Merge pull request #5969 from xuhongxu96/main

Remove unused dependency `allocator-api2`

.github/dependabot.yml

@@ -7,6 +7,5 @@ updates:
     schedule:
       interval: "monthly"
     labels:
-      - 'skip changelog'
       - 'dependencies'
     rebase-strategy: disabled

.github/release-draft-template.yml

@@ -1,33 +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:
-      - 'maintenance'
-      - 'documentation'
-template: |
-  $CHANGES
-
-  ❤️  Huge thanks to our contributors: $CONTRIBUTORS.
-no-changes-template: 'Changes are coming soon 😎'
-sort-direction: 'ascending'
-replacers:
-  - search: '/(?:and )?@dependabot-preview(?:\[bot\])?,?/g'
-    replace: ''
-  - search: '/(?:and )?@dependabot(?:\[bot\])?,?/g'
-    replace: ''

.github/workflows/bench-manual.yml

@@ -18,7 +18,7 @@ jobs:
     timeout-minutes: 180 # 3h
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/bench-pr.yml

@@ -66,7 +66,7 @@ jobs:
           fetch-depth: 0 # fetch full history to be able to get main commit sha
           ref: ${{ steps.comment-branch.outputs.head_ref }}
 
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/bench-push-indexing.yml

@@ -12,7 +12,7 @@ jobs:
     timeout-minutes: 180 # 3h
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/benchmarks-manual.yml

@@ -18,7 +18,7 @@ jobs:
     timeout-minutes: 4320 # 72h
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/benchmarks-pr.yml

@@ -44,7 +44,7 @@ jobs:
             exit 1
           fi
 
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/benchmarks-push-indexing.yml

@@ -16,7 +16,7 @@ jobs:
     timeout-minutes: 4320 # 72h
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/benchmarks-push-search-geo.yml

@@ -15,7 +15,7 @@ jobs:
     runs-on: benchmarks
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/benchmarks-push-search-songs.yml

@@ -15,7 +15,7 @@ jobs:
     runs-on: benchmarks
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/benchmarks-push-search-wiki.yml

@@ -15,7 +15,7 @@ jobs:
     runs-on: benchmarks
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/flaky-tests.yml

@@ -17,7 +17,7 @@ jobs:
         run: |
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Install cargo-flaky
         run: cargo install cargo-flaky
       - name: Run cargo flaky in the dumps

.github/workflows/fuzzer-indexing.yml

@@ -12,7 +12,7 @@ jobs:
     timeout-minutes: 4320 # 72h
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
 

.github/workflows/publish-apt-brew-pkg.yml

@@ -25,7 +25,7 @@ jobs:
         run: |
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Install cargo-deb
         run: cargo install cargo-deb
       - uses: actions/checkout@v5

.github/workflows/publish-docker-images.yml

@@ -65,7 +65,7 @@ jobs:
         uses: docker/setup-buildx-action@v3
 
       - name: Install cosign
-        uses: sigstore/cosign-installer@d58896d6a1865668819e1d91763c7751a165e159 # tag=v3.9.2
+        uses: sigstore/cosign-installer@d7543c93d881b35a8faa02e8e3605f69b7a1ce62 # tag=v3.10.0
 
       - name: Login to Docker Hub
         uses: docker/login-action@v3

.github/workflows/publish-release-assets.yml

@@ -11,7 +11,7 @@ jobs:
   check-version:
     name: Check the version validity
     runs-on: ubuntu-latest
-    # No need to check the version for dry run (cron)
+    # No need to check the version for dry run (cron or workflow_dispatch)
     steps:
       - uses: actions/checkout@v5
       # Check if the tag has the v<nmumber>.<number>.<number> format.
@@ -45,10 +45,10 @@ jobs:
         run: |
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Build
         run: cargo build --release --locked
-      # No need to upload binaries for dry run (cron)
+      # No need to upload binaries for dry run (cron or workflow_dispatch)
       - name: Upload binaries to release
         if: github.event_name == 'release'
         uses: svenstaro/upload-release-action@2.11.2
@@ -75,10 +75,10 @@ jobs:
             asset_name: meilisearch-windows-amd64.exe
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Build
         run: cargo build --release --locked
-      # No need to upload binaries for dry run (cron)
+      # No need to upload binaries for dry run (cron or workflow_dispatch)
       - name: Upload binaries to release
         if: github.event_name == 'release'
         uses: svenstaro/upload-release-action@2.11.2
@@ -101,7 +101,7 @@ jobs:
       - name: Checkout repository
         uses: actions/checkout@v5
       - name: Installing Rust toolchain
-        uses: dtolnay/rust-toolchain@1.85
+        uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
           target: ${{ matrix.target }}
@@ -111,7 +111,7 @@ jobs:
           command: build
           args: --release --target ${{ matrix.target }}
       - name: Upload the binary to release
-        # No need to upload binaries for dry run (cron)
+        # No need to upload binaries for dry run (cron or workflow_dispatch)
         if: github.event_name == 'release'
         uses: svenstaro/upload-release-action@2.11.2
         with:
@@ -148,7 +148,7 @@ jobs:
           add-apt-repository "deb [arch=$(dpkg --print-architecture)] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
           apt-get update -y && apt-get install -y docker-ce
       - name: Installing Rust toolchain
-        uses: dtolnay/rust-toolchain@1.85
+        uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
           target: ${{ matrix.target }}
@@ -176,7 +176,7 @@ jobs:
       - name: List target output files
         run: ls -lR ./target
       - name: Upload the binary to release
-        # No need to upload binaries for dry run (cron)
+        # No need to upload binaries for dry run (cron or workflow_dispatch)
         if: github.event_name == 'release'
         uses: svenstaro/upload-release-action@2.11.2
         with:
@@ -187,6 +187,7 @@ jobs:
 
   publish-openapi-file:
     name: Publish OpenAPI file
+    needs: check-version
     runs-on: ubuntu-latest
     steps:
       - name: Checkout code
@@ -201,7 +202,7 @@ jobs:
           cd crates/openapi-generator
           cargo run --release -- --pretty --output ../../meilisearch.json
       - name: Upload OpenAPI to Release
-        # No need to upload for dry run (cron)
+        # No need to upload for dry run (cron or workflow_dispatch)
         if: github.event_name == 'release'
         uses: svenstaro/upload-release-action@2.11.2
         with:

