nproc@cf0433aa74563aead044a1e395610c96d65a37cf
id: nproc
with:
subtract: ${{ inputs.nproc_subtract }}

View File

@@ -64,13 +64,13 @@ jobs:
steps:
- name: Cleanup workspace (macOS and Windows)
if: ${{ runner.os == 'macOS' || runner.os == 'Windows' }}
uses: XRPLF/actions/cleanup-workspace@2ece4ec6ab7de266859a6f053571425b2bd684b6
uses: XRPLF/actions/cleanup-workspace@cf0433aa74563aead044a1e395610c96d65a37cf
- name: Checkout repository
uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
- name: Prepare runner
uses: XRPLF/actions/prepare-runner@f05cab7b8541eee6473aa42beb9d2fe35608a190
uses: XRPLF/actions/prepare-runner@2cbf481018d930656e9276fcc20dc0e3a0be5b6d
with:
enable_ccache: false
@@ -78,7 +78,7 @@ jobs:
uses: ./.github/actions/print-env
- name: Get number of processors
uses: XRPLF/actions/get-nproc@2ece4ec6ab7de266859a6f053571425b2bd684b6
uses: XRPLF/actions/get-nproc@cf0433aa74563aead044a1e395610c96d65a37cf
id: nproc
with:
subtract: ${{ env.NPROC_SUBTRACT }}

3
.gitignore vendored
View File

@@ -64,6 +64,9 @@ DerivedData
/.vs/
/.vscode/
# zed IDE.
/.zed/
# AI tools.
/.augment
/.claude

