Compare commits

...

36 Commits

Author SHA1 Message Date
Clément Renault
316b4c047f Merge pull request #5940 from meilisearch/update-version-v1.23.0
Update version v1.23.0
2025-10-13 12:50:52 +00:00
Kerollmops
1d701c6980 Fix upgrade tests 2025-10-13 10:40:15 +02:00
Kerollmops
0203adb9cb Add a no-op when upgrading the index scheduler 2025-10-13 10:28:31 +02:00
Kerollmops
0d05c2ad6e Add a no-op when upgrading the index 2025-10-13 10:24:57 +02:00
Kerollmops
b3f44c4abd Bump the version to 1.23.0 2025-10-13 09:47:20 +02:00
Clémentine
62115f57b1 Merge pull request #5938 from meilisearch/attempt-license-fix-again
Try to fix GH license detection again
2025-10-09 16:32:40 +00:00
Louis Dureuil
9023172139 Add a dedicated LICENSE-MIT file containing the unmodified MIT license 2025-10-09 16:24:18 +02:00
Louis Dureuil
59631afd9a Merge pull request #5929 from meilisearch/compaction-task
Introduce a task to compact an index
2025-10-09 11:30:01 +00:00
Clément Renault
c2584c6edd Merge pull request #5936 from meilisearch/merge-v1.22.3-back
Merge v1.22.3 back into main
2025-10-09 08:45:33 +00:00
Louis Dureuil
685663af3c bump cellulite to address backcompat issue from #5307 2025-10-09 10:20:58 +02:00
Louis Dureuil
72b4b41443 Read MEILI_EXPERIMENTAL_REMOTE_SEARCH_TIMEOUT_SECONDS to override the default timeout in remote federated search 2025-10-09 09:34:49 +02:00
Louis Dureuil
70aa768d48 Update ignored test 2025-10-09 09:34:48 +02:00
Louis Dureuil
6029677eec Also raise the global deadline 2025-10-09 09:34:48 +02:00
Louis Dureuil
3c78f4121e Raise timeout to 30secs 2025-10-09 09:34:48 +02:00
Clémentine
89170dd78f Merge pull request #5935 from meilisearch/remove-release-drafter
Remove release-drafter and encourage usage of GitHub generated notes
2025-10-08 16:42:51 +00:00
Many the fish
6379a62d95 Merge pull request #5933 from meilisearch/fix-ranking-score-with-sort
Fix ranking score bug when sort is present
2025-10-08 16:23:12 +00:00
curquiza
4c05c0cf96 Remove release-drafter and encourage usage of GitHub generated notes 2025-10-08 17:35:33 +02:00
ManyTheFish
ce832da16c Add a function documentation 2025-10-08 17:19:40 +02:00
Louis Dureuil
14de657d36 Use the "currently_processing_index" to avoid potentially blocking the search during compaction 2025-10-08 15:45:38 +02:00
Kerollmops
9a36c090bf Do not return the EnvClosingEvent 2025-10-08 15:38:45 +02:00
Kerollmops
3aca010b42 Recompute the stats 2025-10-08 15:33:12 +02:00
Clément Renault
62c11ce3f3 Fix comments 2025-10-08 15:33:12 +02:00
Clément Renault
f358538f4f Improve the pre-compaction size information 2025-10-08 15:33:12 +02:00
Clément Renault
9068857ba1 Make the tests pass 2025-10-08 15:33:12 +02:00
Clément Renault
d241157084 Make Clippy happy 2025-10-08 15:33:12 +02:00
Clément Renault
69f73b1d74 Introduce a function to effectively close an index 2025-10-08 15:33:12 +02:00
Clément Renault
202794f620 Expose the env closing event so we can wait for the index to close 2025-10-08 15:33:12 +02:00
Kerollmops
38cbd54604 Implement the index compaction task 2025-10-08 15:33:12 +02:00
Kerollmops
3877e0043c Rename operation to IndexCompaction 2025-10-08 15:33:12 +02:00
Clément Renault
f95398420b Add the necessary batches and tasks in the process 2025-10-08 15:33:11 +02:00
Clément Renault
53905c1362 Add a new CompactIndex action 2025-10-08 15:33:11 +02:00
Clément Renault
113aac8815 Introduce a new /indexes/{indexUid}/compact route 2025-10-08 15:33:11 +02:00
ManyTheFish
d2071dde1f Fix ranking score bug when sort is present
- Fix global_score function to properly handle semantic scores and ranking scores
- Prioritize semantic scores (vector/embedding) when available, fall back to ranking scores
- Exclude sort and geo sort details from relevance scoring
- Use Rank::global_score to properly merge ranking scores
- Add test case with insta snapshots to reproduce and verify the fix
- When sorting is present, ranking scores now properly reflect search relevance
- Previously all ranking scores were 1.0 when sort was present, now they show actual relevance scores
2025-10-08 11:23:43 +02:00
Many the fish
4502af5aed Merge pull request #5930 from meilisearch/synonym-performance-fix
Synonym performance fix
2025-10-07 15:17:34 +00:00
ManyTheFish
06af68aa07 Get rid of upwrap in get_synonym, We can't use get_or_insert_with because the index.synonyms(..) returns a Result 2025-10-07 14:37:13 +02:00
ManyTheFish
6d378c6397 PERFORMANCE: Implement synonym caching to eliminate repeated database access
- Added SynonymCache to SearchContext to cache synonyms in memory
- Modified synonym retrieval to use cached synonyms after first load
- Eliminated redundant database calls for multi-word queries
- Performance improvement: 87% → 0ms for subsequent synonym processing
- Complex queries now process in 40ms vs 495ms (92% improvement)
2025-10-06 14:26:30 +02:00
55 changed files with 657 additions and 175 deletions

View File

@@ -1,29 +0,0 @@
name-template: 'v$RESOLVED_VERSION'
tag-template: 'v$RESOLVED_VERSION'
exclude-labels:
- 'skip changelog'
version-resolver:
minor:
labels:
- 'enhancement'
default: patch
categories:
- title: '⚠️ Breaking changes'
label: 'breaking-change'
- title: '🚀 Enhancements'
label: 'enhancement'
- title: '🐛 Bug Fixes'
label: 'bug'
- title: '🔒 Security'
label: 'security'
- title: '⚙️ Maintenance/misc'
label:
- 'dependencies'
- 'maintenance'
- 'documentation'
template: |
$CHANGES
❤️ Huge thanks to our contributors: $CONTRIBUTORS.
no-changes-template: 'Changes are coming soon 😎'
sort-direction: 'ascending'

View File

@@ -1,20 +0,0 @@
name: Release Drafter
permissions:
contents: read
pull-requests: write
on:
push:
branches:
- main
jobs:
update_release_draft:
runs-on: ubuntu-latest
steps:
- uses: release-drafter/release-drafter@v6
with:
config-name: release-draft-template.yml
env:
GITHUB_TOKEN: ${{ secrets.RELEASE_DRAFTER_TOKEN }}