.github/workflows/release-drafter.yml

@@ -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 }}

.github/workflows/sdks-tests.yml

@@ -50,7 +50,7 @@ jobs:
         with:
           repository: meilisearch/meilisearch-dotnet
       - name: Setup .NET Core
-        uses: actions/setup-dotnet@v4
+        uses: actions/setup-dotnet@v5
         with:
           dotnet-version: "8.0.x"
       - name: Install dependencies
@@ -100,7 +100,7 @@ jobs:
           - '7700:7700'
     steps:
       - name: Set up Go
-        uses: actions/setup-go@v5
+        uses: actions/setup-go@v6
         with:
           go-version: stable
       - uses: actions/checkout@v5
@@ -135,13 +135,13 @@ jobs:
       - name: Set up Java
         uses: actions/setup-java@v5
         with:
-          java-version: 8
-          distribution: 'zulu'
+          java-version: 17
+          distribution: 'temurin'
           cache: gradle
       - name: Grant execute permission for gradlew
         run: chmod +x gradlew
       - name: Build and run unit and integration tests
-        run: ./gradlew build integrationTest
+        run: ./gradlew build integrationTest --info
 
   meilisearch-js-tests:
     needs: define-docker-image
@@ -160,7 +160,7 @@ jobs:
         with:
           repository: meilisearch/meilisearch-js
       - name: Setup node
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v5
         with:
           cache: 'yarn'
       - name: Install dependencies
@@ -224,7 +224,7 @@ jobs:
         with:
           repository: meilisearch/meilisearch-python
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
       - name: Install pipenv
         uses: dschep/install-pipenv-action@v1
       - name: Install dependencies
@@ -318,7 +318,7 @@ jobs:
         with:
           repository: meilisearch/meilisearch-js-plugins
       - name: Setup node
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v5
         with:
           cache: yarn
       - name: Install dependencies

.github/workflows/test-suite.yml

@@ -27,7 +27,7 @@ jobs:
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
       - name: Setup test with Rust stable
-        uses: dtolnay/rust-toolchain@1.85
+        uses: dtolnay/rust-toolchain@1.89
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
       - name: Run cargo check without any default features
@@ -52,7 +52,7 @@ jobs:
       - uses: actions/checkout@v5
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Run cargo check without any default features
         uses: actions-rs/cargo@v1
         with:
@@ -77,7 +77,7 @@ jobs:
         run: |
           apt-get update
           apt-get install --assume-yes build-essential curl
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Run cargo build with almost all features
         run: |
           cargo build --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda,test-ollama)"
@@ -129,7 +129,7 @@ jobs:
         run: |
           apt-get update
           apt-get install --assume-yes build-essential curl
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Run cargo tree without default features and check lindera is not present
         run: |
           if cargo tree -f '{p} {f}' -e normal --no-default-features | grep -qz lindera; then
@@ -153,7 +153,7 @@ jobs:
         run: |
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
       - name: Run tests in debug
@@ -167,7 +167,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
           components: clippy
@@ -184,7 +184,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
           toolchain: nightly-2024-07-09

.github/workflows/update-cargo-toml-version.yml

@@ -18,7 +18,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v5
-      - uses: dtolnay/rust-toolchain@1.85
+      - uses: dtolnay/rust-toolchain@1.89
         with:
           profile: minimal
       - name: Install sd

Cargo.toml

@@ -23,7 +23,7 @@ members = [
 ]
 
 [workspace.package]
-version = "1.21.0"
+version = "1.26.0"
 authors = [
     "Quentin de Quelen <quentin@dequelen.me>",
     "Clément Renault <clement@meilisearch.com>",

Dockerfile

@@ -1,5 +1,5 @@
 # Compile
-FROM    rust:1.85-alpine3.20 AS compiler
+FROM    rust:1.89-alpine3.22 AS compiler
 
 RUN     apk add -q --no-cache build-base openssl-dev
 
@@ -20,7 +20,7 @@ RUN     set -eux; \
         cargo build --release -p meilisearch -p meilitool
 
 # Run
-FROM    alpine:3.20
+FROM    alpine:3.22
 LABEL   org.opencontainers.image.source="https://github.com/meilisearch/meilisearch"
 
 ENV     MEILI_HTTP_ADDR 0.0.0.0:7700

LICENSE

@@ -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`

LICENSE-MIT

@@ -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.