View File

@@ -6,96 +6,85 @@ For info about how [API versioning](https://xrpl.org/request-formatting.html#api
The API version controls the API behavior you see. This includes what properties you see in responses, what parameters you're permitted to send in requests, and so on. You specify the API version in each of your requests. When a breaking change is introduced to the `rippled` API, a new version is released. To avoid breaking your code, you should set (or increase) your version when you're ready to upgrade.
The [commandline](https://xrpl.org/docs/references/http-websocket-apis/api-conventions/request-formatting/#commandline-format) always uses the latest API version. The command line is intended for ad-hoc usage by humans, not programs or automated scripts. The command line is not meant for use in production code.
For a log of breaking changes, see the **API Version [number]** headings. In general, breaking changes are associated with a particular API Version number. For non-breaking changes, scroll to the **XRP Ledger version [x.y.z]** headings. Non-breaking changes are associated with a particular XRP Ledger (`rippled`) release.
## API Version 3 (Beta)
API version 3 is currently a beta API. It requires enabling `[beta_rpc_api]` in the rippled configuration to use. See [API-VERSION-3.md](API-VERSION-3.md) for the full list of changes in API version 3.
## API Version 2
API version 2 is available in `rippled` version 2.0.0 and later. To use this API, clients specify `"api_version" : 2` in each request.
#### Removed methods
In API version 2, the following deprecated methods are no longer available: (https://github.com/XRPLF/rippled/pull/4759)
- `tx_history` - Instead, use other methods such as `account_tx` or `ledger` with the `transactions` field set to `true`.
- `ledger_header` - Instead, use the `ledger` method.
#### Modifications to JSON transaction element in V2
In API version 2, JSON elements for transaction output have been changed and made consistent for all methods which output transactions. (https://github.com/XRPLF/rippled/pull/4775)
This helps to unify the JSON serialization format of transactions. (https://github.com/XRPLF/clio/issues/722, https://github.com/XRPLF/rippled/issues/4727)
- JSON transaction element is named `tx_json`
- Binary transaction element is named `tx_blob`
- JSON transaction metadata element is named `meta`
- Binary transaction metadata element is named `meta_blob`
Additionally, these elements are now consistently available next to `tx_json` (i.e. sibling elements), where possible:
- `hash` - Transaction ID. This data was stored inside transaction output in API version 1, but in API version 2 is a sibling element.
- `ledger_index` - Ledger index (only set on validated ledgers)
- `ledger_hash` - Ledger hash (only set on closed or validated ledgers)
- `close_time_iso` - Ledger close time expressed in ISO 8601 time format (only set on validated ledgers)
- `validated` - Bool element set to `true` if the transaction is in a validated ledger, otherwise `false`
This change affects the following methods:
- `tx` - Transaction data moved into element `tx_json` (was inline inside `result`) or, if binary output was requested, moved from `tx` to `tx_blob`. Renamed binary transaction metadata element (if it was requested) from `meta` to `meta_blob`. Changed location of `hash` and added new elements
- `account_tx` - Renamed transaction element from `tx` to `tx_json`. Renamed binary transaction metadata element (if it was requested) from `meta` to `meta_blob`. Changed location of `hash` and added new elements
- `transaction_entry` - Renamed transaction metadata element from `metadata` to `meta`. Changed location of `hash` and added new elements
- `subscribe` - Renamed transaction element from `transaction` to `tx_json`. Changed location of `hash` and added new elements
- `sign`, `sign_for`, `submit` and `submit_multisigned` - Changed location of `hash` element.
#### Modification to `Payment` transaction JSON schema
When reading Payments, the `Amount` field should generally **not** be used. Instead, use [delivered_amount](https://xrpl.org/partial-payments.html#the-delivered_amount-field) to see the amount that the Payment delivered. To clarify its meaning, the `Amount` field is being renamed to `DeliverMax`. (https://github.com/XRPLF/rippled/pull/4733)
- In `Payment` transaction type, JSON RPC field `Amount` is renamed to `DeliverMax`. To enable smooth client transition, `Amount` is still handled, as described below: (https://github.com/XRPLF/rippled/pull/4733)
- On JSON RPC input (e.g. `submit_multisigned` etc. methods), `Amount` is recognized as an alias to `DeliverMax` for both API version 1 and version 2 clients.
- On JSON RPC input, submitting both `Amount` and `DeliverMax` fields is allowed _only_ if they are identical; otherwise such input is rejected with `rpcINVALID_PARAMS` error.
- On JSON RPC output (e.g. `subscribe`, `account_tx` etc. methods), `DeliverMax` is present in both API version 1 and version 2.
- On JSON RPC output, `Amount` is only present in API version 1 and _not_ in version 2.
#### Modifications to account_info response
- `signer_lists` is returned in the root of the response. (In API version 1, it was nested under `account_data`.) (https://github.com/XRPLF/rippled/pull/3770)
- When using an invalid `signer_lists` value, the API now returns an "invalidParams" error. (https://github.com/XRPLF/rippled/pull/4585)
- (`signer_lists` must be a boolean. In API version 1, strings were accepted and may return a normal response - i.e. as if `signer_lists` were `true`.)
#### Modifications to [account_tx](https://xrpl.org/account_tx.html#account_tx) response
- Using `ledger_index_min`, `ledger_index_max`, and `ledger_index` returns `invalidParams` because if you use `ledger_index_min` or `ledger_index_max`, then it does not make sense to also specify `ledger_index`. In API version 1, no error was returned. (https://github.com/XRPLF/rippled/pull/4571)
- The same applies for `ledger_index_min`, `ledger_index_max`, and `ledger_hash`. (https://github.com/XRPLF/rippled/issues/4545#issuecomment-1565065579)
- Using a `ledger_index_min` or `ledger_index_max` beyond the range of ledgers that the server has:
- returns `lgrIdxMalformed` in API version 2. Previously, in API version 1, no error was returned. (https://github.com/XRPLF/rippled/issues/4288)
- Attempting to use a non-boolean value (such as a string) for the `binary` or `forward` parameters returns `invalidParams` (`rpcINVALID_PARAMS`). Previously, in API version 1, no error was returned. (https://github.com/XRPLF/rippled/pull/4620)
#### Modifications to [noripple_check](https://xrpl.org/noripple_check.html#noripple_check) response
- Attempting to use a non-boolean value (such as a string) for the `transactions` parameter returns `invalidParams` (`rpcINVALID_PARAMS`). Previously, in API version 1, no error was returned. (https://github.com/XRPLF/rippled/pull/4620)
API version 2 is available in `rippled` version 2.0.0 and later. See [API-VERSION-2.md](API-VERSION-2.md) for the full list of changes in API version 2.
## API Version 1
This version is supported by all `rippled` versions. For WebSocket and HTTP JSON-RPC requests, it is currently the default API version used when no `api_version` is specified.
The [commandline](https://xrpl.org/docs/references/http-websocket-apis/api-conventions/request-formatting/#commandline-format) always uses the latest API version. The command line is intended for ad-hoc usage by humans, not programs or automated scripts. The command line is not meant for use in production code.
## XRP Ledger server version 3.1.0
### Inconsistency: server_info - network_id
[Version 3.1.0](https://github.com/XRPLF/rippled/releases/tag/3.1.0) was released on Jan 27, 2026.
The `network_id` field was added in the `server_info` response in version 1.5.0 (2019), but it is not returned in [reporting mode](https://xrpl.org/rippled-server-modes.html#reporting-mode). However, use of reporting mode is now discouraged, in favor of using [Clio](https://github.com/XRPLF/clio) instead.
### Additions in 3.1.0
## Unreleased Changes
- `vault_info`: New RPC method to retrieve information about a specific vault (part of XLS-66 Lending Protocol). ([#6156](https://github.com/XRPLF/rippled/pull/6156))