38
Cargo.lock generated
View File

@@ -589,7 +589,7 @@ source = "git+https://github.com/meilisearch/bbqueue#cbb87cc707b5af415ef203bdaf2
[[package]]
name = "benchmarks"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"anyhow",
"bumpalo",
@@ -799,7 +799,7 @@ dependencies = [
[[package]]
name = "build-info"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"anyhow",
"time",
@@ -1075,9 +1075,9 @@ dependencies = [
[[package]]
name = "cellulite"
version = "0.3.1-nested-rtxns"
version = "0.3.1-nested-rtxns-2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "db298d57a80b9284327800b394ee3921307c2fdda87c6d37202f5cf400478981"
checksum = "f22d721963ead1a144f10cb8b53dc9469e760723b069123c7c7bc675c7354270"
dependencies = [
"crossbeam",
"geo",
@@ -1829,7 +1829,7 @@ dependencies = [
[[package]]
name = "dump"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"anyhow",
"big_s",
@@ -2072,7 +2072,7 @@ checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be"
[[package]]
name = "file-store"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"tempfile",
"thiserror 2.0.16",
@@ -2094,7 +2094,7 @@ dependencies = [
[[package]]
name = "filter-parser"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"insta",
"levenshtein_automata",
@@ -2122,7 +2122,7 @@ dependencies = [
[[package]]
name = "flatten-serde-json"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"criterion",
"serde_json",
@@ -2279,7 +2279,7 @@ dependencies = [
[[package]]
name = "fuzzers"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"arbitrary",
"bumpalo",
@@ -3233,7 +3233,7 @@ dependencies = [
[[package]]
name = "index-scheduler"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"anyhow",
"backoff",
@@ -3487,7 +3487,7 @@ dependencies = [
[[package]]
name = "json-depth-checker"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"criterion",
"serde_json",
@@ -3996,7 +3996,7 @@ checksum = "490cc448043f947bae3cbee9c203358d62dbee0db12107a74be5c30ccfd09771"
[[package]]
name = "meili-snap"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"insta",
"md5",
@@ -4007,7 +4007,7 @@ dependencies = [
[[package]]
name = "meilisearch"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"actix-cors",
"actix-http",
@@ -4104,7 +4104,7 @@ dependencies = [
[[package]]
name = "meilisearch-auth"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"base64 0.22.1",
"enum-iterator",
@@ -4123,7 +4123,7 @@ dependencies = [
[[package]]
name = "meilisearch-types"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"actix-web",
"anyhow",
@@ -4158,7 +4158,7 @@ dependencies = [
[[package]]
name = "meilitool"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"anyhow",
"clap",
@@ -4192,7 +4192,7 @@ dependencies = [
[[package]]
name = "milli"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"allocator-api2 0.3.1",
"arroy",
@@ -4773,7 +4773,7 @@ checksum = "9b4f627cb1b25917193a259e49bdad08f671f8d9708acfd5fe0a8c1455d87220"
[[package]]
name = "permissive-json-pointer"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"big_s",
"serde_json",
@@ -7820,7 +7820,7 @@ dependencies = [
[[package]]
name = "xtask"
version = "1.22.1"
version = "1.23.0"
dependencies = [
"anyhow",
"build-info",

View File

@@ -23,7 +23,7 @@ members = [
]
[workspace.package]
version = "1.22.1"
version = "1.23.0"
authors = [
"Quentin de Quelen <quentin@dequelen.me>",
"Clément Renault <clement@meilisearch.com>",

28
LICENSE
View File

@@ -1,29 +1,9 @@
MIT License
# License
Copyright (c) 2019-2025 Meili SAS
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
Part of this work fall under the Meilisearch Enterprise Edition (EE) and are licensed under the Business Source License 1.1, please refer to [LICENSE-EE](./LICENSE-EE) for details.
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
The other parts of this work are licensed under the [MIT license](./LICENSE-MIT).
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
---
🔒 Meilisearch Enterprise Edition (EE)
Certain parts of this codebase are not licensed under the MIT license and governed by the Business Source License 1.1.
See the LICENSE-EE file for details.
`SPDX-License-Identifier: MIT AND BUSL-1.1`

21
LICENSE-MIT Normal file
View File

@@ -0,0 +1,21 @@
MIT License
Copyright (c) 2019-2025 Meili SAS
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

View File

@@ -158,6 +158,9 @@ pub enum KindDump {
UpgradeDatabase {
from: (u32, u32, u32),
},
IndexCompaction {
index_uid: String,
},
}
impl From<Task> for TaskDump {
@@ -240,6 +243,9 @@ impl From<KindWithContent> for KindDump {
KindWithContent::UpgradeDatabase { from: version } => {
KindDump::UpgradeDatabase { from: version }
}
KindWithContent::IndexCompaction { index_uid } => {
KindDump::IndexCompaction { index_uid }
}
}
}
}

View File

@@ -234,6 +234,9 @@ impl<'a> Dump<'a> {
}
}
KindDump::UpgradeDatabase { from } => KindWithContent::UpgradeDatabase { from },
KindDump::IndexCompaction { index_uid } => {
KindWithContent::IndexCompaction { index_uid }
}
},
};

View File

@@ -341,6 +341,26 @@ impl IndexMapper {
Ok(())
}
/// Closes the specified index.
///
/// This operation involves closing the underlying environment and so can take a long time to complete.
///
/// # Panics
///
/// - If the Index corresponding to the passed name is concurrently being deleted/resized or cannot be found in the
/// in memory hash map.
pub fn close_index(&self, rtxn: &RoTxn, name: &str) -> Result<()> {
let uuid = self
.index_mapping
.get(rtxn, name)?
.ok_or_else(|| Error::IndexNotFound(name.to_string()))?;
// We remove the index from the in-memory index map.
self.index_map.write().unwrap().close_for_resize(&uuid, self.enable_mdb_writemap, 0);
Ok(())
}
/// Return an index, may open it if it wasn't already opened.
pub fn index(&self, rtxn: &RoTxn, name: &str) -> Result<Index> {
if let Some((current_name, current_index)) =

View File

@@ -317,6 +317,9 @@ fn snapshot_details(d: &Details) -> String {
Details::UpgradeDatabase { from, to } => {
format!("{{ from: {from:?}, to: {to:?} }}")
}
Details::IndexCompaction { index_uid, pre_compaction_size, post_compaction_size } => {
format!("{{ index_uid: {index_uid:?}, pre_compaction_size: {pre_compaction_size:?}, post_compaction_size: {post_compaction_size:?} }}")
}
}
}

View File

@@ -138,6 +138,17 @@ make_enum_progress! {
}
}
make_enum_progress! {
pub enum IndexCompaction {
RetrieveTheIndex,
CreateTemporaryFile,
CopyAndCompactTheIndex,
PersistTheCompactedIndex,
CloseTheIndex,
ReopenTheIndex,
}
}
make_enum_progress! {
pub enum InnerSwappingTwoIndexes {
RetrieveTheTasks,

View File

@@ -25,6 +25,7 @@ enum AutobatchKind {
IndexDeletion,
IndexUpdate,
IndexSwap,
IndexCompaction,
}
impl AutobatchKind {
@@ -68,6 +69,7 @@ impl From<KindWithContent> for AutobatchKind {
KindWithContent::IndexCreation { .. } => AutobatchKind::IndexCreation,
KindWithContent::IndexUpdate { .. } => AutobatchKind::IndexUpdate,
KindWithContent::IndexSwap { .. } => AutobatchKind::IndexSwap,
KindWithContent::IndexCompaction { .. } => AutobatchKind::IndexCompaction,
KindWithContent::TaskCancelation { .. }
| KindWithContent::TaskDeletion { .. }
| KindWithContent::DumpCreation { .. }
@@ -118,6 +120,9 @@ pub enum BatchKind {
IndexSwap {
id: TaskId,
},
IndexCompaction {
id: TaskId,
},
}
impl BatchKind {
@@ -183,6 +188,13 @@ impl BatchKind {
)),
false,
),
K::IndexCompaction => (
Break((
BatchKind::IndexCompaction { id: task_id },
BatchStopReason::TaskCannotBeBatched { kind, id: task_id },
)),
false,
),
K::DocumentClear => (Continue(BatchKind::DocumentClear { ids: vec![task_id] }), false),
K::DocumentImport { allow_index_creation, primary_key: pk }
if primary_key.is_none() || pk.is_none() || primary_key == pk.as_deref() =>
@@ -287,8 +299,10 @@ impl BatchKind {
};
match (self, autobatch_kind) {
// We don't batch any of these operations
(this, K::IndexCreation | K::IndexUpdate | K::IndexSwap | K::DocumentEdition) => Break((this, BatchStopReason::TaskCannotBeBatched { kind, id })),
// We don't batch any of these operations
(this, K::IndexCreation | K::IndexUpdate | K::IndexSwap | K::DocumentEdition | K::IndexCompaction) => {
Break((this, BatchStopReason::TaskCannotBeBatched { kind, id }))
},
// We must not batch tasks that don't have the same index creation rights if the index doesn't already exists.
(this, kind) if !index_already_exists && this.allow_index_creation() == Some(false) && kind.allow_index_creation() == Some(true) => {
Break((this, BatchStopReason::IndexCreationMismatch { id }))
@@ -483,6 +497,7 @@ impl BatchKind {
| BatchKind::IndexDeletion { .. }
| BatchKind::IndexUpdate { .. }
| BatchKind::IndexSwap { .. }
| BatchKind::IndexCompaction { .. }
| BatchKind::DocumentEdition { .. },
_,
) => {