README.md

@@ -39,6 +39,7 @@
 ## 🖥 Examples
 
 - [**Movies**](https://where2watch.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=organization) — An application to help you find streaming platforms to watch movies using [hybrid search](https://www.meilisearch.com/solutions/hybrid-search?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos).
+- [**Flickr**](https://flickr.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=organization) — Search and explore one hundred million Flickr images with semantic search.
 - [**Ecommerce**](https://ecommerce.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) — Ecommerce website using disjunctive [facets](https://www.meilisearch.com/docs/learn/fine_tuning_results/faceted_search?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos), range and rating filtering, and pagination.
 - [**Songs**](https://music.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) — Search through 47 million of songs.
 - [**SaaS**](https://saas.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) — Search for contacts, deals, and companies in this [multi-tenant](https://www.meilisearch.com/docs/learn/security/multitenancy_tenant_tokens?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) CRM application.
@@ -121,7 +122,7 @@ If you want to know more about the kind of data we collect and what we use it fo
 
 Meilisearch is a search engine created by [Meili](https://www.meilisearch.com/careers), a software development company headquartered in France and with team members all over the world. Want to know more about us? [Check out our blog!](https://blog.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=contact)
 
-🗞 [Subscribe to our newsletter](https://meilisearch.us2.list-manage.com/subscribe?u=27870f7b71c908a8b359599fb&id=79582d828e) if you don't want to miss any updates! We promise we won't clutter your mailbox: we only send one edition every two months.
+🗞 [Subscribe to our newsletter](https://share-eu1.hsforms.com/1LN5N0x_GQgq7ss7tXmSykwfg3aq) if you don't want to miss any updates! We promise we won't clutter your mailbox: we only send one edition every two months.
 
 💌 Want to make a suggestion or give feedback? Here are some of the channels where you can reach us:
 

crates/dump/src/lib.rs

@@ -96,6 +96,8 @@ pub struct TaskDump {
     pub finished_at: Option<OffsetDateTime>,
     #[serde(default, skip_serializing_if = "Option::is_none")]
     pub network: Option<TaskNetwork>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub custom_metadata: Option<String>,
 }
 
 // A `Kind` specific version made for the dump. If modified you may break the dump.
@@ -158,6 +160,9 @@ pub enum KindDump {
     UpgradeDatabase {
         from: (u32, u32, u32),
     },
+    IndexCompaction {
+        index_uid: String,
+    },
 }
 
 impl From<Task> for TaskDump {
@@ -175,6 +180,7 @@ impl From<Task> for TaskDump {
             started_at: task.started_at,
             finished_at: task.finished_at,
             network: task.network,
+            custom_metadata: task.custom_metadata,
         }
     }
 }
@@ -240,6 +246,9 @@ impl From<KindWithContent> for KindDump {
             KindWithContent::UpgradeDatabase { from: version } => {
                 KindDump::UpgradeDatabase { from: version }
             }
+            KindWithContent::IndexCompaction { index_uid } => {
+                KindDump::IndexCompaction { index_uid }
+            }
         }
     }
 }
@@ -390,6 +399,7 @@ pub(crate) mod test {
                     started_at: Some(datetime!(2022-11-20 0:00 UTC)),
                     finished_at: Some(datetime!(2022-11-21 0:00 UTC)),
                     network: None,
+                    custom_metadata: None,
                 },
                 None,
             ),
@@ -415,6 +425,7 @@ pub(crate) mod test {
                     started_at: None,
                     finished_at: None,
                     network: None,
+                    custom_metadata: None,
                 },
                 Some(vec![
                     json!({ "id": 4, "race": "leonberg" }).as_object().unwrap().clone(),
@@ -435,6 +446,7 @@ pub(crate) mod test {
                     started_at: None,
                     finished_at: None,
                     network: None,
+                    custom_metadata: None,
                 },
                 None,
             ),

crates/dump/src/reader/compat/v2_to_v3.rs

@@ -97,6 +97,7 @@ impl CompatV2ToV3 {
     }
 }
 
+#[allow(clippy::large_enum_variant)]
 pub enum CompatIndexV2ToV3 {
     V2(v2::V2IndexReader),
     Compat(Box<CompatIndexV1ToV2>),

crates/dump/src/reader/compat/v5_to_v6.rs

@@ -164,6 +164,7 @@ impl CompatV5ToV6 {
                     started_at: task_view.started_at,
                     finished_at: task_view.finished_at,
                     network: None,
+                    custom_metadata: None,
                 };
 
                 (task, content_file)

crates/file-store/src/lib.rs

@@ -60,7 +60,7 @@ impl FileStore {
 
     /// Returns the file corresponding to the requested uuid.
     pub fn get_update(&self, uuid: Uuid) -> Result<StdFile> {
-        let path = self.get_update_path(uuid);
+        let path = self.update_path(uuid);
         let file = match StdFile::open(path) {
             Ok(file) => file,
             Err(e) => {
@@ -72,7 +72,7 @@ impl FileStore {
     }
 
     /// Returns the path that correspond to this uuid, the path could not exists.
-    pub fn get_update_path(&self, uuid: Uuid) -> PathBuf {
+    pub fn update_path(&self, uuid: Uuid) -> PathBuf {
         self.path.join(uuid.to_string())
     }
 

crates/filter-parser/src/condition.rs

@@ -7,23 +7,14 @@
 
 use nom::branch::alt;
 use nom::bytes::complete::tag;
-use nom::character::complete::char;
-use nom::character::complete::multispace0;
-use nom::character::complete::multispace1;
-use nom::combinator::cut;
-use nom::combinator::map;
-use nom::combinator::value;
-use nom::sequence::preceded;
-use nom::sequence::{terminated, tuple};
+use nom::character::complete::{char, multispace0, multispace1};
+use nom::combinator::{cut, map, value};
+use nom::sequence::{preceded, terminated, tuple};
 use Condition::*;
 
 use crate::error::IResultExt;
-use crate::value::parse_vector_value;
-use crate::value::parse_vector_value_cut;
-use crate::Error;
-use crate::ErrorKind;
-use crate::VectorFilter;
-use crate::{parse_value, FilterCondition, IResult, Span, Token};
+use crate::value::{parse_vector_value, parse_vector_value_cut};
+use crate::{parse_value, Error, ErrorKind, FilterCondition, IResult, Span, Token, VectorFilter};
 