### Additions and bugfixes
## XRP Ledger server version 3.0.0
- `submit`: Augmented response fields (`accepted`, `applied`, `broadcast`, `queued`, `kept`, `account_sequence_next`, `account_sequence_available`, `open_ledger_cost`, `validated_ledger_index`) are now included in sign-and-submit mode. Previously, these fields were only returned when submitting a binary transaction blob. ([#6304](https://github.com/XRPLF/rippled/pull/6304))
[Version 3.0.0](https://github.com/XRPLF/rippled/releases/tag/3.0.0) was released on Dec 9, 2025.
### Additions in 3.0.0
- `ledger_entry`: Supports all ledger entry types with dedicated parsers. ([#5237](https://github.com/XRPLF/rippled/pull/5237))
- `ledger_entry`: New error codes `entryNotFound` and `unexpectedLedgerType` for more specific error handling. ([#5237](https://github.com/XRPLF/rippled/pull/5237))
- `ledger_entry`: Improved error messages with more context (e.g., specifying which field is invalid or missing). ([#5237](https://github.com/XRPLF/rippled/pull/5237))
- `ledger_entry`: Assorted bug fixes in RPC processing. ([#5237](https://github.com/XRPLF/rippled/pull/5237))
- `simulate`: Supports additional metadata in the response. ([#5754](https://github.com/XRPLF/rippled/pull/5754))
## XRP Ledger server version 2.6.2
[Version 2.6.2](https://github.com/XRPLF/rippled/releases/tag/2.6.2) was released on Nov 19, 2025.
This release contains bug fixes only and no API changes.
## XRP Ledger server version 2.6.1
[Version 2.6.1](https://github.com/XRPLF/rippled/releases/tag/2.6.1) was released on Sep 30, 2025.
This release contains bug fixes only and no API changes.
## XRP Ledger server version 2.6.0
[Version 2.6.0](https://github.com/XRPLF/rippled/releases/tag/2.6.0) was released on Aug 27, 2025.
### Additions in 2.6.0
- `account_info`: Added `allowTrustLineLocking` flag in response. ([#5525](https://github.com/XRPLF/rippled/pull/5525))
- `ledger`: Removed the type filter from the RPC command. ([#4934](https://github.com/XRPLF/rippled/pull/4934))
- `subscribe` (`validations` stream): `network_id` is now included. ([#5579](https://github.com/XRPLF/rippled/pull/5579))
- `subscribe` (`transactions` stream): `nftoken_id`, `nftoken_ids`, and `offer_id` are now included in transaction metadata. ([#5230](https://github.com/XRPLF/rippled/pull/5230))
## XRP Ledger server version 2.5.1
[Version 2.5.1](https://github.com/XRPLF/rippled/releases/tag/2.5.1) was released on Sep 17, 2025.
This release contains bug fixes only and no API changes.
## XRP Ledger server version 2.5.0
As of 2025-04-04, version 2.5.0 is in development. You can use a pre-release version by building from source or [using the `nightly` package](https://xrpl.org/docs/infrastructure/installation/install-rippled-on-ubuntu).
[Version 2.5.0](https://github.com/XRPLF/rippled/releases/tag/2.5.0) was released on Jun 24, 2025.
### Additions and bugfixes in 2.5.0
- `channel_authorize`: If `signing_support` is not enabled in the config, the RPC is disabled.
- `tx`: Added `ctid` field to the response and improved error handling. ([#4738](https://github.com/XRPLF/rippled/pull/4738))
- `ledger_entry`: Improved error messages in `permissioned_domain`. ([#5344](https://github.com/XRPLF/rippled/pull/5344))
- `simulate`: Improved multi-sign usage. ([#5479](https://github.com/XRPLF/rippled/pull/5479))
- `channel_authorize`: If `signing_support` is not enabled in the config, the RPC is disabled. ([#5385](https://github.com/XRPLF/rippled/pull/5385))
- `subscribe` (admin): Removed webhook queue limit to prevent dropping notifications; reduced HTTP timeout from 10 minutes to 30 seconds. ([#5163](https://github.com/XRPLF/rippled/pull/5163))
- `ledger_data` (gRPC): Fixed crashing issue with some invalid markers. ([#5137](https://github.com/XRPLF/rippled/pull/5137))
- `account_lines`: Fixed error with `no_ripple` and `no_ripple_peer` sometimes showing up incorrectly. ([#5345](https://github.com/XRPLF/rippled/pull/5345))
- `account_tx`: Fixed issue with incorrect CTIDs. ([#5408](https://github.com/XRPLF/rippled/pull/5408))
## XRP Ledger server version 2.4.0
@@ -103,11 +92,19 @@ As of 2025-04-04, version 2.5.0 is in development. You can use a pre-release ver
### Additions and bugfixes in 2.4.0
- `ledger_entry`: `state` is added an alias for `ripple_state`.
- `ledger_entry`: Enables case-insensitive filtering by canonical name in addition to case-sensitive filtering by RPC name.
- `validators`: Added new field `validator_list_threshold` in response.
- `simulate`: A new RPC that executes a [dry run of a transaction submission](https://github.com/XRPLF/XRPL-Standards/tree/master/XLS-0069d-simulate#2-rpc-simulate)
- Signing methods autofill fees better and properly handle transactions that don't have a base fee, and will also autofill the `NetworkID` field.
- `simulate`: A new RPC that executes a [dry run of a transaction submission](https://github.com/XRPLF/XRPL-Standards/tree/master/XLS-0069d-simulate#2-rpc-simulate). ([#5069](https://github.com/XRPLF/rippled/pull/5069))
- Signing methods (`sign`, `sign_for`, `submit`): Autofill fees better, properly handle transactions without a base fee, and autofill the `NetworkID` field. ([#5069](https://github.com/XRPLF/rippled/pull/5069))
- `ledger_entry`: `state` is added as an alias for `ripple_state`. ([#5199](https://github.com/XRPLF/rippled/pull/5199))
- `ledger`, `ledger_data`, `account_objects`: Support filtering ledger entry types by their canonical names (case-insensitive). ([#5271](https://github.com/XRPLF/rippled/pull/5271))
- `validators`: Added new field `validator_list_threshold` in response. ([#5112](https://github.com/XRPLF/rippled/pull/5112))
- `server_info`: Added git commit hash info on admin connection. ([#5225](https://github.com/XRPLF/rippled/pull/5225))
- `server_definitions`: Changed larger `UInt` serialized types to `Hash`. ([#5231](https://github.com/XRPLF/rippled/pull/5231))
## XRP Ledger server version 2.3.1
[Version 2.3.1](https://github.com/XRPLF/rippled/releases/tag/2.3.1) was released on Jan 29, 2025.
This release contains bug fixes only and no API changes.
## XRP Ledger server version 2.3.0
@@ -115,19 +112,30 @@ As of 2025-04-04, version 2.5.0 is in development. You can use a pre-release ver
### Breaking changes in 2.3.0
- `book_changes`: If the requested ledger version is not available on this node, a `ledgerNotFound` error is returned and the node does not attempt to acquire the ledger from the p2p network (as with other non-admin RPCs).
Admins can still attempt to retrieve old ledgers with the `ledger_request` RPC.
- `book_changes`: If the requested ledger version is not available on this node, a `ledgerNotFound` error is returned and the node does not attempt to acquire the ledger from the p2p network (as with other non-admin RPCs). Admins can still attempt to retrieve old ledgers with the `ledger_request` RPC.
### Additions and bugfixes in 2.3.0
- `book_changes`: Returns a `validated` field in its response, which was missing in prior versions.
- `book_changes`: Returns a `validated` field in its response. ([#5096](https://github.com/XRPLF/rippled/pull/5096))
- `book_changes`: Accepts shortcut strings (`current`, `closed`, `validated`) for the `ledger_index` parameter. ([#5096](https://github.com/XRPLF/rippled/pull/5096))
- `server_definitions`: Include `index` in response. ([#5190](https://github.com/XRPLF/rippled/pull/5190))
- `account_nfts`: Fix issue where unassociated marker would return incorrect results. ([#5045](https://github.com/XRPLF/rippled/pull/5045))
- `account_objects`: Fix issue where invalid marker would not return an error. ([#5046](https://github.com/XRPLF/rippled/pull/5046))
- `account_objects`: Disallow filtering by ledger entry types that an account cannot hold. ([#5056](https://github.com/XRPLF/rippled/pull/5056))
- `tx`: Allow lowercase CTID. ([#5049](https://github.com/XRPLF/rippled/pull/5049))
- `feature`: Better error handling for invalid values of `feature`. ([#5063](https://github.com/XRPLF/rippled/pull/5063))
## XRP Ledger server version 2.2.0
[Version 2.2.0](https://github.com/XRPLF/rippled/releases/tag/2.2.0) was released on Jun 5, 2024. The following additions are non-breaking (because they are purely additive):
- The `feature` method now has a non-admin mode for users. (It was previously only available to admin connections.) The method returns an updated list of amendments, including their names and other information. ([#4781](https://github.com/XRPLF/rippled/pull/4781))