View File

@@ -55,6 +55,10 @@ pub(crate) enum Batch {
UpgradeDatabase {
tasks: Vec<Task>,
},
IndexCompaction {
index_uid: String,
task: Task,
},
}
#[derive(Debug)]
@@ -110,7 +114,8 @@ impl Batch {
| Batch::Dump(task)
| Batch::IndexCreation { task, .. }
| Batch::Export { task }
| Batch::IndexUpdate { task, .. } => {
| Batch::IndexUpdate { task, .. }
| Batch::IndexCompaction { task, .. } => {
RoaringBitmap::from_sorted_iter(std::iter::once(task.uid)).unwrap()
}
Batch::SnapshotCreation(tasks)
@@ -155,7 +160,8 @@ impl Batch {
IndexOperation { op, .. } => Some(op.index_uid()),
IndexCreation { index_uid, .. }
| IndexUpdate { index_uid, .. }
| IndexDeletion { index_uid, .. } => Some(index_uid),
| IndexDeletion { index_uid, .. }
| IndexCompaction { index_uid, .. } => Some(index_uid),
}
}
}
@@ -175,6 +181,7 @@ impl fmt::Display for Batch {
Batch::IndexUpdate { .. } => f.write_str("IndexUpdate")?,
Batch::IndexDeletion { .. } => f.write_str("IndexDeletion")?,
Batch::IndexSwap { .. } => f.write_str("IndexSwap")?,
Batch::IndexCompaction { .. } => f.write_str("IndexCompaction")?,
Batch::Export { .. } => f.write_str("Export")?,
Batch::UpgradeDatabase { .. } => f.write_str("UpgradeDatabase")?,
};
@@ -430,6 +437,12 @@ impl IndexScheduler {
current_batch.processing(Some(&mut task));
Ok(Some(Batch::IndexSwap { task }))
}
BatchKind::IndexCompaction { id } => {
let mut task =
self.queue.tasks.get_task(rtxn, id)?.ok_or(Error::CorruptedTaskQueue)?;
current_batch.processing(Some(&mut task));
Ok(Some(Batch::IndexCompaction { index_uid, task }))
}
}
}

View File

@@ -1,22 +1,26 @@
use std::collections::{BTreeSet, HashMap, HashSet};
use std::io::{Seek, SeekFrom};
use std::panic::{catch_unwind, AssertUnwindSafe};
use std::sync::atomic::Ordering;
use byte_unit::Byte;
use meilisearch_types::batches::{BatchEnqueuedAt, BatchId};
use meilisearch_types::heed::{RoTxn, RwTxn};
use meilisearch_types::milli::heed::CompactionOption;
use meilisearch_types::milli::progress::{Progress, VariableNameStep};
use meilisearch_types::milli::{self, ChannelCongestion};
use meilisearch_types::tasks::{Details, IndexSwap, Kind, KindWithContent, Status, Task};
use meilisearch_types::versioning::{VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
use milli::update::Settings as MilliSettings;
use roaring::RoaringBitmap;
use tempfile::PersistError;
use time::OffsetDateTime;
use super::create_batch::Batch;
use crate::processing::{
AtomicBatchStep, AtomicTaskStep, CreateIndexProgress, DeleteIndexProgress, FinalizingIndexStep,
InnerSwappingTwoIndexes, SwappingTheIndexes, TaskCancelationProgress, TaskDeletionProgress,
UpdateIndexProgress,
IndexCompaction, InnerSwappingTwoIndexes, SwappingTheIndexes, TaskCancelationProgress,
TaskDeletionProgress, UpdateIndexProgress,
};
use crate::utils::{
self, remove_n_tasks_datetime_earlier_than, remove_task_datetime, swap_index_uid_in_task,
@@ -418,6 +422,47 @@ impl IndexScheduler {
task.status = Status::Succeeded;
Ok((vec![task], ProcessBatchInfo::default()))
}
Batch::IndexCompaction { index_uid: _, mut task } => {
let KindWithContent::IndexCompaction { index_uid } = &task.kind else {
unreachable!()
};
let rtxn = self.env.read_txn()?;
let ret = catch_unwind(AssertUnwindSafe(|| {
self.apply_compaction(&rtxn, &progress, index_uid)
}));
let (pre_size, post_size) = match ret {
Ok(Ok(stats)) => stats,
Ok(Err(Error::AbortedTask)) => return Err(Error::AbortedTask),
Ok(Err(e)) => return Err(e),
Err(e) => {
let msg = match e.downcast_ref::<&'static str>() {
Some(s) => *s,
None => match e.downcast_ref::<String>() {
Some(s) => &s[..],
None => "Box<dyn Any>",
},
};
return Err(Error::Export(Box::new(Error::ProcessBatchPanicked(
msg.to_string(),
))));
}
};
task.status = Status::Succeeded;
if let Some(Details::IndexCompaction {
index_uid: _,
pre_compaction_size,
post_compaction_size,
}) = task.details.as_mut()
{
*pre_compaction_size = Some(Byte::from_u64(pre_size));
*post_compaction_size = Some(Byte::from_u64(post_size));
}
Ok((vec![task], ProcessBatchInfo::default()))
}
Batch::Export { mut task } => {
let KindWithContent::Export { url, api_key, payload_size, indexes } = &task.kind
else {
@@ -493,6 +538,91 @@ impl IndexScheduler {
}
}
fn apply_compaction(
&self,
rtxn: &RoTxn,
progress: &Progress,
index_uid: &str,
) -> Result<(u64, u64)> {
// 1. Verify that the index exists
if !self.index_mapper.index_exists(rtxn, index_uid)? {
return Err(Error::IndexNotFound(index_uid.to_owned()));
}
// 2. We retrieve the index and create a temporary file in the index directory
progress.update_progress(IndexCompaction::RetrieveTheIndex);
let index = self.index_mapper.index(rtxn, index_uid)?;
// the index operation can take a long time, so save this handle to make it available to the search for the duration of the tick
self.index_mapper
.set_currently_updating_index(Some((index_uid.to_string(), index.clone())));
progress.update_progress(IndexCompaction::CreateTemporaryFile);
let pre_size = std::fs::metadata(index.path().join("data.mdb"))?.len();
let mut file = tempfile::Builder::new()
.suffix("data.")
.prefix(".mdb.cpy")
.tempfile_in(index.path())?;
// 3. We copy the index data to the temporary file
progress.update_progress(IndexCompaction::CopyAndCompactTheIndex);
index
.copy_to_file(file.as_file_mut(), CompactionOption::Enabled)
.map_err(|error| Error::Milli { error, index_uid: Some(index_uid.to_string()) })?;
// ...and reset the file position as specified in the documentation
file.seek(SeekFrom::Start(0))?;
// 4. We replace the index data file with the temporary file
progress.update_progress(IndexCompaction::PersistTheCompactedIndex);
match file.persist(index.path().join("data.mdb")) {
Ok(file) => file.sync_all()?,
// TODO see if we have a _resource busy_ error and probably handle this by:
// 1. closing the index, 2. replacing and 3. reopening it
Err(PersistError { error, file: _ }) => return Err(Error::IoError(error)),
};
// 5. Prepare to close the index
progress.update_progress(IndexCompaction::CloseTheIndex);
// unmark that the index is the processing one so we don't keep a handle to it, preventing its closing
self.index_mapper.set_currently_updating_index(None);
self.index_mapper.close_index(rtxn, index_uid)?;
drop(index);
progress.update_progress(IndexCompaction::ReopenTheIndex);
// 6. Reopen the index
// The index will use the compacted data file when being reopened
let index = self.index_mapper.index(rtxn, index_uid)?;
// if the update processed successfully, we're going to store the new
// stats of the index. Since the tasks have already been processed and
// this is a non-critical operation. If it fails, we should not fail
// the entire batch.
let res = || -> Result<_> {
let mut wtxn = self.env.write_txn()?;
let index_rtxn = index.read_txn()?;
let stats = crate::index_mapper::IndexStats::new(&index, &index_rtxn)
.map_err(|e| Error::from_milli(e, Some(index_uid.to_string())))?;
self.index_mapper.store_stats_of(&mut wtxn, index_uid, &stats)?;
wtxn.commit()?;
Ok(stats.database_size)
}();
let post_size = match res {
Ok(post_size) => post_size,
Err(e) => {
tracing::error!(
error = &e as &dyn std::error::Error,
"Could not write the stats of the index"
);
0
}
};
Ok((pre_size, post_size))
}
/// Swap the index `lhs` with the index `rhs`.
fn apply_index_swap(
&self,