 #[derive(Debug, Clone, PartialEq, Eq)]
 pub enum Condition<'a> {
@@ -124,7 +115,7 @@ pub fn parse_not_exists(input: Span) -> IResult<FilterCondition> {
     Ok((input, FilterCondition::Not(Box::new(FilterCondition::Condition { fid: key, op: Exists }))))
 }
 
-fn parse_vectors(input: Span) -> IResult<(Token, Option<Token>, VectorFilter<'_>)> {
+fn parse_vectors(input: Span) -> IResult<(Token, Option<Token>, VectorFilter)> {
     let (input, _) = multispace0(input)?;
     let (input, fid) = tag("_vectors")(input)?;
 

crates/filter-parser/src/error.rs

@@ -75,7 +75,11 @@ pub enum ExpectedValueKind {
 pub enum ErrorKind<'a> {
     ReservedGeo(&'a str),
     GeoRadius,
+    GeoRadiusArgumentCount(usize),
     GeoBoundingBox,
+    GeoPolygon,
+    GeoPolygonNotEnoughPoints(usize),
+    GeoCoordinatesNotPair(usize),
     MisusedGeoRadius,
     MisusedGeoBoundingBox,
     VectorFilterLeftover,
@@ -189,7 +193,7 @@ impl Display for Error<'_> {
             }
             ErrorKind::InvalidPrimary => {
                 let text = if input.trim().is_empty() { "but instead got nothing.".to_string() } else { format!("at `{}`.", escaped_input) };
-                writeln!(f, "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` {}", text)?
+                writeln!(f, "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` {text}")?
             }
             ErrorKind::InvalidEscapedNumber => {
                 writeln!(f, "Found an invalid escaped sequence number: `{}`.", escaped_input)?
@@ -198,11 +202,23 @@ impl Display for Error<'_> {
                 writeln!(f, "Found unexpected characters at the end of the filter: `{}`. You probably forgot an `OR` or an `AND` rule.", escaped_input)?
             }
             ErrorKind::GeoRadius => {
-                writeln!(f, "The `_geoRadius` filter expects three arguments: `_geoRadius(latitude, longitude, radius)`.")?
+                writeln!(f, "The `_geoRadius` filter must be in the form: `_geoRadius(latitude, longitude, radius, optionalResolution)`.")?
+            }
+            ErrorKind::GeoRadiusArgumentCount(count) => {
+                writeln!(f, "Was expecting 3 or 4 arguments for `_geoRadius`, but instead found {count}.")?
             }
             ErrorKind::GeoBoundingBox => {
                 writeln!(f, "The `_geoBoundingBox` filter expects two pairs of arguments: `_geoBoundingBox([latitude, longitude], [latitude, longitude])`.")?
             }
+            ErrorKind::GeoPolygon => {
+                writeln!(f, "The `_geoPolygon` filter doesn't match the expected format: `_geoPolygon([latitude, longitude], [latitude, longitude])`.")?
+            }
+            ErrorKind::GeoPolygonNotEnoughPoints(n) => {
+                writeln!(f, "The `_geoPolygon` filter expects at least 3 points but only {n} were specified")?;
+            }
+            ErrorKind::GeoCoordinatesNotPair(number) => {
+                writeln!(f, "Was expecting 2 coordinates but instead found {number}.")?
+            }
             ErrorKind::ReservedGeo(name) => {
                 writeln!(f, "`{}` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.", name.escape_debug())?
             }

crates/filter-parser/src/lib.rs

@@ -19,6 +19,7 @@
 //! word           = (alphanumeric | _ | - | .)+
 //! geoRadius      = "_geoRadius(" WS* float WS* "," WS* float WS* "," float WS* ")"
 //! geoBoundingBox = "_geoBoundingBox([" WS * float WS* "," WS* float WS* "], [" WS* float WS* "," WS* float WS* "]")
+//! geoPolygon     = "_geoPolygon([[" WS* float WS* "," WS* float WS* "],+])"
 //! ```
 //!
 //! Other BNF grammar used to handle some specific errors:
@@ -116,7 +117,7 @@ impl<'a> Token<'a> {
         self.span
     }
 
-    pub fn parse_finite_float(&self) -> Result<f64, Error> {
+    pub fn parse_finite_float(&self) -> Result<f64, Error<'a>> {
         let value: f64 = self.value().parse().map_err(|e| self.as_external_error(e))?;
         if value.is_finite() {
             Ok(value)
@@ -156,8 +157,9 @@ pub enum FilterCondition<'a> {
     Or(Vec<Self>),
     And(Vec<Self>),
     VectorExists { fid: Token<'a>, embedder: Option<Token<'a>>, filter: VectorFilter<'a> },
-    GeoLowerThan { point: [Token<'a>; 2], radius: Token<'a> },
+    GeoLowerThan { point: [Token<'a>; 2], radius: Token<'a>, resolution: Option<Token<'a>> },
     GeoBoundingBox { top_right_point: [Token<'a>; 2], bottom_left_point: [Token<'a>; 2] },
+    GeoPolygon { points: Vec<[Token<'a>; 2]> },
 }
 
 pub enum TraversedElement<'a> {
@@ -166,7 +168,7 @@ pub enum TraversedElement<'a> {
 }
 
 impl<'a> FilterCondition<'a> {
-    pub fn use_contains_operator(&self) -> Option<&Token> {
+    pub fn use_contains_operator(&self) -> Option<&Token<'a>> {
         match self {
             FilterCondition::Condition { fid: _, op } => match op {
                 Condition::GreaterThan(_)
@@ -189,11 +191,12 @@ impl<'a> FilterCondition<'a> {
             FilterCondition::VectorExists { .. }
             | FilterCondition::GeoLowerThan { .. }
             | FilterCondition::GeoBoundingBox { .. }
+            | FilterCondition::GeoPolygon { .. }
             | FilterCondition::In { .. } => None,
         }
     }
 