- `feature`: Add a non-admin mode for users. (It was previously only available to admin connections.) The method returns an updated list of amendments, including their names and other information. ([#4781](https://github.com/XRPLF/rippled/pull/4781))
## XRP Ledger server version 2.0.1
[Version 2.0.1](https://github.com/XRPLF/rippled/releases/tag/2.0.1) was released on Jan 29, 2024. The following additions are non-breaking:
- `path_find`: Fixes unbounded memory growth. ([#4822](https://github.com/XRPLF/rippled/pull/4822))
## XRP Ledger server version 2.0.0
@@ -135,24 +143,18 @@ Admins can still attempt to retrieve old ledgers with the `ledger_request` RPC.
- `server_definitions`: A new RPC that generates a `definitions.json`-like output that can be used in XRPL libraries.
- In `Payment` transactions, `DeliverMax` has been added. This is a replacement for the `Amount` field, which should not be used. Typically, the `delivered_amount` (in transaction metadata) should be used. To ease the transition, `DeliverMax` is present regardless of API version, since adding a field is non-breaking.
- API version 2 has been moved from beta to supported, meaning that it is generally available (regardless of the `beta_rpc_api` setting).
## XRP Ledger server version 2.2.0
The following is a non-breaking addition to the API.
- The `feature` method now has a non-admin mode for users. (It was previously only available to admin connections.) The method returns an updated list of amendments, including their names and other information. ([#4781](https://github.com/XRPLF/rippled/pull/4781))
- API version 2 has been moved from beta to supported, meaning that it is generally available (regardless of the `beta_rpc_api` setting). The full list of changes is in [API-VERSION-2.md](API-VERSION-2.md).
## XRP Ledger server version 1.12.0
[Version 1.12.0](https://github.com/XRPLF/rippled/releases/tag/1.12.0) was released on Sep 6, 2023. The following additions are non-breaking (because they are purely additive).
[Version 1.12.0](https://github.com/XRPLF/rippled/releases/tag/1.12.0) was released on Sep 6, 2023. The following additions are non-breaking (because they are purely additive):
- `server_info`: Added `ports`, an array which advertises the RPC and WebSocket ports. This information is also included in the `/crawl` endpoint (which calls `server_info` internally). `grpc` and `peer` ports are also included. (https://github.com/XRPLF/rippled/pull/4427)
- `server_info`: Added `ports`, an array which advertises the RPC and WebSocket ports. This information is also included in the `/crawl` endpoint (which calls `server_info` internally). `grpc` and `peer` ports are also included. ([#4427](https://github.com/XRPLF/rippled/pull/4427))
- `ports` contains objects, each containing a `port` for the listening port (a number string), and a `protocol` array listing the supported protocols on that port.
- This allows crawlers to build a more detailed topology without needing to port-scan nodes.
- (For peers and other non-admin clients, the info about admin ports is excluded.)
- Clawback: The following additions are gated by the Clawback amendment (`featureClawback`). (https://github.com/XRPLF/rippled/pull/4553)
- Adds an [AccountRoot flag](https://xrpl.org/accountroot.html#accountroot-flags) called `lsfAllowTrustLineClawback` (https://github.com/XRPLF/rippled/pull/4617)
- Clawback: The following additions are gated by the Clawback amendment (`featureClawback`). ([#4553](https://github.com/XRPLF/rippled/pull/4553))
- Adds an [AccountRoot flag](https://xrpl.org/accountroot.html#accountroot-flags) called `lsfAllowTrustLineClawback`. ([#4617](https://github.com/XRPLF/rippled/pull/4617))
- Adds the corresponding `asfAllowTrustLineClawback` [AccountSet Flag](https://xrpl.org/accountset.html#accountset-flags) as well.
- Clawback is disabled by default, so if an issuer desires the ability to claw back funds, they must use an `AccountSet` transaction to set the AllowTrustLineClawback flag. They must do this before creating any trust lines, offers, escrows, payment channels, or checks.
- Adds the [Clawback transaction type](https://github.com/XRPLF/XRPL-Standards/blob/master/XLS-39d-clawback/README.md#331-clawback-transaction), containing these fields:
@@ -187,16 +189,16 @@ The following is a non-breaking addition to the API.
### Breaking changes in 1.11
- Added the ability to mark amendments as obsolete. For the `feature` admin API, there is a new possible value for the `vetoed` field. (https://github.com/XRPLF/rippled/pull/4291)
- Added the ability to mark amendments as obsolete. For the `feature` admin API, there is a new possible value for the `vetoed` field. ([#4291](https://github.com/XRPLF/rippled/pull/4291))
- The value of `vetoed` can now be `true`, `false`, or `"Obsolete"`.
- Removed the acceptance of seeds or public keys in place of account addresses. (https://github.com/XRPLF/rippled/pull/4404)
- Removed the acceptance of seeds or public keys in place of account addresses. ([#4404](https://github.com/XRPLF/rippled/pull/4404))
- This simplifies the API and encourages better security practices (i.e. seeds should never be sent over the network).
- For the `ledger_data` method, when all entries are filtered out, the `state` field of the response is now an empty list (in other words, an empty array, `[]`). (Previously, it would return `null`.) While this is technically a breaking change, the new behavior is consistent with the documentation, so this is considered only a bug fix. (https://github.com/XRPLF/rippled/pull/4398)
- For the `ledger_data` method, when all entries are filtered out, the `state` field of the response is now an empty list (in other words, an empty array, `[]`). (Previously, it would return `null`.) While this is technically a breaking change, the new behavior is consistent with the documentation, so this is considered only a bug fix. ([#4398](https://github.com/XRPLF/rippled/pull/4398))
- If and when the `fixNFTokenRemint` amendment activates, there will be a new AccountRoot field, `FirstNFTSequence`. This field is set to the current account sequence when the account issues their first NFT. If an account has not issued any NFTs, then the field is not set. ([#4406](https://github.com/XRPLF/rippled/pull/4406))
- There is a new account deletion restriction: an account can only be deleted if `FirstNFTSequence` + `MintedNFTokens` + `256` is less than the current ledger sequence.
- This is potentially a breaking change if clients have logic for determining whether an account can be deleted.
- NetworkID
- For sidechains and networks with a network ID greater than 1024, there is a new [transaction common field](https://xrpl.org/transaction-common-fields.html), `NetworkID`. (https://github.com/XRPLF/rippled/pull/4370)
- For sidechains and networks with a network ID greater than 1024, there is a new [transaction common field](https://xrpl.org/transaction-common-fields.html), `NetworkID`. ([#4370](https://github.com/XRPLF/rippled/pull/4370))
- This field helps to prevent replay attacks and is now required for chains whose network ID is 1025 or higher.
- The field must be omitted for Mainnet, so there is no change for Mainnet users.
- There are three new local error codes:
@@ -206,10 +208,10 @@ The following is a non-breaking addition to the API.
### Additions and bug fixes in 1.11
- Added `nftoken_id`, `nftoken_ids` and `offer_id` meta fields into NFT `tx` and `account_tx` responses. (https://github.com/XRPLF/rippled/pull/4447)
- Added an `account_flags` object to the `account_info` method response. (https://github.com/XRPLF/rippled/pull/4459)
- Added `NFTokenPages` to the `account_objects` RPC. (https://github.com/XRPLF/rippled/pull/4352)
- Fixed: `marker` returned from the `account_lines` command would not work on subsequent commands. (https://github.com/XRPLF/rippled/pull/4361)
- Added `nftoken_id`, `nftoken_ids` and `offer_id` meta fields into NFT `tx` and `account_tx` responses. ([#4447](https://github.com/XRPLF/rippled/pull/4447))
- Added an `account_flags` object to the `account_info` method response. ([#4459](https://github.com/XRPLF/rippled/pull/4459))
- Added `NFTokenPages` to the `account_objects` RPC. ([#4352](https://github.com/XRPLF/rippled/pull/4352))
- Fixed: `marker` returned from the `account_lines` command would not work on subsequent commands. ([#4361](https://github.com/XRPLF/rippled/pull/4361))
## XRP Ledger server version 1.10.0