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 23, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, batch_uid: 1, status: succeeded, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
2 {uid: 2, batch_uid: 2, status: succeeded, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
3 {uid: 3, batch_uid: 3, status: failed, error: ResponseError { code: 200, message: "Index `doggo` already exists.", error_code: "index_already_exists", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#index_already_exists" }, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
@@ -57,7 +57,7 @@ girafo: { number_of_documents: 0, field_distribution: {} }
[timestamp] [4,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.23.0"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
1 {uid: 1, details: {"primaryKey":"mouse"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"indexCreation":1},"indexUids":{"catto":1}}, stop reason: "created batch containing only task with id 1 of type `indexCreation` that cannot be batched with any other task.", }
2 {uid: 2, details: {"primaryKey":"bone"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"indexCreation":1},"indexUids":{"doggo":1}}, stop reason: "created batch containing only task with id 2 of type `indexCreation` that cannot be batched with any other task.", }
3 {uid: 3, details: {"primaryKey":"bone"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"indexCreation":1},"indexUids":{"doggo":1}}, stop reason: "created batch containing only task with id 3 of type `indexCreation` that cannot be batched with any other task.", }

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 23, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
----------------------------------------------------------------------
### Status:
enqueued [0,]

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 23, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
----------------------------------------------------------------------
### Status:

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 23, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
----------------------------------------------------------------------
### Status:
@@ -37,7 +37,7 @@ catto [1,]
[timestamp] [0,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.23.0"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
----------------------------------------------------------------------
### Batch to tasks mapping:
0 [0,]

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 23, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
2 {uid: 2, status: enqueued, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
----------------------------------------------------------------------
@@ -40,7 +40,7 @@ doggo [2,]
[timestamp] [0,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.23.0"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
----------------------------------------------------------------------
### Batch to tasks mapping:
0 [0,]

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 23, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
2 {uid: 2, status: enqueued, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
3 {uid: 3, status: enqueued, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
@@ -43,7 +43,7 @@ doggo [2,3,]
[timestamp] [0,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.23.0"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
----------------------------------------------------------------------
### Batch to tasks mapping:
0 [0,]

View File

@@ -722,7 +722,7 @@ fn basic_get_stats() {
let kind = index_creation_task("whalo", "fish");
let _task = index_scheduler.register(kind, None, false).unwrap();
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r#"
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r###"
{
"indexes": {
"catto": 1,
@@ -742,6 +742,7 @@ fn basic_get_stats() {
"documentEdition": 0,
"dumpCreation": 0,
"export": 0,
"indexCompaction": 0,
"indexCreation": 3,
"indexDeletion": 0,
"indexSwap": 0,
@@ -753,10 +754,10 @@ fn basic_get_stats() {
"upgradeDatabase": 0
}
}
"#);
"###);
handle.advance_till([Start, BatchCreated]);
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r#"
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r###"
{
"indexes": {
"catto": 1,
@@ -776,6 +777,7 @@ fn basic_get_stats() {
"documentEdition": 0,
"dumpCreation": 0,
"export": 0,
"indexCompaction": 0,
"indexCreation": 3,
"indexDeletion": 0,
"indexSwap": 0,
@@ -787,7 +789,7 @@ fn basic_get_stats() {
"upgradeDatabase": 0
}
}
"#);
"###);
handle.advance_till([
InsideProcessBatch,
@@ -797,7 +799,7 @@ fn basic_get_stats() {
Start,
BatchCreated,
]);
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r#"
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r###"
{
"indexes": {
"catto": 1,
@@ -817,6 +819,7 @@ fn basic_get_stats() {
"documentEdition": 0,
"dumpCreation": 0,
"export": 0,
"indexCompaction": 0,
"indexCreation": 3,
"indexDeletion": 0,
"indexSwap": 0,
@@ -828,7 +831,7 @@ fn basic_get_stats() {
"upgradeDatabase": 0
}
}
"#);
"###);
// now we make one more batch, the started_at field of the new tasks will be past `second_start_time`
handle.advance_till([
@@ -839,7 +842,7 @@ fn basic_get_stats() {
Start,
BatchCreated,
]);
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r#"
snapshot!(json_string!(index_scheduler.get_stats().unwrap()), @r###"
{
"indexes": {
"catto": 1,
@@ -859,6 +862,7 @@ fn basic_get_stats() {
"documentEdition": 0,
"dumpCreation": 0,
"export": 0,
"indexCompaction": 0,
"indexCreation": 3,
"indexDeletion": 0,
"indexSwap": 0,
@@ -870,7 +874,7 @@ fn basic_get_stats() {
"upgradeDatabase": 0
}
}
"#);
"###);
}
#[test]

View File

@@ -46,6 +46,7 @@ pub fn upgrade_index_scheduler(
(1, 20, _) => 0,
(1, 21, _) => 0,
(1, 22, _) => 0,
(1, 23, _) => 0,
(major, minor, patch) => {
if major > current_major
|| (major == current_major && minor > current_minor)

View File

@@ -256,14 +256,15 @@ pub fn swap_index_uid_in_task(task: &mut Task, swap: (&str, &str)) {
use KindWithContent as K;
let mut index_uids = vec![];
match &mut task.kind {
K::DocumentAdditionOrUpdate { index_uid, .. } => index_uids.push(index_uid),
K::DocumentEdition { index_uid, .. } => index_uids.push(index_uid),
K::DocumentDeletion { index_uid, .. } => index_uids.push(index_uid),
K::DocumentDeletionByFilter { index_uid, .. } => index_uids.push(index_uid),
K::DocumentClear { index_uid } => index_uids.push(index_uid),
K::SettingsUpdate { index_uid, .. } => index_uids.push(index_uid),
K::IndexDeletion { index_uid } => index_uids.push(index_uid),
K::IndexCreation { index_uid, .. } => index_uids.push(index_uid),
K::DocumentAdditionOrUpdate { index_uid, .. }
| K::DocumentEdition { index_uid, .. }
| K::DocumentDeletion { index_uid, .. }
| K::DocumentDeletionByFilter { index_uid, .. }
| K::DocumentClear { index_uid }
| K::SettingsUpdate { index_uid, .. }
| K::IndexDeletion { index_uid }
| K::IndexCreation { index_uid, .. }
| K::IndexCompaction { index_uid, .. } => index_uids.push(index_uid),
K::IndexUpdate { index_uid, new_index_uid, .. } => {
index_uids.push(index_uid);
if let Some(new_uid) = new_index_uid {
@@ -618,6 +619,13 @@ impl crate::IndexScheduler {
Details::UpgradeDatabase { from: _, to: _ } => {
assert_eq!(kind.as_kind(), Kind::UpgradeDatabase);
}
Details::IndexCompaction {
index_uid: _,
pre_compaction_size: _,
post_compaction_size: _,
} => {
assert_eq!(kind.as_kind(), Kind::IndexCompaction);
}
}
}

View File

@@ -109,6 +109,7 @@ impl HeedAuthStore {
Action::IndexesGet,
Action::IndexesUpdate,
Action::IndexesSwap,
Action::IndexesCompact,
]
.iter(),
);