-    pub fn use_vector_filter(&self) -> Option<&Token> {
+    pub fn use_vector_filter(&self) -> Option<&Token<'a>> {
         match self {
             FilterCondition::Condition { .. } => None,
             FilterCondition::Not(this) => this.use_vector_filter(),
@@ -202,12 +205,13 @@ impl<'a> FilterCondition<'a> {
             }
             FilterCondition::GeoLowerThan { .. }
             | FilterCondition::GeoBoundingBox { .. }
+            | FilterCondition::GeoPolygon { .. }
             | FilterCondition::In { .. } => None,
             FilterCondition::VectorExists { fid, .. } => Some(fid),
         }
     }
 
-    pub fn fids(&self, depth: usize) -> Box<dyn Iterator<Item = &Token> + '_> {
+    pub fn fids(&self, depth: usize) -> Box<dyn Iterator<Item = &Token<'a>> + '_> {
         if depth == 0 {
             return Box::new(std::iter::empty());
         }
@@ -228,7 +232,7 @@ impl<'a> FilterCondition<'a> {
     }
 
     /// Returns the first token found at the specified depth, `None` if no token at this depth.
-    pub fn token_at_depth(&self, depth: usize) -> Option<&Token> {
+    pub fn token_at_depth(&self, depth: usize) -> Option<&Token<'a>> {
         match self {
             FilterCondition::Condition { fid, .. } if depth == 0 => Some(fid),
             FilterCondition::Or(subfilters) => {
@@ -396,23 +400,27 @@ fn parse_not(input: Span, depth: usize) -> IResult<FilterCondition> {
 /// If we parse `_geoRadius` we MUST parse the rest of the expression.
 fn parse_geo_radius(input: Span) -> IResult<FilterCondition> {
     // we want to allow space BEFORE the _geoRadius but not after
-    let parsed = preceded(
-        tuple((multispace0, word_exact("_geoRadius"))),
-        // if we were able to parse `_geoRadius` and can't parse the rest of the input we return a failure
-        cut(delimited(char('('), separated_list1(tag(","), ws(recognize_float)), char(')'))),
-    )(input)
-    .map_err(|e| e.map(|_| Error::new_from_kind(input, ErrorKind::GeoRadius)));
+
+    let (input, _) = tuple((multispace0, word_exact("_geoRadius")))(input)?;
+
+    // if we were able to parse `_geoRadius` and can't parse the rest of the input we return a failure
+
+    let parsed =
+        delimited(char('('), separated_list1(tag(","), ws(recognize_float)), char(')'))(input)
+            .map_cut(ErrorKind::GeoRadius);
 
     let (input, args) = parsed?;
 
-    if args.len() != 3 {
-        return Err(nom::Err::Failure(Error::new_from_kind(input, ErrorKind::GeoRadius)));
+    if !(3..=4).contains(&args.len()) {
+        return Err(Error::failure_from_kind(input, ErrorKind::GeoRadiusArgumentCount(args.len())));
     }
 
     let res = FilterCondition::GeoLowerThan {
         point: [args[0].into(), args[1].into()],
         radius: args[2].into(),
+        resolution: args.get(3).cloned().map(Token::from),
     };
+
     Ok((input, res))
 }
 
@@ -420,26 +428,33 @@ fn parse_geo_radius(input: Span) -> IResult<FilterCondition> {
 /// If we parse `_geoBoundingBox` we MUST parse the rest of the expression.
 fn parse_geo_bounding_box(input: Span) -> IResult<FilterCondition> {
     // we want to allow space BEFORE the _geoBoundingBox but not after
-    let parsed = preceded(
-        tuple((multispace0, word_exact("_geoBoundingBox"))),
-        // if we were able to parse `_geoBoundingBox` and can't parse the rest of the input we return a failure
-        cut(delimited(
-            char('('),
-            separated_list1(
-                tag(","),
-                ws(delimited(char('['), separated_list1(tag(","), ws(recognize_float)), char(']'))),
-            ),
-            char(')'),
-        )),
+
+    let (input, _) = tuple((multispace0, word_exact("_geoBoundingBox")))(input)?;
+
+    // if we were able to parse `_geoBoundingBox` and can't parse the rest of the input we return a failure
+
+    let (input, args) = delimited(
+        char('('),
+        separated_list1(
+            tag(","),
+            ws(delimited(char('['), separated_list1(tag(","), ws(recognize_float)), char(']'))),
+        ),
+        char(')'),
     )(input)
-    .map_err(|e| e.map(|_| Error::new_from_kind(input, ErrorKind::GeoBoundingBox)));
+    .map_cut(ErrorKind::GeoBoundingBox)?;
 
-    let (input, args) = parsed?;
-
-    if args.len() != 2 || args[0].len() != 2 || args[1].len() != 2 {
+    if args.len() != 2 {
         return Err(Error::failure_from_kind(input, ErrorKind::GeoBoundingBox));
     }
 
+    if let Some(offending) = args.iter().find(|a| a.len() != 2) {
+        let context = offending.first().unwrap_or(&input);
+        return Err(Error::failure_from_kind(
+            *context,
+            ErrorKind::GeoCoordinatesNotPair(offending.len()),
+        ));
+    }
+
     let res = FilterCondition::GeoBoundingBox {
         top_right_point: [args[0][0].into(), args[0][1].into()],
         bottom_left_point: [args[1][0].into(), args[1][1].into()],
@@ -447,6 +462,47 @@ fn parse_geo_bounding_box(input: Span) -> IResult<FilterCondition> {
     Ok((input, res))
 }
 