66
API-VERSION-2.md Normal file
View File

@@ -0,0 +1,66 @@
# API Version 2
API version 2 is available in `rippled` version 2.0.0 and later. To use this API, clients specify `"api_version" : 2` in each request.
For info about how [API versioning](https://xrpl.org/request-formatting.html#api-versioning) works, including examples, please view the [XLS-22d spec](https://github.com/XRPLF/XRPL-Standards/discussions/54). For details about the implementation of API versioning, view the [implementation PR](https://github.com/XRPLF/rippled/pull/3155). API versioning ensures existing integrations and users continue to receive existing behavior, while those that request a higher API version will experience new behavior.
## Removed methods
In API version 2, the following deprecated methods are no longer available: ([#4759](https://github.com/XRPLF/rippled/pull/4759))
- `tx_history` - Instead, use other methods such as `account_tx` or `ledger` with the `transactions` field set to `true`.
- `ledger_header` - Instead, use the `ledger` method.
## Modifications to JSON transaction element in API version 2
In API version 2, JSON elements for transaction output have been changed and made consistent for all methods which output transactions. ([#4775](https://github.com/XRPLF/rippled/pull/4775))
This helps to unify the JSON serialization format of transactions. ([clio#722](https://github.com/XRPLF/clio/issues/722), [#4727](https://github.com/XRPLF/rippled/issues/4727))
- JSON transaction element is named `tx_json`
- Binary transaction element is named `tx_blob`
- JSON transaction metadata element is named `meta`
- Binary transaction metadata element is named `meta_blob`
Additionally, these elements are now consistently available next to `tx_json` (i.e. sibling elements), where possible:
- `hash` - Transaction ID. This data was stored inside transaction output in API version 1, but in API version 2 is a sibling element.
- `ledger_index` - Ledger index (only set on validated ledgers)
- `ledger_hash` - Ledger hash (only set on closed or validated ledgers)
- `close_time_iso` - Ledger close time expressed in ISO 8601 time format (only set on validated ledgers)
- `validated` - Bool element set to `true` if the transaction is in a validated ledger, otherwise `false`
This change affects the following methods:
- `tx` - Transaction data moved into element `tx_json` (was inline inside `result`) or, if binary output was requested, moved from `tx` to `tx_blob`. Renamed binary transaction metadata element (if it was requested) from `meta` to `meta_blob`. Changed location of `hash` and added new elements
- `account_tx` - Renamed transaction element from `tx` to `tx_json`. Renamed binary transaction metadata element (if it was requested) from `meta` to `meta_blob`. Changed location of `hash` and added new elements
- `transaction_entry` - Renamed transaction metadata element from `metadata` to `meta`. Changed location of `hash` and added new elements
- `subscribe` - Renamed transaction element from `transaction` to `tx_json`. Changed location of `hash` and added new elements
- `sign`, `sign_for`, `submit` and `submit_multisigned` - Changed location of `hash` element.
## Modifications to `Payment` transaction JSON schema
When reading Payments, the `Amount` field should generally **not** be used. Instead, use [delivered_amount](https://xrpl.org/partial-payments.html#the-delivered_amount-field) to see the amount that the Payment delivered. To clarify its meaning, the `Amount` field is being renamed to `DeliverMax`. ([#4733](https://github.com/XRPLF/rippled/pull/4733))
- In `Payment` transaction type, JSON RPC field `Amount` is renamed to `DeliverMax`. To enable smooth client transition, `Amount` is still handled, as described below: ([#4733](https://github.com/XRPLF/rippled/pull/4733))
- On JSON RPC input (e.g. `submit_multisigned` etc. methods), `Amount` is recognized as an alias to `DeliverMax` for both API version 1 and version 2 clients.
- On JSON RPC input, submitting both `Amount` and `DeliverMax` fields is allowed _only_ if they are identical; otherwise such input is rejected with `rpcINVALID_PARAMS` error.
- On JSON RPC output (e.g. `subscribe`, `account_tx` etc. methods), `DeliverMax` is present in both API version 1 and version 2.
- On JSON RPC output, `Amount` is only present in API version 1 and _not_ in version 2.
## Modifications to account_info response
- `signer_lists` is returned in the root of the response. (In API version 1, it was nested under `account_data`.) ([#3770](https://github.com/XRPLF/rippled/pull/3770))
- When using an invalid `signer_lists` value, the API now returns an "invalidParams" error. ([#4585](https://github.com/XRPLF/rippled/pull/4585))
- (`signer_lists` must be a boolean. In API version 1, strings were accepted and may return a normal response - i.e. as if `signer_lists` were `true`.)
## Modifications to [account_tx](https://xrpl.org/account_tx.html#account_tx) response
- Using `ledger_index_min`, `ledger_index_max`, and `ledger_index` returns `invalidParams` because if you use `ledger_index_min` or `ledger_index_max`, then it does not make sense to also specify `ledger_index`. In API version 1, no error was returned. ([#4571](https://github.com/XRPLF/rippled/pull/4571))
- The same applies for `ledger_index_min`, `ledger_index_max`, and `ledger_hash`. ([#4545](https://github.com/XRPLF/rippled/issues/4545#issuecomment-1565065579))
- Using a `ledger_index_min` or `ledger_index_max` beyond the range of ledgers that the server has:
- returns `lgrIdxMalformed` in API version 2. Previously, in API version 1, no error was returned. ([#4288](https://github.com/XRPLF/rippled/issues/4288))
- Attempting to use a non-boolean value (such as a string) for the `binary` or `forward` parameters returns `invalidParams` (`rpcINVALID_PARAMS`). Previously, in API version 1, no error was returned. ([#4620](https://github.com/XRPLF/rippled/pull/4620))
## Modifications to [noripple_check](https://xrpl.org/noripple_check.html#noripple_check) response
- Attempting to use a non-boolean value (such as a string) for the `transactions` parameter returns `invalidParams` (`rpcINVALID_PARAMS`). Previously, in API version 1, no error was returned. ([#4620](https://github.com/XRPLF/rippled/pull/4620))

27
API-VERSION-3.md Normal file
View File

@@ -0,0 +1,27 @@
# API Version 3
API version 3 is currently a **beta API**. It requires enabling `[beta_rpc_api]` in the rippled configuration to use. To use this API, clients specify `"api_version" : 3` in each request.
For info about how [API versioning](https://xrpl.org/request-formatting.html#api-versioning) works, including examples, please view the [XLS-22d spec](https://github.com/XRPLF/XRPL-Standards/discussions/54). For details about the implementation of API versioning, view the [implementation PR](https://github.com/XRPLF/rippled/pull/3155). API versioning ensures existing integrations and users continue to receive existing behavior, while those that request a higher API version will experience new behavior.
## Breaking Changes
### Modifications to `amm_info`
The order of error checks has been changed to provide more specific error messages. ([#4924](https://github.com/XRPLF/rippled/pull/4924))
- **Before (API v2)**: When sending an invalid account or asset to `amm_info` while other parameters are not set as expected, the method returns a generic `rpcINVALID_PARAMS` error.
- **After (API v3)**: The same scenario returns a more specific error: `rpcISSUE_MALFORMED` for malformed assets or `rpcACT_MALFORMED` for malformed accounts.
### Modifications to `ledger_entry`
Added support for string shortcuts to look up fixed-location ledger entries using the `"index"` parameter. ([#5644](https://github.com/XRPLF/rippled/pull/5644))
In API version 3, the following string values can be used with the `"index"` parameter:
- `"index": "amendments"` - Returns the `Amendments` ledger entry
- `"index": "fee"` - Returns the `FeeSettings` ledger entry
- `"index": "nunl"` - Returns the `NegativeUNL` ledger entry
- `"index": "hashes"` - Returns the "short" `LedgerHashes` ledger entry (recent ledger hashes)
These shortcuts are only available in API version 3 and later. In API versions 1 and 2, these string values would result in an error.

View File

@@ -872,7 +872,8 @@ git push --delete upstream-push master-next
11. [Create a new release on
Github](https://github.com/XRPLF/rippled/releases). Be sure that
"Set as the latest release" is checked.
12. Finally [reverse merge the release into `develop`](#follow-up-reverse-merge).
12. Open a PR to update the [API-CHANGELOG](API-CHANGELOG.md) and `API-VERSION-[n].md` with the changes for this release (if any are missing).
13. Finally, [reverse merge the release into `develop`](#follow-up-reverse-merge).
#### Special cases: point releases, hotfixes, etc.