View File

@@ -380,6 +380,9 @@ pub enum Action {
#[serde(rename = "webhooks.*")]
#[deserr(rename = "webhooks.*")]
WebhooksAll,
#[serde(rename = "indexes.compact")]
#[deserr(rename = "indexes.compact")]
IndexesCompact,
}
impl Action {
@@ -398,6 +401,7 @@ impl Action {
INDEXES_UPDATE => Some(Self::IndexesUpdate),
INDEXES_DELETE => Some(Self::IndexesDelete),
INDEXES_SWAP => Some(Self::IndexesSwap),
INDEXES_COMPACT => Some(Self::IndexesCompact),
TASKS_ALL => Some(Self::TasksAll),
TASKS_CANCEL => Some(Self::TasksCancel),
TASKS_DELETE => Some(Self::TasksDelete),
@@ -462,6 +466,7 @@ impl Action {
IndexesUpdate => false,
IndexesDelete => false,
IndexesSwap => false,
IndexesCompact => false,
TasksCancel => false,
TasksDelete => false,
TasksGet => true,
@@ -513,6 +518,7 @@ pub mod actions {
pub const INDEXES_UPDATE: u8 = IndexesUpdate.repr();
pub const INDEXES_DELETE: u8 = IndexesDelete.repr();
pub const INDEXES_SWAP: u8 = IndexesSwap.repr();
pub const INDEXES_COMPACT: u8 = IndexesCompact.repr();
pub const TASKS_ALL: u8 = TasksAll.repr();
pub const TASKS_CANCEL: u8 = TasksCancel.repr();
pub const TASKS_DELETE: u8 = TasksDelete.repr();
@@ -614,6 +620,7 @@ pub(crate) mod test {
assert!(WebhooksDelete.repr() == 47 && WEBHOOKS_DELETE == 47);
assert!(WebhooksCreate.repr() == 48 && WEBHOOKS_CREATE == 48);
assert!(WebhooksAll.repr() == 49 && WEBHOOKS_ALL == 49);
assert!(IndexesCompact.repr() == 50 && INDEXES_COMPACT == 50);
}
#[test]

View File

@@ -142,6 +142,11 @@ pub struct DetailsView {
pub old_index_uid: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub new_index_uid: Option<String>,
// index compaction
#[serde(skip_serializing_if = "Option::is_none")]
pub pre_compaction_size: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub post_compaction_size: Option<String>,
}
impl DetailsView {
@@ -314,6 +319,24 @@ impl DetailsView {
// We should never be able to batch multiple renames at the same time.
(Some(left), Some(_right)) => Some(left),
},
pre_compaction_size: match (
self.pre_compaction_size.clone(),
other.pre_compaction_size.clone(),
) {
(None, None) => None,
(None, Some(size)) | (Some(size), None) => Some(size),
// We should never be able to batch multiple compactions at the same time.
(Some(left), Some(_right)) => Some(left),
},
post_compaction_size: match (
self.post_compaction_size.clone(),
other.post_compaction_size.clone(),
) {
(None, None) => None,
(None, Some(size)) | (Some(size), None) => Some(size),
// We should never be able to batch multiple compactions at the same time.
(Some(left), Some(_right)) => Some(left),
},
}
}
}
@@ -415,6 +438,15 @@ impl From<Details> for DetailsView {
upgrade_to: Some(format!("v{}.{}.{}", to.0, to.1, to.2)),
..Default::default()
},
Details::IndexCompaction { pre_compaction_size, post_compaction_size, .. } => {
DetailsView {
pre_compaction_size: pre_compaction_size
.map(|size| size.get_appropriate_unit(UnitType::Both).to_string()),
post_compaction_size: post_compaction_size
.map(|size| size.get_appropriate_unit(UnitType::Both).to_string()),
..Default::default()
}
}
}
}
}

View File

@@ -67,7 +67,8 @@ impl Task {
| SettingsUpdate { index_uid, .. }
| IndexCreation { index_uid, .. }
| IndexUpdate { index_uid, .. }
| IndexDeletion { index_uid } => Some(index_uid),
| IndexDeletion { index_uid }
| IndexCompaction { index_uid } => Some(index_uid),
}
}
@@ -94,7 +95,8 @@ impl Task {
| KindWithContent::DumpCreation { .. }
| KindWithContent::SnapshotCreation
| KindWithContent::Export { .. }
| KindWithContent::UpgradeDatabase { .. } => None,
| KindWithContent::UpgradeDatabase { .. }
| KindWithContent::IndexCompaction { .. } => None,
}
}
}
@@ -170,6 +172,9 @@ pub enum KindWithContent {
UpgradeDatabase {
from: (u32, u32, u32),
},
IndexCompaction {
index_uid: String,
},
}
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize, ToSchema)]
@@ -206,6 +211,7 @@ impl KindWithContent {
KindWithContent::SnapshotCreation => Kind::SnapshotCreation,
KindWithContent::Export { .. } => Kind::Export,
KindWithContent::UpgradeDatabase { .. } => Kind::UpgradeDatabase,
KindWithContent::IndexCompaction { .. } => Kind::IndexCompaction,
}
}
@@ -226,7 +232,8 @@ impl KindWithContent {
| DocumentClear { index_uid }
| SettingsUpdate { index_uid, .. }
| IndexCreation { index_uid, .. }
| IndexDeletion { index_uid } => vec![index_uid],
| IndexDeletion { index_uid }
| IndexCompaction { index_uid } => vec![index_uid],
IndexUpdate { index_uid, new_index_uid, .. } => {
let mut indexes = vec![index_uid.as_str()];
if let Some(new_uid) = new_index_uid {
@@ -325,6 +332,11 @@ impl KindWithContent {
versioning::VERSION_PATCH,
),
}),
KindWithContent::IndexCompaction { index_uid } => Some(Details::IndexCompaction {
index_uid: index_uid.clone(),
pre_compaction_size: None,
post_compaction_size: None,
}),
}
}
@@ -407,6 +419,11 @@ impl KindWithContent {
versioning::VERSION_PATCH,
),
}),
KindWithContent::IndexCompaction { index_uid } => Some(Details::IndexCompaction {
index_uid: index_uid.clone(),
pre_compaction_size: None,
post_compaction_size: None,
}),
}
}
}
@@ -469,6 +486,11 @@ impl From<&KindWithContent> for Option<Details> {
versioning::VERSION_PATCH,
),
}),
KindWithContent::IndexCompaction { index_uid } => Some(Details::IndexCompaction {
index_uid: index_uid.clone(),
pre_compaction_size: None,
post_compaction_size: None,
}),
}
}
}
@@ -579,6 +601,7 @@ pub enum Kind {
SnapshotCreation,
Export,
UpgradeDatabase,
IndexCompaction,
}
impl Kind {
@@ -590,7 +613,8 @@ impl Kind {
| Kind::SettingsUpdate
| Kind::IndexCreation
| Kind::IndexDeletion
| Kind::IndexUpdate => true,
| Kind::IndexUpdate
| Kind::IndexCompaction => true,
Kind::IndexSwap
| Kind::TaskCancelation
| Kind::TaskDeletion
@@ -618,6 +642,7 @@ impl Display for Kind {
Kind::SnapshotCreation => write!(f, "snapshotCreation"),
Kind::Export => write!(f, "export"),
Kind::UpgradeDatabase => write!(f, "upgradeDatabase"),
Kind::IndexCompaction => write!(f, "indexCompaction"),
}
}
}
@@ -653,6 +678,8 @@ impl FromStr for Kind {
Ok(Kind::Export)
} else if kind.eq_ignore_ascii_case("upgradeDatabase") {
Ok(Kind::UpgradeDatabase)
} else if kind.eq_ignore_ascii_case("indexCompaction") {
Ok(Kind::IndexCompaction)
} else {
Err(ParseTaskKindError(kind.to_owned()))
}
@@ -738,6 +765,11 @@ pub enum Details {
from: (u32, u32, u32),
to: (u32, u32, u32),
},
IndexCompaction {
index_uid: String,
pre_compaction_size: Option<Byte>,
post_compaction_size: Option<Byte>,
},
}
#[derive(Debug, PartialEq, Clone, Serialize, Deserialize, ToSchema)]
@@ -800,6 +832,10 @@ impl Details {
Self::ClearAll { deleted_documents } => *deleted_documents = Some(0),
Self::TaskCancelation { canceled_tasks, .. } => *canceled_tasks = Some(0),
Self::TaskDeletion { deleted_tasks, .. } => *deleted_tasks = Some(0),
Self::IndexCompaction { pre_compaction_size, post_compaction_size, .. } => {
*pre_compaction_size = None;
*post_compaction_size = None;
}
Self::SettingsUpdate { .. }
| Self::IndexInfo { .. }
| Self::Dump { .. }