+/// geoPolygon     = "_geoPolygon([[" WS* float WS* "," WS* float WS* "],+])"
+/// If we parse `_geoPolygon` we MUST parse the rest of the expression.
+fn parse_geo_polygon(input: Span) -> IResult<FilterCondition> {
+    // we want to allow space BEFORE the _geoPolygon but not after
+
+    let (input, _) = tuple((multispace0, word_exact("_geoPolygon")))(input)?;
+
+    // if we were able to parse `_geoPolygon` and can't parse the rest of the input we return a failure
+
+    let (input, args): (_, Vec<Vec<LocatedSpan<_, _>>>) = delimited(
+        char('('),
+        separated_list1(
+            tag(","),
+            ws(delimited(char('['), separated_list1(tag(","), ws(recognize_float)), char(']'))),
+        ),
+        preceded(opt(ws(char(','))), char(')')), // Tolerate trailing comma
+    )(input)
+    .map_cut(ErrorKind::GeoPolygon)?;
+
+    if args.len() < 3 {
+        let context = args.last().and_then(|a| a.last()).unwrap_or(&input);
+        return Err(Error::failure_from_kind(
+            *context,
+            ErrorKind::GeoPolygonNotEnoughPoints(args.len()),
+        ));
+    }
+
+    if let Some(offending) = args.iter().find(|a| a.len() != 2) {
+        let context = offending.first().unwrap_or(&input);
+        return Err(Error::failure_from_kind(
+            *context,
+            ErrorKind::GeoCoordinatesNotPair(offending.len()),
+        ));
+    }
+
+    let res = FilterCondition::GeoPolygon {
+        points: args.into_iter().map(|a| [a[0].into(), a[1].into()]).collect(),
+    };
+    Ok((input, res))
+}
+
 /// geoPoint      = WS* "_geoPoint(float WS* "," WS* float WS* "," WS* float)
 fn parse_geo_point(input: Span) -> IResult<FilterCondition> {
     // we want to forbid space BEFORE the _geoPoint but not after
@@ -516,8 +572,8 @@ fn parse_primary(input: Span, depth: usize) -> IResult<FilterCondition> {
                 Error::new_from_kind(input, ErrorKind::MissingClosingDelimiter(c.char()))
             }),
         ),
-        parse_geo_radius,
-        parse_geo_bounding_box,
+        // Made a random block of functions because we reached the maximum number of elements per alt
+        alt((parse_geo_radius, parse_geo_bounding_box, parse_geo_polygon)),
         parse_in,
         parse_not_in,
         parse_condition,
@@ -597,9 +653,12 @@ impl std::fmt::Display for FilterCondition<'_> {
                 }
                 write!(f, " EXISTS")
             }
-            FilterCondition::GeoLowerThan { point, radius } => {
+            FilterCondition::GeoLowerThan { point, radius, resolution: None } => {
                 write!(f, "_geoRadius({}, {}, {})", point[0], point[1], radius)
             }
+            FilterCondition::GeoLowerThan { point, radius, resolution: Some(resolution) } => {
+                write!(f, "_geoRadius({}, {}, {}, {})", point[0], point[1], radius, resolution)
+            }
             FilterCondition::GeoBoundingBox {
                 top_right_point: top_left_point,
                 bottom_left_point: bottom_right_point,
@@ -613,6 +672,13 @@ impl std::fmt::Display for FilterCondition<'_> {
                     bottom_right_point[1]
                 )
             }
+            FilterCondition::GeoPolygon { points } => {
+                write!(f, "_geoPolygon([")?;
+                for point in points {
+                    write!(f, "[{}, {}], ", point[0], point[1])?;
+                }
+                write!(f, "])")
+            }
         }
     }
 }
@@ -651,7 +717,7 @@ pub mod tests {
     /// Create a raw [Token]. You must specify the string that appear BEFORE your element followed by your element
     pub fn rtok<'a>(before: &'a str, value: &'a str) -> Token<'a> {
         // if the string is empty we still need to return 1 for the line number
-        let lines = before.is_empty().then_some(1).unwrap_or_else(|| before.lines().count());
+        let lines = if before.is_empty() { 1 } else { before.lines().count() };
         let offset = before.chars().count();
         // the extra field is not checked in the tests so we can set it to nothing
         unsafe { Span::new_from_raw_offset(offset, lines as u32, value, "") }.into()
@@ -776,12 +842,17 @@ pub mod tests {
         insta::assert_snapshot!(p("_geoRadius(12, 13, 14)"), @"_geoRadius({12}, {13}, {14})");
         insta::assert_snapshot!(p("NOT _geoRadius(12, 13, 14)"), @"NOT (_geoRadius({12}, {13}, {14}))");
         insta::assert_snapshot!(p("_geoRadius(12,13,14)"), @"_geoRadius({12}, {13}, {14})");
+        insta::assert_snapshot!(p("_geoRadius(12,13,14,1000)"), @"_geoRadius({12}, {13}, {14}, {1000})");
 
         // Test geo bounding box
         insta::assert_snapshot!(p("_geoBoundingBox([12, 13], [14, 15])"), @"_geoBoundingBox([{12}, {13}], [{14}, {15}])");
         insta::assert_snapshot!(p("NOT _geoBoundingBox([12, 13], [14, 15])"), @"NOT (_geoBoundingBox([{12}, {13}], [{14}, {15}]))");
         insta::assert_snapshot!(p("_geoBoundingBox([12,13],[14,15])"), @"_geoBoundingBox([{12}, {13}], [{14}, {15}])");
 
