snowykami/meilisearch
1
0
Fork 0
mirror of https://github.com/meilisearch/meilisearch.git synced 2025-10-10 05:36:35 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #5929 from meilisearch/compaction-task

Browse Source 
Introduce a task to compact an index
This commit is contained in:
Louis Dureuil
2025-10-09 11:30:01 +00:00
committed by GitHub
parent c2584c6edd 685663af3c
commit 59631afd9a
24 changed files with 424 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
Cargo.lock generated
View File

@@ -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",

6
crates/dump/src/lib.rs
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 }
}
}
}
}

3
crates/index-scheduler/src/dump.rs
View File

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

20
crates/index-scheduler/src/index_mapper/mod.rs
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)) =

3
crates/index-scheduler/src/insta_snapshot.rs
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:?} }}")
}
}
}

11
crates/index-scheduler/src/processing.rs
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,

19
crates/index-scheduler/src/scheduler/autobatcher.rs
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 { .. },
_,
) => {

17
crates/index-scheduler/src/scheduler/create_batch.rs
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 }))
}
}
}

134
crates/index-scheduler/src/scheduler/process_batch.rs
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,

20
crates/index-scheduler/src/scheduler/test.rs
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]

24
crates/index-scheduler/src/utils.rs
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);
}
}
}

1
crates/meilisearch-auth/src/store.rs
View File

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

7
crates/meilisearch-types/src/keys.rs
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]

32
crates/meilisearch-types/src/task_view.rs
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()
}
}
}
}
}

44
crates/meilisearch-types/src/tasks.rs
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 { .. }

84
crates/meilisearch/src/routes/indexes/compact.rs Normal file
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");

7
crates/meilisearch/src/routes/indexes/mod.rs
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)),
);
}

6
crates/meilisearch/src/routes/tasks_test.rs
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]

6
crates/meilisearch/tests/auth/api_keys.rs
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]

6
crates/meilisearch/tests/auth/errors.rs
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]

6
crates/meilisearch/tests/batches/errors.rs
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]

6
crates/meilisearch/tests/tasks/errors.rs
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"

4
crates/meilitool/src/main.rs
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)
}

2
crates/milli/Cargo.toml
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"