View File

@@ -0,0 +1,84 @@
use actix_web::web::{self, Data};
use actix_web::{HttpRequest, HttpResponse};
use index_scheduler::IndexScheduler;
use meilisearch_types::error::ResponseError;
use meilisearch_types::index_uid::IndexUid;
use meilisearch_types::keys::actions;
use meilisearch_types::tasks::KindWithContent;
use tracing::debug;
use utoipa::OpenApi;
use super::ActionPolicy;
use crate::analytics::Analytics;
use crate::extractors::authentication::GuardedData;
use crate::extractors::sequential_extractor::SeqHandler;
use crate::routes::SummarizedTaskView;
#[derive(OpenApi)]
#[openapi(
paths(compact),
tags(
(
name = "Compact an index",
description = "The /compact route uses compacts the database to reorganize and make it smaller and more efficient.",
external_docs(url = "https://www.meilisearch.com/docs/reference/api/compact"),
),
),
)]
pub struct CompactApi;
pub fn configure(cfg: &mut web::ServiceConfig) {
cfg.service(web::resource("").route(web::post().to(SeqHandler(compact))));
}
/// Compact an index
#[utoipa::path(
post,
path = "{indexUid}/compact",
tag = "Compact an index",
security(("Bearer" = ["search", "*"])),
params(("indexUid" = String, Path, example = "movies", description = "Index Unique Identifier", nullable = false)),
responses(
(status = ACCEPTED, description = "Task successfully enqueued", body = SummarizedTaskView, content_type = "application/json", example = json!(
{
"taskUid": 147,
"indexUid": null,
"status": "enqueued",
"type": "documentDeletion",
"enqueuedAt": "2024-08-08T17:05:55.791772Z"
}
)),
(status = 401, description = "The authorization header is missing", body = ResponseError, content_type = "application/json", example = json!(
{
"message": "The Authorization header is missing. It must use the bearer authorization method.",
"code": "missing_authorization_header",
"type": "auth",
"link": "https://docs.meilisearch.com/errors#missing_authorization_header"
}
)),
)
)]
pub async fn compact(
index_scheduler: GuardedData<ActionPolicy<{ actions::INDEXES_COMPACT }>, Data<IndexScheduler>>,
index_uid: web::Path<String>,
req: HttpRequest,
analytics: web::Data<Analytics>,
) -> Result<HttpResponse, ResponseError> {
let index_uid = IndexUid::try_from(index_uid.into_inner())?;
analytics.publish(IndexCompacted::default(), &req);
let task = KindWithContent::IndexCompaction { index_uid: index_uid.to_string() };
let task =
match tokio::task::spawn_blocking(move || index_scheduler.register(task, None, false))
.await?
{
Ok(task) => task,
Err(e) => return Err(e.into()),
};
debug!(returns = ?task, "Compact the {index_uid} index");
Ok(HttpResponse::Accepted().json(SummarizedTaskView::from(task)))
}
crate::empty_analytics!(IndexCompacted, "Index Compacted");

View File

@@ -28,6 +28,7 @@ use crate::extractors::sequential_extractor::SeqHandler;
use crate::routes::is_dry_run;
use crate::Opt;
pub mod compact;
pub mod documents;
mod enterprise_edition;
pub mod facet_search;
@@ -49,8 +50,9 @@ pub use enterprise_edition::proxy::{PROXY_ORIGIN_REMOTE_HEADER, PROXY_ORIGIN_TAS
(path = "/", api = facet_search::FacetSearchApi),
(path = "/", api = similar::SimilarApi),
(path = "/", api = settings::SettingsApi),
(path = "/", api = compact::CompactApi),
),
paths(list_indexes, create_index, get_index, update_index, delete_index, get_index_stats),
paths(list_indexes, create_index, get_index, update_index, delete_index, get_index_stats, compact::compact),
tags(
(
name = "Indexes",
@@ -80,7 +82,8 @@ pub fn configure(cfg: &mut web::ServiceConfig) {
.service(web::scope("/search").configure(search::configure))
.service(web::scope("/facet-search").configure(facet_search::configure))
.service(web::scope("/similar").configure(similar::configure))
.service(web::scope("/settings").configure(settings::configure)),
.service(web::scope("/settings").configure(settings::configure))
.service(web::scope("/compact").configure(compact::configure)),
);
}

View File

@@ -226,14 +226,14 @@ mod tests {
{
let params = "types=createIndex";
let err = deserr_query_params::<TaskDeletionOrCancelationQuery>(params).unwrap_err();
snapshot!(meili_snap::json_string!(err), @r#"
snapshot!(meili_snap::json_string!(err), @r###"
{
"message": "Invalid value in parameter `types`: `createIndex` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`.",
"message": "Invalid value in parameter `types`: `createIndex` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`, `indexCompaction`.",
"code": "invalid_task_types",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_task_types"
}
"#);
"###);
}
}
#[test]

View File

@@ -46,7 +46,13 @@ pub async fn perform_federated_search(
features.check_network("Performing a remote federated search")?;
}
let before_search = std::time::Instant::now();
let deadline = before_search + std::time::Duration::from_secs(9);
let timeout = std::env::var("MEILI_EXPERIMENTAL_REMOTE_SEARCH_TIMEOUT_SECONDS")
.ok()
.map(|p| p.parse().unwrap())
.unwrap_or(25);
let deadline = before_search + std::time::Duration::from_secs(timeout);
let required_hit_count = federation.limit + federation.offset;
let retrieve_vectors = queries.iter().any(|q| q.retrieve_vectors);

View File

@@ -105,7 +105,12 @@ pub async fn proxy_search(
let search_api_key = node.search_api_key.as_deref();
let max_deadline = std::time::Instant::now() + std::time::Duration::from_secs(5);
let timeout = std::env::var("MEILI_EXPERIMENTAL_REMOTE_SEARCH_TIMEOUT_SECONDS")
.ok()
.map(|p| p.parse().unwrap())
.unwrap_or(25);
let max_deadline = std::time::Instant::now() + std::time::Duration::from_secs(timeout);
let deadline = if let Some(deadline) = params.deadline {
std::time::Instant::min(deadline, max_deadline)

View File

@@ -419,14 +419,14 @@ async fn error_add_api_key_invalid_parameters_actions() {
let (response, code) = server.add_api_key(content).await;
meili_snap::snapshot!(code, @"400 Bad Request");
meili_snap::snapshot!(meili_snap::json_string!(response, { ".createdAt" => "[ignored]", ".updatedAt" => "[ignored]" }), @r#"
meili_snap::snapshot!(meili_snap::json_string!(response, { ".createdAt" => "[ignored]", ".updatedAt" => "[ignored]" }), @r###"
{
"message": "Unknown value `doc.add` at `.actions[0]`: expected one of `*`, `search`, `documents.*`, `documents.add`, `documents.get`, `documents.delete`, `indexes.*`, `indexes.create`, `indexes.get`, `indexes.update`, `indexes.delete`, `indexes.swap`, `tasks.*`, `tasks.cancel`, `tasks.delete`, `tasks.get`, `settings.*`, `settings.get`, `settings.update`, `stats.*`, `stats.get`, `metrics.*`, `metrics.get`, `dumps.*`, `dumps.create`, `snapshots.*`, `snapshots.create`, `version`, `keys.create`, `keys.get`, `keys.update`, `keys.delete`, `experimental.get`, `experimental.update`, `export`, `network.get`, `network.update`, `chatCompletions`, `chats.*`, `chats.get`, `chats.delete`, `chatsSettings.*`, `chatsSettings.get`, `chatsSettings.update`, `*.get`, `webhooks.get`, `webhooks.update`, `webhooks.delete`, `webhooks.create`, `webhooks.*`",
"message": "Unknown value `doc.add` at `.actions[0]`: expected one of `*`, `search`, `documents.*`, `documents.add`, `documents.get`, `documents.delete`, `indexes.*`, `indexes.create`, `indexes.get`, `indexes.update`, `indexes.delete`, `indexes.swap`, `tasks.*`, `tasks.cancel`, `tasks.delete`, `tasks.get`, `settings.*`, `settings.get`, `settings.update`, `stats.*`, `stats.get`, `metrics.*`, `metrics.get`, `dumps.*`, `dumps.create`, `snapshots.*`, `snapshots.create`, `version`, `keys.create`, `keys.get`, `keys.update`, `keys.delete`, `experimental.get`, `experimental.update`, `export`, `network.get`, `network.update`, `chatCompletions`, `chats.*`, `chats.get`, `chats.delete`, `chatsSettings.*`, `chatsSettings.get`, `chatsSettings.update`, `*.get`, `webhooks.get`, `webhooks.update`, `webhooks.delete`, `webhooks.create`, `webhooks.*`, `indexes.compact`",
"code": "invalid_api_key_actions",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_api_key_actions"
}
"#);
"###);
}
#[actix_rt::test]