+        // Test geo polygon
+        insta::assert_snapshot!(p("_geoPolygon([12, 13], [14, 15], [16, 17])"), @"_geoPolygon([[{12}, {13}], [{14}, {15}], [{16}, {17}], ])");
+        insta::assert_snapshot!(p("_geoPolygon([12, 13], [14, 15], [-1.2,2939.2], [1,1])"), @"_geoPolygon([[{12}, {13}], [{14}, {15}], [{-1.2}, {2939.2}], [{1}, {1}], ])");
+
         // Test OR + AND
         insta::assert_snapshot!(p("channel = ponce AND 'dog race' != 'bernese mountain'"), @"AND[{channel} = {ponce}, {dog race} != {bernese mountain}, ]");
         insta::assert_snapshot!(p("channel = ponce OR 'dog race' != 'bernese mountain'"), @"OR[{channel} = {ponce}, {dog race} != {bernese mountain}, ]");
@@ -838,50 +909,80 @@ pub mod tests {
         11:12 channel = 🐻 AND followers < 100
         "###);
 
-        insta::assert_snapshot!(p("'OR'"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `\'OR\'`.
+        insta::assert_snapshot!(p("'OR'"), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `\'OR\'`.
         1:5 'OR'
-        "###);
+        ");
 
         insta::assert_snapshot!(p("OR"), @r###"
         Was expecting a value but instead got `OR`, which is a reserved keyword. To use `OR` as a field name or a value, surround it by quotes.
         1:3 OR
         "###);
 
-        insta::assert_snapshot!(p("channel Ponce"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `channel Ponce`.
+        insta::assert_snapshot!(p("channel Ponce"), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `channel Ponce`.
         1:14 channel Ponce
-        "###);
+        ");
 
-        insta::assert_snapshot!(p("channel = Ponce OR"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` but instead got nothing.
+        insta::assert_snapshot!(p("channel = Ponce OR"), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` but instead got nothing.
         19:19 channel = Ponce OR
-        "###);
+        ");
 
-        insta::assert_snapshot!(p("_geoRadius"), @r###"
-        The `_geoRadius` filter expects three arguments: `_geoRadius(latitude, longitude, radius)`.
-        1:11 _geoRadius
-        "###);
+        insta::assert_snapshot!(p("_geoRadius"), @r"
+        The `_geoRadius` filter must be in the form: `_geoRadius(latitude, longitude, radius, optionalResolution)`.
+        11:11 _geoRadius
+        ");
 
-        insta::assert_snapshot!(p("_geoRadius = 12"), @r###"
-        The `_geoRadius` filter expects three arguments: `_geoRadius(latitude, longitude, radius)`.
-        1:16 _geoRadius = 12
-        "###);
+        insta::assert_snapshot!(p("_geoRadius = 12"), @r"
+        The `_geoRadius` filter must be in the form: `_geoRadius(latitude, longitude, radius, optionalResolution)`.
+        11:16 _geoRadius = 12
+        ");
 
-        insta::assert_snapshot!(p("_geoBoundingBox"), @r###"
+        insta::assert_snapshot!(p("_geoBoundingBox"), @r"
         The `_geoBoundingBox` filter expects two pairs of arguments: `_geoBoundingBox([latitude, longitude], [latitude, longitude])`.
-        1:16 _geoBoundingBox
-        "###);
+        16:16 _geoBoundingBox
+        ");
 
-        insta::assert_snapshot!(p("_geoBoundingBox = 12"), @r###"
+        insta::assert_snapshot!(p("_geoBoundingBox = 12"), @r"
         The `_geoBoundingBox` filter expects two pairs of arguments: `_geoBoundingBox([latitude, longitude], [latitude, longitude])`.
-        1:21 _geoBoundingBox = 12
-        "###);
+        16:21 _geoBoundingBox = 12
+        ");
 
-        insta::assert_snapshot!(p("_geoBoundingBox(1.0, 1.0)"), @r###"
+        insta::assert_snapshot!(p("_geoBoundingBox(1.0, 1.0)"), @r"
         The `_geoBoundingBox` filter expects two pairs of arguments: `_geoBoundingBox([latitude, longitude], [latitude, longitude])`.
-        1:26 _geoBoundingBox(1.0, 1.0)
-        "###);
+        17:26 _geoBoundingBox(1.0, 1.0)
+        ");
+
+        insta::assert_snapshot!(p("_geoPolygon([1,2,3])"), @r"
+        The `_geoPolygon` filter expects at least 3 points but only 1 were specified
+        18:19 _geoPolygon([1,2,3])
+        ");
+
+        insta::assert_snapshot!(p("_geoPolygon(1,2,3)"), @r"
+        The `_geoPolygon` filter doesn't match the expected format: `_geoPolygon([latitude, longitude], [latitude, longitude])`.
+        13:19 _geoPolygon(1,2,3)
+        ");
+
+        insta::assert_snapshot!(p("_geoPolygon([1,2],[1,2],[1,2,3])"), @r"
+        Was expecting 2 coordinates but instead found 3.
+        26:27 _geoPolygon([1,2],[1,2],[1,2,3])
+        ");
+
+        insta::assert_snapshot!(p("_geoPolygon([1,2],[1,2,3])"), @r"
+        The `_geoPolygon` filter expects at least 3 points but only 2 were specified
+        24:25 _geoPolygon([1,2],[1,2,3])
+        ");
+
+        insta::assert_snapshot!(p("_geoPolygon(1)"), @r"
+        The `_geoPolygon` filter doesn't match the expected format: `_geoPolygon([latitude, longitude], [latitude, longitude])`.
+        13:15 _geoPolygon(1)
+        ");
+
+        insta::assert_snapshot!(p("_geoPolygon([1,2)"), @r"
+        The `_geoPolygon` filter doesn't match the expected format: `_geoPolygon([latitude, longitude], [latitude, longitude])`.
+        17:18 _geoPolygon([1,2)
+        ");
 
         insta::assert_snapshot!(p("_geoPoint(12, 13, 14)"), @r###"
         `_geoPoint` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.
@@ -938,15 +1039,15 @@ pub mod tests {
         34:35 channel = mv OR followers >= 1000)
         "###);
 