View File

@@ -940,23 +940,7 @@
#
# path Location to store the database
#
# Optional keys
#
# cache_size Size of cache for database records. Default is 16384.
# Setting this value to 0 will use the default value.
#
# cache_age Length of time in minutes to keep database records
# cached. Default is 5 minutes. Setting this value to
# 0 will use the default value.
#
# Note: if neither cache_size nor cache_age is
# specified, the cache for database records will not
# be created. If only one of cache_size or cache_age
# is specified, the cache will be created using the
# default value for the unspecified parameter.
#
# Note: the cache will not be created if online_delete
# is specified.
# Optional keys for NuDB and RocksDB:
#
# fast_load Boolean. If set, load the last persisted ledger
# from disk upon process start before syncing to
@@ -964,8 +948,6 @@
# if sufficient IOPS capacity is available.
# Default 0.
#
# Optional keys for NuDB or RocksDB:
#
# earliest_seq The default is 32570 to match the XRP ledger
# network's earliest allowed sequence. Alternate
# networks may set this value. Minimum value of 1.

View File

@@ -51,7 +51,7 @@ set(is_amd64 FALSE)
set(is_arm64 FALSE)
if (CMAKE_SYSTEM_PROCESSOR MATCHES "x86_64|AMD64")
set(is_amd64 TRUE)
elseif (CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64|arm64")
elseif (CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64|arm64|ARM64")
set(is_arm64 TRUE)
else ()
message(FATAL_ERROR "Unknown architecture: ${CMAKE_SYSTEM_PROCESSOR}")

View File

@@ -134,10 +134,6 @@ public:
std::uint32_t ledgerSeq,
std::function<void(std::shared_ptr<NodeObject> const&)>&& callback);
/** Remove expired entries from the positive and negative caches. */
virtual void
sweep() = 0;
/** Gather statistics pertaining to read and write activities.
*
* @param obj Json object reference into which to place counters.

View File

@@ -24,32 +24,6 @@ public:
beast::Journal j)
: Database(scheduler, readThreads, config, j), backend_(std::move(backend))
{
std::optional<int> cacheSize, cacheAge;
if (config.exists("cache_size"))
{
cacheSize = get<int>(config, "cache_size");
if (cacheSize.value() < 0)
{
Throw<std::runtime_error>("Specified negative value for cache_size");
}
}
if (config.exists("cache_age"))
{
cacheAge = get<int>(config, "cache_age");
if (cacheAge.value() < 0)
{
Throw<std::runtime_error>("Specified negative value for cache_age");
}
}
if (cacheSize != 0 || cacheAge != 0)
{
cache_ = std::make_shared<TaggedCache<uint256, NodeObject>>(
"DatabaseNodeImp", cacheSize.value_or(0), std::chrono::minutes(cacheAge.value_or(0)), stopwatch(), j);
}
XRPL_ASSERT(
backend_,
"xrpl::NodeStore::DatabaseNodeImp::DatabaseNodeImp : non-null "
@@ -104,13 +78,7 @@ public:
std::uint32_t ledgerSeq,
std::function<void(std::shared_ptr<NodeObject> const&)>&& callback) override;
void
sweep() override;
private:
// Cache for database objects. This cache is not always initialized. Check
// for null before using.
std::shared_ptr<TaggedCache<uint256, NodeObject>> cache_;
// Persistent key/value storage
std::shared_ptr<Backend> backend_;

View File

@@ -56,9 +56,6 @@ public:
void
sync() override;
void
sweep() override;
private:
std::shared_ptr<Backend> writableBackend_;
std::shared_ptr<Backend> archiveBackend_;

View File

@@ -10,11 +10,6 @@ DatabaseNodeImp::store(NodeObjectType type, Blob&& data, uint256 const& hash, st
auto obj = NodeObject::createObject(type, std::move(data), hash);
backend_->store(obj);
if (cache_)
{
// After the store, replace a negative cache entry if there is one
cache_->canonicalize(hash, obj, [](std::shared_ptr<NodeObject> const& n) { return n->getType() == hotDUMMY; });
}
}
void
@@ -23,77 +18,36 @@ DatabaseNodeImp::asyncFetch(
std::uint32_t ledgerSeq,
std::function<void(std::shared_ptr<NodeObject> const&)>&& callback)
{
if (cache_)
{
std::shared_ptr<NodeObject> obj = cache_->fetch(hash);
if (obj)
{
callback(obj->getType() == hotDUMMY ? nullptr : obj);
return;
}
}
Database::asyncFetch(hash, ledgerSeq, std::move(callback));
}
void
DatabaseNodeImp::sweep()
{
if (cache_)
cache_->sweep();
}
std::shared_ptr<NodeObject>
DatabaseNodeImp::fetchNodeObject(uint256 const& hash, std::uint32_t, FetchReport& fetchReport, bool duplicate)
{
std::shared_ptr<NodeObject> nodeObject = cache_ ? cache_->fetch(hash) : nullptr;
std::shared_ptr<NodeObject> nodeObject = nullptr;
Status status;
if (!nodeObject)
try
{
JLOG(j_.trace()) << "fetchNodeObject " << hash << ": record not " << (cache_ ? "cached" : "found");
Status status;
try
{
status = backend_->fetch(hash.data(), &nodeObject);
}
catch (std::exception const& e)
{
JLOG(j_.fatal()) << "fetchNodeObject " << hash << ": Exception fetching from backend: " << e.what();
Rethrow();
}
switch (status)
{
case ok:
if (cache_)
{
if (nodeObject)
cache_->canonicalize_replace_client(hash, nodeObject);
else
{
auto notFound = NodeObject::createObject(hotDUMMY, {}, hash);
cache_->canonicalize_replace_client(hash, notFound);
if (notFound->getType() != hotDUMMY)
nodeObject = notFound;
}
}
break;
case notFound:
break;
case dataCorrupt:
JLOG(j_.fatal()) << "fetchNodeObject " << hash << ": nodestore data is corrupted";
break;
default:
JLOG(j_.warn()) << "fetchNodeObject " << hash << ": backend returns unknown result " << status;
break;
}
status = backend_->fetch(hash.data(), &nodeObject);
}
else
catch (std::exception const& e)
{
JLOG(j_.trace()) << "fetchNodeObject " << hash << ": record found in cache";
if (nodeObject->getType() == hotDUMMY)
nodeObject.reset();
JLOG(j_.fatal()) << "fetchNodeObject " << hash << ": Exception fetching from backend: " << e.what();
Rethrow();
}
switch (status)
{
case ok:
case notFound:
break;
case dataCorrupt:
JLOG(j_.fatal()) << "fetchNodeObject " << hash << ": nodestore data is corrupted";
break;
default:
JLOG(j_.warn()) << "fetchNodeObject " << hash << ": backend returns unknown result " << status;
break;
}
if (nodeObject)
@@ -105,66 +59,29 @@ DatabaseNodeImp::fetchNodeObject(uint256 const& hash, std::uint32_t, FetchReport
std::vector<std::shared_ptr<NodeObject>>
DatabaseNodeImp::fetchBatch(std::vector<uint256> const& hashes)
{
std::vector<std::shared_ptr<NodeObject>> results{hashes.size()};
using namespace std::chrono;
auto const before = steady_clock::now();
std::unordered_map<uint256 const*, size_t> indexMap;
std::vector<uint256 const*> cacheMisses;
uint64_t hits = 0;
uint64_t fetches = 0;
std::vector<uint256 const*> batch{};
batch.reserve(hashes.size());
for (size_t i = 0; i < hashes.size(); ++i)
{
auto const& hash = hashes[i];
// See if the object already exists in the cache
auto nObj = cache_ ? cache_->fetch(hash) : nullptr;
++fetches;
if (!nObj)
{
// Try the database
indexMap[&hash] = i;
cacheMisses.push_back(&hash);
}
else
{
results[i] = nObj->getType() == hotDUMMY ? nullptr : nObj;
// It was in the cache.
++hits;
}
batch.push_back(&hash);
}
JLOG(j_.debug()) << "fetchBatch - cache hits = " << (hashes.size() - cacheMisses.size())
<< " - cache misses = " << cacheMisses.size();
auto dbResults = backend_->fetchBatch(cacheMisses).first;
for (size_t i = 0; i < dbResults.size(); ++i)
auto results = backend_->fetchBatch(batch).first;
for (size_t i = 0; i < results.size(); ++i)
{
auto nObj = std::move(dbResults[i]);
size_t index = indexMap[cacheMisses[i]];
auto const& hash = hashes[index];
if (nObj)
{
// Ensure all threads get the same object
if (cache_)
cache_->canonicalize_replace_client(hash, nObj);
}
else
if (!results[i])
{
JLOG(j_.error()) << "fetchBatch - "
<< "record not found in db or cache. hash = " << strHex(hash);
if (cache_)
{
auto notFound = NodeObject::createObject(hotDUMMY, {}, hash);
cache_->canonicalize_replace_client(hash, notFound);
if (notFound->getType() != hotDUMMY)
nObj = std::move(notFound);
}
<< "record not found in db. hash = " << strHex(hashes[i]);
}
results[index] = std::move(nObj);
}
auto fetchDurationUs = std::chrono::duration_cast<std::chrono::microseconds>(steady_clock::now() - before).count();
updateFetchMetrics(fetches, hits, fetchDurationUs);
updateFetchMetrics(hashes.size(), 0, fetchDurationUs);
return results;
}