View File

@@ -91,14 +91,14 @@ async fn create_api_key_bad_actions() {
// can't parse
let (response, code) = server.add_api_key(json!({ "actions": ["doggo"] })).await;
snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r#"
snapshot!(json_string!(response), @r###"
{
"message": "Unknown value `doggo` at `.actions[0]`: expected one of `*`, `search`, `documents.*`, `documents.add`, `documents.get`, `documents.delete`, `indexes.*`, `indexes.create`, `indexes.get`, `indexes.update`, `indexes.delete`, `indexes.swap`, `tasks.*`, `tasks.cancel`, `tasks.delete`, `tasks.get`, `settings.*`, `settings.get`, `settings.update`, `stats.*`, `stats.get`, `metrics.*`, `metrics.get`, `dumps.*`, `dumps.create`, `snapshots.*`, `snapshots.create`, `version`, `keys.create`, `keys.get`, `keys.update`, `keys.delete`, `experimental.get`, `experimental.update`, `export`, `network.get`, `network.update`, `chatCompletions`, `chats.*`, `chats.get`, `chats.delete`, `chatsSettings.*`, `chatsSettings.get`, `chatsSettings.update`, `*.get`, `webhooks.get`, `webhooks.update`, `webhooks.delete`, `webhooks.create`, `webhooks.*`",
"message": "Unknown value `doggo` at `.actions[0]`: expected one of `*`, `search`, `documents.*`, `documents.add`, `documents.get`, `documents.delete`, `indexes.*`, `indexes.create`, `indexes.get`, `indexes.update`, `indexes.delete`, `indexes.swap`, `tasks.*`, `tasks.cancel`, `tasks.delete`, `tasks.get`, `settings.*`, `settings.get`, `settings.update`, `stats.*`, `stats.get`, `metrics.*`, `metrics.get`, `dumps.*`, `dumps.create`, `snapshots.*`, `snapshots.create`, `version`, `keys.create`, `keys.get`, `keys.update`, `keys.delete`, `experimental.get`, `experimental.update`, `export`, `network.get`, `network.update`, `chatCompletions`, `chats.*`, `chats.get`, `chats.delete`, `chatsSettings.*`, `chatsSettings.get`, `chatsSettings.update`, `*.get`, `webhooks.get`, `webhooks.update`, `webhooks.delete`, `webhooks.create`, `webhooks.*`, `indexes.compact`",
"code": "invalid_api_key_actions",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_api_key_actions"
}
"#);
"###);
}
#[actix_rt::test]

View File

@@ -40,14 +40,14 @@ async fn batch_bad_types() {
let (response, code) = server.batches_filter("types=doggo").await;
snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r#"
snapshot!(json_string!(response), @r###"
{
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`.",
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`, `indexCompaction`.",
"code": "invalid_task_types",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_task_types"
}
"#);
"###);
}
#[actix_rt::test]

View File

@@ -2128,3 +2128,102 @@ async fn simple_search_changing_unrelated_settings() {
})
.await;
}
#[actix_rt::test]
async fn ranking_score_bug_with_sort() {
let server = Server::new_shared();
let index = server.unique_index();
// Create documents with a "created" field for sorting
let documents = json!([
{
"id": "1",
"title": "Coffee Mug",
"created": "2023-01-01T00:00:00Z"
},
{
"id": "2",
"title": "Water Bottle",
"created": "2023-01-02T00:00:00Z"
},
{
"id": "3",
"title": "Tumbler Cup",
"created": "2023-01-03T00:00:00Z"
},
{
"id": "4",
"title": "Stainless Steel Tumbler",
"created": "2023-01-04T00:00:00Z"
}
]);
// Add documents
let (task, code) = index.add_documents(documents, None).await;
assert_eq!(code, 202, "{task}");
server.wait_task(task.uid()).await.succeeded();
// Configure sortable attributes
let (task, code) = index
.update_settings(json!({
"sortableAttributes": ["created"]
}))
.await;
assert_eq!(code, 202, "{task}");
server.wait_task(task.uid()).await.succeeded();
// Test 1: Search without sort - should have proper ranking scores
index
.search(
json!({
"q": "tumbler",
"showRankingScore": true,
"rankingScoreThreshold": 0.0,
"attributesToRetrieve": ["title"]
}),
|response, code| {
assert_eq!(code, 200, "{response}");
snapshot!(json_string!(response["hits"]), @r###"
[
{
"title": "Tumbler Cup",
"_rankingScore": 0.9848484848484848
},
{
"title": "Stainless Steel Tumbler",
"_rankingScore": 0.8787878787878788
}
]
"###);
},
)
.await;
// Test 2: Search with sort - this is where the bug occurs
index
.search(
json!({
"q": "tumbler",
"showRankingScore": true,
"rankingScoreThreshold": 0.0,
"sort": ["created:desc"],
"attributesToRetrieve": ["title"]
}),
|response, code| {
assert_eq!(code, 200, "{response}");
snapshot!(json_string!(response["hits"]), @r###"
[
{
"title": "Tumbler Cup",
"_rankingScore": 0.9848484848484848
},
{
"title": "Stainless Steel Tumbler",
"_rankingScore": 0.8787878787878788
}
]
"###);
},
)
.await;
}

View File

@@ -2866,7 +2866,8 @@ async fn error_remote_timeout() {
snapshot!(json_string!(response), @r###"
{
"self": "ms0",
"remotes": {}
"remotes": {},
"sharding": false
}
"###);
let (response, code) = ms1.set_network(json!({"self": "ms1"})).await;
@@ -2874,7 +2875,8 @@ async fn error_remote_timeout() {
snapshot!(json_string!(response), @r###"
{
"self": "ms1",
"remotes": {}
"remotes": {},
"sharding": false
}
"###);
@@ -2895,7 +2897,7 @@ async fn error_remote_timeout() {
let rms0 = LocalMeili::new(ms0.clone()).await;
let rms1 = LocalMeili::with_params(
ms1.clone(),
LocalMeiliParams { delay: Some(std::time::Duration::from_secs(6)), ..Default::default() },
LocalMeiliParams { delay: Some(std::time::Duration::from_secs(31)), ..Default::default() },
)
.await;

View File

@@ -97,7 +97,7 @@ async fn task_bad_types() {
snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r#"
{
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`.",
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`, `indexCompaction`.",
"code": "invalid_task_types",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_task_types"
@@ -108,7 +108,7 @@ async fn task_bad_types() {
snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r#"
{
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`.",
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`, `indexCompaction`.",
"code": "invalid_task_types",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_task_types"
@@ -119,7 +119,7 @@ async fn task_bad_types() {
snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r#"
{
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`.",
"message": "Invalid value in parameter `types`: `doggo` is not a valid task type. Available types are `documentAdditionOrUpdate`, `documentEdition`, `documentDeletion`, `settingsUpdate`, `indexCreation`, `indexDeletion`, `indexUpdate`, `indexSwap`, `taskCancelation`, `taskDeletion`, `dumpCreation`, `snapshotCreation`, `export`, `upgradeDatabase`, `indexCompaction`.",
"code": "invalid_task_types",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_task_types"

View File

@@ -43,7 +43,7 @@ async fn version_too_old() {
std::fs::write(db_path.join("VERSION"), "1.11.9999").unwrap();
let options = Opt { experimental_dumpless_upgrade: true, ..default_settings };
let err = Server::new_with_options(options).await.map(|_| ()).unwrap_err();
snapshot!(err, @"Database version 1.11.9999 is too old for the experimental dumpless upgrade feature. Please generate a dump using the v1.11.9999 and import it in the v1.22.1");
snapshot!(err, @"Database version 1.11.9999 is too old for the experimental dumpless upgrade feature. Please generate a dump using the v1.11.9999 and import it in the v1.23.0");
}
#[actix_rt::test]
@@ -58,7 +58,7 @@ async fn version_requires_downgrade() {
std::fs::write(db_path.join("VERSION"), format!("{major}.{minor}.{patch}")).unwrap();
let options = Opt { experimental_dumpless_upgrade: true, ..default_settings };
let err = Server::new_with_options(options).await.map(|_| ()).unwrap_err();
snapshot!(err, @"Database version 1.22.2 is higher than the Meilisearch version 1.22.1. Downgrade is not supported");
snapshot!(err, @"Database version 1.23.1 is higher than the Meilisearch version 1.23.0. Downgrade is not supported");
}
#[actix_rt::test]