View File

@@ -93,12 +93,6 @@ DatabaseRotatingImp::store(NodeObjectType type, Blob&& data, uint256 const& hash
storeStats(1, nObj->getData().size());
}
void
DatabaseRotatingImp::sweep()
{
// nothing to do
}
std::shared_ptr<NodeObject>
DatabaseRotatingImp::fetchNodeObject(uint256 const& hash, std::uint32_t, FetchReport& fetchReport, bool duplicate)
{

View File

@@ -490,19 +490,8 @@ public:
Env env(*this, envconfig(onlineDelete));
/////////////////////////////////////////////////////////////
// Create the backend. Normally, SHAMapStoreImp handles all these
// details
auto nscfg = env.app().config().section(ConfigSection::nodeDatabase());
// Provide default values:
if (!nscfg.exists("cache_size"))
nscfg.set(
"cache_size", std::to_string(env.app().config().getValueFor(SizedItem::treeCacheSize, std::nullopt)));
if (!nscfg.exists("cache_age"))
nscfg.set(
"cache_age", std::to_string(env.app().config().getValueFor(SizedItem::treeCacheAge, std::nullopt)));
// Create NodeStore with two backends to allow online deletion of data.
// Normally, SHAMapStoreImp handles all these details.
NodeStoreScheduler scheduler(env.app().getJobQueue());
std::string const writableDb = "write";
@@ -510,9 +499,8 @@ public:
auto writableBackend = makeBackendRotating(env, scheduler, writableDb);
auto archiveBackend = makeBackendRotating(env, scheduler, archiveDb);
// Create NodeStore with two backends to allow online deletion of
// data
constexpr int readThreads = 4;
auto nscfg = env.app().config().section(ConfigSection::nodeDatabase());
auto dbr = std::make_unique<NodeStore::DatabaseRotatingImp>(
scheduler,
readThreads,

View File

@@ -1,96 +0,0 @@
#include <test/jtx.h>
#include <xrpld/core/ConfigSections.h>
#include <xrpl/protocol/jss.h>
namespace xrpl {
class Submit_test : public beast::unit_test::suite
{
public:
void
testAugmentedFields()
{
testcase("Augmented fields in sign-and-submit mode");
using namespace test::jtx;
// Enable signing support in config
Env env{*this, envconfig([](std::unique_ptr<Config> cfg) {
cfg->loadFromString("[" SECTION_SIGNING_SUPPORT "]\ntrue");
return cfg;
})};
Account const alice{"alice"};
Account const bob{"bob"};
env.fund(XRP(10000), alice, bob);
env.close();
// Test 1: Sign-and-submit mode should return augmented fields
{
Json::Value jv;
jv[jss::tx_json][jss::TransactionType] = jss::Payment;
jv[jss::tx_json][jss::Account] = alice.human();
jv[jss::tx_json][jss::Destination] = bob.human();
jv[jss::tx_json][jss::Amount] = XRP(100).value().getJson(JsonOptions::none);
jv[jss::secret] = alice.name();
auto const result = env.rpc("json", "submit", to_string(jv))[jss::result];
// These are the augmented fields that should be present
BEAST_EXPECT(result.isMember(jss::engine_result));
BEAST_EXPECT(result.isMember(jss::engine_result_code));
BEAST_EXPECT(result.isMember(jss::engine_result_message));
// New augmented fields from issue #3125
BEAST_EXPECT(result.isMember(jss::accepted));
BEAST_EXPECT(result.isMember(jss::applied));
BEAST_EXPECT(result.isMember(jss::broadcast));
BEAST_EXPECT(result.isMember(jss::queued));
BEAST_EXPECT(result.isMember(jss::kept));
// Current ledger state fields
BEAST_EXPECT(result.isMember(jss::account_sequence_next));
BEAST_EXPECT(result.isMember(jss::account_sequence_available));
BEAST_EXPECT(result.isMember(jss::open_ledger_cost));
BEAST_EXPECT(result.isMember(jss::validated_ledger_index));
// Verify basic transaction fields
BEAST_EXPECT(result.isMember(jss::tx_blob));
BEAST_EXPECT(result.isMember(jss::tx_json));
}
// Test 2: Binary blob mode should also return augmented fields (regression test)
{
auto jt = env.jt(pay(alice, bob, XRP(100)));
Serializer s;
jt.stx->add(s);
auto const result = env.rpc("submit", strHex(s.slice()))[jss::result];
// Verify augmented fields are present in binary mode too
BEAST_EXPECT(result.isMember(jss::engine_result));
BEAST_EXPECT(result.isMember(jss::accepted));
BEAST_EXPECT(result.isMember(jss::applied));
BEAST_EXPECT(result.isMember(jss::broadcast));
BEAST_EXPECT(result.isMember(jss::queued));
BEAST_EXPECT(result.isMember(jss::kept));
BEAST_EXPECT(result.isMember(jss::account_sequence_next));
BEAST_EXPECT(result.isMember(jss::account_sequence_available));
BEAST_EXPECT(result.isMember(jss::open_ledger_cost));
BEAST_EXPECT(result.isMember(jss::validated_ledger_index));
}
}
void
run() override
{
testAugmentedFields();
}
};
BEAST_DEFINE_TESTSUITE(Submit, rpc, xrpl);
} // namespace xrpl