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"stats": {
"totalNbTasks": 1,

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"stats": {
"totalNbTasks": 1,

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"stats": {
"totalNbTasks": 1,

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"stats": {
"totalNbTasks": 1,

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.22.1"
"upgradeTo": "v1.23.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -126,7 +126,7 @@ enum Command {
/// before running the copy and compaction. This way the current indexation must finish before
/// the compaction operation can start. Once the compaction is done, the big index is replaced
/// by the compacted one and the mutable transaction is released.
CompactIndex { index_name: String },
IndexCompaction { index_name: String },
/// Uses the hair dryer the dedicate pages hot in cache
///
@@ -165,7 +165,7 @@ fn main() -> anyhow::Result<()> {
let target_version = parse_version(&target_version).context("While parsing `--target-version`. Make sure `--target-version` is in the format MAJOR.MINOR.PATCH")?;
OfflineUpgrade { db_path, current_version: detected_version, target_version }.upgrade()
}
Command::CompactIndex { index_name } => compact_index(db_path, &index_name),
Command::IndexCompaction { index_name } => compact_index(db_path, &index_name),
Command::HairDryer { index_name, index_part } => {
hair_dryer(db_path, &index_name, &index_part)
}

View File

@@ -19,7 +19,7 @@ bstr = "1.12.0"
bytemuck = { version = "1.23.1", features = ["extern_crate_alloc"] }
byteorder = "1.5.0"
charabia = { version = "0.9.7", default-features = false }
cellulite = "0.3.1-nested-rtxns"
cellulite = "0.3.1-nested-rtxns-2"
concat-arrays = "0.1.2"
convert_case = "0.8.0"
crossbeam-channel = "0.5.15"

View File

@@ -66,15 +66,29 @@ impl ScoreDetails {
}
}
/// Calculate the global score of the details.
///
/// It is computed from the ranks of the ranking rules, excluding the sort/geo sort rules.
/// If the details contain a semantic score (ScoreDetails::Vector), it is used instead of the ranking score.
///
/// note: this function expects a maximum of one semantic score, otherwise only the last one will be used.
pub fn global_score<'a>(details: impl Iterator<Item = &'a Self> + 'a) -> f64 {
Self::score_values(details)
.find_map(|x| {
let ScoreValue::Score(score) = x else {
return None;
};
Some(score)
})
.unwrap_or(1.0f64)
// Filter out only the ranking scores (Rank values) and exclude sort/geo sort
let mut semantic_score = None;
let ranking_ranks = details.filter_map(|detail| match detail.rank_or_value() {
RankOrValue::Rank(rank) => Some(rank),
RankOrValue::Score(score) => {
semantic_score = Some(score);
None
}
RankOrValue::Sort(_) => None,
RankOrValue::GeoSort(_) => None,
});
let ranking_score = Rank::global_score(ranking_ranks);
// If we have semantic score, use it, otherwise use ranking score
semantic_score.unwrap_or(ranking_score)
}
pub fn score_values<'a>(

View File

@@ -21,7 +21,7 @@ mod vector_sort;
#[cfg(test)]
mod tests;
use std::collections::HashSet;
use std::collections::{HashMap, HashSet};
use std::ops::AddAssign;
use std::time::Duration;
@@ -64,6 +64,12 @@ use crate::{
UserError, Weight,
};
/// Cache for synonyms to avoid repeated database access
#[derive(Default)]
pub struct SynonymCache {
pub cache: Option<HashMap<Vec<String>, Vec<Vec<String>>>>,
}
/// A structure used throughout the execution of a search query.
pub struct SearchContext<'ctx> {
pub index: &'ctx Index,
@@ -73,6 +79,7 @@ pub struct SearchContext<'ctx> {
pub phrase_interner: DedupInterner<Phrase>,
pub term_interner: Interner<QueryTerm>,
pub phrase_docids: PhraseDocIdsCache,
pub synonym_cache: SynonymCache,
pub restricted_fids: Option<RestrictedFids>,
pub prefix_search: PrefixSearch,
pub vector_store_stats: Option<VectorStoreStats>,
@@ -103,6 +110,7 @@ impl<'ctx> SearchContext<'ctx> {
phrase_interner: <_>::default(),
term_interner: <_>::default(),
phrase_docids: <_>::default(),
synonym_cache: <_>::default(),
restricted_fids: None,
prefix_search,
vector_store_stats: None,
@@ -113,6 +121,17 @@ impl<'ctx> SearchContext<'ctx> {
self.prefix_search != PrefixSearch::Disabled
}
/// Get synonyms with caching to avoid repeated database access
pub fn get_synonyms(&mut self) -> Result<&HashMap<Vec<String>, Vec<Vec<String>>>> {
match self.synonym_cache.cache {
Some(ref synonyms) => Ok(synonyms),
None => {
let synonyms = self.index.synonyms(self.txn)?;
Ok(self.synonym_cache.cache.insert(synonyms))
}
}
}
pub fn attributes_to_search_on(
&mut self,
attributes_to_search_on: &'ctx [String],

View File

@@ -214,7 +214,7 @@ pub fn partially_initialized_term_from_word(
if is_prefix && use_prefix_db.is_none() {
find_zero_typo_prefix_derivations(ctx, word_interned, &mut prefix_of)?;
}
let synonyms = ctx.index.synonyms(ctx.txn)?;
let synonyms = ctx.get_synonyms()?;
let mut synonym_word_count = 0;
let synonyms = synonyms
.get(&vec![word.to_owned()])

View File

@@ -258,7 +258,7 @@ pub fn make_ngram(
partially_initialized_term_from_word(ctx, &ngram_str, max_nbr_typos, is_prefix, true)?;
// Now add the synonyms
let index_synonyms = ctx.index.synonyms(ctx.txn)?;
let index_synonyms = ctx.get_synonyms()?;
term.zero_typo.synonyms.extend(
index_synonyms.get(&words).cloned().unwrap_or_default().into_iter().map(|words| {

View File

@@ -40,6 +40,7 @@ const UPGRADE_FUNCTIONS: &[&dyn UpgradeIndex] = &[
&ToTargetNoOp { target: (1, 20, 0) },
&ToTargetNoOp { target: (1, 21, 0) },
&ToTargetNoOp { target: (1, 22, 0) },
&ToTargetNoOp { target: (1, 23, 0) },
// This is the last upgrade function, it will be called when the index is up to date.
// any other upgrade function should be added before this one.
&ToCurrentNoOp {},
@@ -73,6 +74,7 @@ const fn start(from: (u32, u32, u32)) -> Option<usize> {
(1, 20, _) => function_index!(10),
(1, 21, _) => function_index!(11),
(1, 22, _) => function_index!(12),
(1, 23, _) => function_index!(13),
// We deliberately don't add a placeholder with (VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH) here to force manually
// considering dumpless upgrade.
(_major, _minor, _patch) => return None,

View File

@@ -58,6 +58,12 @@ Why? GitHub Merge Queue does not work with branch patterns yet, so we have to ad
5. Go to the GitHub interface, in the [`Release` section](https://github.com/meilisearch/meilisearch/releases) and click on `Draft a new release`
⚠️⚠️⚠️ Publish on `release-vX.Y.Z+1` branch, not on `main`!
📝 <ins>About the changelogs</s>
- Use the "Generate release notes" button in the GitHub interface to get the exhaustive list of PRs.
- Separate the PRs into different categories: Enhancement/Features, Bug fixes, Maintenance.
- Ensure each line makes sense for external people reading the changelogs. Add more details of usage if needed.
- Thank the external contributors at the end of the changelogs.
⚠️ <ins>If doing a patch release that should NOT be the `latest` release</s>:
- Do NOT check `Set as the latest release` when creating the GitHub release. If you did, quickly interrupt all CIs and delete the GitHub release!