View File

@@ -908,10 +908,6 @@ public:
JLOG(m_journal.debug()) << "MasterTransaction sweep. Size before: " << oldMasterTxSize
<< "; size after: " << masterTxCache.size();
}
{
// Does not appear to have an associated cache.
getNodeStore().sweep();
}
{
std::size_t const oldLedgerMasterCacheSize = getLedgerMaster().getFetchPackCacheSize();

View File

@@ -130,14 +130,6 @@ std::unique_ptr<NodeStore::Database>
SHAMapStoreImp::makeNodeStore(int readThreads)
{
auto nscfg = app_.config().section(ConfigSection::nodeDatabase());
// Provide default values:
if (!nscfg.exists("cache_size"))
nscfg.set("cache_size", std::to_string(app_.config().getValueFor(SizedItem::treeCacheSize, std::nullopt)));
if (!nscfg.exists("cache_age"))
nscfg.set("cache_age", std::to_string(app_.config().getValueFor(SizedItem::treeCacheAge, std::nullopt)));
std::unique_ptr<NodeStore::Database> db;
if (deleteInterval_)
@@ -226,8 +218,6 @@ SHAMapStoreImp::run()
LedgerIndex lastRotated = state_db_.getState().lastRotated;
netOPs_ = &app_.getOPs();
ledgerMaster_ = &app_.getLedgerMaster();
fullBelowCache_ = &(*app_.getNodeFamily().getFullBelowCache());
treeNodeCache_ = &(*app_.getNodeFamily().getTreeNodeCache());
if (advisoryDelete_)
canDelete_ = state_db_.getCanDelete();
@@ -490,16 +480,19 @@ void
SHAMapStoreImp::clearCaches(LedgerIndex validatedSeq)
{
ledgerMaster_->clearLedgerCachePrior(validatedSeq);
fullBelowCache_->clear();
// Also clear the FullBelowCache so its generation counter is bumped.
// This prevents stale "full below" markers from persisting across
// backend rotation/online deletion and interfering with SHAMap sync.
app_.getNodeFamily().getFullBelowCache()->clear();
}
void
SHAMapStoreImp::freshenCaches()
{
if (freshenCache(*treeNodeCache_))
return;
if (freshenCache(app_.getMasterTransaction().getCache()))
if (freshenCache(*app_.getNodeFamily().getTreeNodeCache()))
return;
freshenCache(app_.getMasterTransaction().getCache());
}
void

View File

@@ -94,8 +94,6 @@ private:
// as of run() or before
NetworkOPs* netOPs_ = nullptr;
LedgerMaster* ledgerMaster_ = nullptr;
FullBelowCache* fullBelowCache_ = nullptr;
TreeNodeCache* treeNodeCache_ = nullptr;
static constexpr auto nodeStoreName_ = "NodeStore";

View File

@@ -358,9 +358,11 @@ Config::setup(std::string const& strConf, bool bQuiet, bool bSilent, bool bStand
std::string const dbPath(legacy("database_path"));
if (!dbPath.empty())
dataDir = boost::filesystem::path(dbPath);
else if (RUN_STANDALONE)
dataDir.clear();
}
if (!RUN_STANDALONE)
if (!dataDir.empty())
{
boost::system::error_code ec;
boost::filesystem::create_directories(dataDir, ec);

View File

@@ -699,8 +699,6 @@ transactionFormatResultImpl(Transaction::pointer tpTrans, unsigned apiVersion)
jvResult[jss::engine_result] = sToken;
jvResult[jss::engine_result_code] = tpTrans->getResult();
jvResult[jss::engine_result_message] = sHuman;
RPC::populateAugmentedSubmitFields(jvResult, tpTrans);
}
}
catch (std::exception&)
@@ -714,28 +712,6 @@ transactionFormatResultImpl(Transaction::pointer tpTrans, unsigned apiVersion)
//------------------------------------------------------------------------------
void
populateAugmentedSubmitFields(Json::Value& jvResult, std::shared_ptr<Transaction> const& transaction)
{
auto const submitResult = transaction->getSubmitResult();
jvResult[jss::accepted] = submitResult.any();
jvResult[jss::applied] = submitResult.applied;
jvResult[jss::broadcast] = submitResult.broadcast;
jvResult[jss::queued] = submitResult.queued;
jvResult[jss::kept] = submitResult.kept;
if (auto currentLedgerState = transaction->getCurrentLedgerState())
{
jvResult[jss::account_sequence_next] = safe_cast<Json::Value::UInt>(currentLedgerState->accountSeqNext);
jvResult[jss::account_sequence_available] = safe_cast<Json::Value::UInt>(currentLedgerState->accountSeqAvail);
jvResult[jss::open_ledger_cost] = to_string(currentLedgerState->minFeeRequired);
jvResult[jss::validated_ledger_index] = safe_cast<Json::Value::UInt>(currentLedgerState->validatedLedger);
}
}
//------------------------------------------------------------------------------
[[nodiscard]] static XRPAmount
getTxFee(Application const& app, Config const& config, Json::Value tx)
{

View File

@@ -16,18 +16,6 @@ class TxQ;
namespace RPC {
/** Populate augmented submit fields into a JSON result.
This helper populates the submit result flags (accepted, applied,
broadcast, queued, kept) and current ledger state fields
(account_sequence_next, account_sequence_available, open_ledger_cost,
validated_ledger_index) from a Transaction pointer.
@param jvResult The JSON result to populate
@param transaction The transaction containing the submit result and state
*/
void
populateAugmentedSubmitFields(Json::Value& jvResult, std::shared_ptr<Transaction> const& transaction);
Json::Value
getCurrentNetworkFee(
Role const role,

View File

@@ -127,7 +127,23 @@ doSubmit(RPC::JsonContext& context)
jvResult[jss::engine_result_code] = transaction->getResult();
jvResult[jss::engine_result_message] = sHuman;
RPC::populateAugmentedSubmitFields(jvResult, transaction);
auto const submitResult = transaction->getSubmitResult();
jvResult[jss::accepted] = submitResult.any();
jvResult[jss::applied] = submitResult.applied;
jvResult[jss::broadcast] = submitResult.broadcast;
jvResult[jss::queued] = submitResult.queued;
jvResult[jss::kept] = submitResult.kept;
if (auto currentLedgerState = transaction->getCurrentLedgerState())
{
jvResult[jss::account_sequence_next] = safe_cast<Json::Value::UInt>(currentLedgerState->accountSeqNext);
jvResult[jss::account_sequence_available] =
safe_cast<Json::Value::UInt>(currentLedgerState->accountSeqAvail);
jvResult[jss::open_ledger_cost] = to_string(currentLedgerState->minFeeRequired);
jvResult[jss::validated_ledger_index] =
safe_cast<Json::Value::UInt>(currentLedgerState->validatedLedger);
}
}
return jvResult;