-        insta::assert_snapshot!(p("colour NOT EXIST"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `colour NOT EXIST`.
+        insta::assert_snapshot!(p("colour NOT EXIST"), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `colour NOT EXIST`.
         1:17 colour NOT EXIST
-        "###);
+        ");
 
-        insta::assert_snapshot!(p("subscribers 100 TO1000"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `subscribers 100 TO1000`.
+        insta::assert_snapshot!(p("subscribers 100 TO1000"), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `subscribers 100 TO1000`.
         1:23 subscribers 100 TO1000
-        "###);
+        ");
 
         insta::assert_snapshot!(p("channel = ponce ORdog != 'bernese mountain'"), @r###"
         Found unexpected characters at the end of the filter: `ORdog != \'bernese mountain\'`. You probably forgot an `OR` or an `AND` rule.
@@ -1071,38 +1172,38 @@ pub mod tests {
         5:7 NOT OR EXISTS AND EXISTS NOT EXISTS
         "###);
 
-        insta::assert_snapshot!(p(r#"value NULL"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value NULL`.
+        insta::assert_snapshot!(p(r#"value NULL"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value NULL`.
         1:11 value NULL
-        "###);
-        insta::assert_snapshot!(p(r#"value NOT NULL"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value NOT NULL`.
+        ");
+        insta::assert_snapshot!(p(r#"value NOT NULL"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value NOT NULL`.
         1:15 value NOT NULL
-        "###);
-        insta::assert_snapshot!(p(r#"value EMPTY"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value EMPTY`.
+        ");
+        insta::assert_snapshot!(p(r#"value EMPTY"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value EMPTY`.
         1:12 value EMPTY
-        "###);
-        insta::assert_snapshot!(p(r#"value NOT EMPTY"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value NOT EMPTY`.
+        ");
+        insta::assert_snapshot!(p(r#"value NOT EMPTY"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value NOT EMPTY`.
         1:16 value NOT EMPTY
-        "###);
-        insta::assert_snapshot!(p(r#"value IS"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS`.
+        ");
+        insta::assert_snapshot!(p(r#"value IS"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value IS`.
         1:9 value IS
-        "###);
-        insta::assert_snapshot!(p(r#"value IS NOT"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS NOT`.
+        ");
+        insta::assert_snapshot!(p(r#"value IS NOT"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value IS NOT`.
         1:13 value IS NOT
-        "###);
-        insta::assert_snapshot!(p(r#"value IS EXISTS"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS EXISTS`.
+        ");
+        insta::assert_snapshot!(p(r#"value IS EXISTS"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value IS EXISTS`.
         1:16 value IS EXISTS
-        "###);
-        insta::assert_snapshot!(p(r#"value IS NOT EXISTS"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS NOT EXISTS`.
+        ");
+        insta::assert_snapshot!(p(r#"value IS NOT EXISTS"#), @r"
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, `_geoBoundingBox` or `_geoPolygon` at `value IS NOT EXISTS`.
         1:20 value IS NOT EXISTS
-        "###);
+        ");
     }
 
     #[test]

crates/index-scheduler/Cargo.toml

@@ -14,6 +14,7 @@ license.workspace = true
 anyhow = "1.0.98"
 bincode = "1.3.3"
 byte-unit = "5.1.6"
+bytes = "1.10.1"
 bumpalo = "3.18.1"
 bumparaw-collections = "0.1.4"
 convert_case = "0.8.0"
@@ -32,6 +33,7 @@ rayon = "1.10.0"
 roaring = { version = "0.10.12", features = ["serde"] }
 serde = { version = "1.0.219", features = ["derive"] }
 serde_json = { version = "1.0.140", features = ["preserve_order"] }
+tar = "0.4.44"
 synchronoise = "1.0.1"
 tempfile = "3.20.0"
 thiserror = "2.0.12"
@@ -45,6 +47,9 @@ tracing = "0.1.41"
 ureq = "2.12.1"
 uuid = { version = "1.17.0", features = ["serde", "v4"] }
 backoff = "0.4.0"
+reqwest = { version = "0.12.23", features = ["rustls-tls", "http2"], default-features = false }
+rusty-s3 = "0.8.1"
+tokio = { version = "1.47.1", features = ["full"] }
 
 [dev-dependencies]
 big_s = "1.0.2"

crates/index-scheduler/src/dump.rs

@@ -1,3 +1,5 @@
+#![allow(clippy::result_large_err)]
+
 use std::collections::HashMap;
 use std::io;
 
@@ -148,6 +150,7 @@ impl<'a> Dump<'a> {
             details: task.details,
             status: task.status,
             network: task.network,
+            custom_metadata: task.custom_metadata,
             kind: match task.kind {
                 KindDump::DocumentImport {
                     primary_key,
@@ -232,6 +235,9 @@ impl<'a> Dump<'a> {
                     }
                 }
                 KindDump::UpgradeDatabase { from } => KindWithContent::UpgradeDatabase { from },
+                KindDump::IndexCompaction { index_uid } => {
+                    KindWithContent::IndexCompaction { index_uid }
+                }
             },
         };
 

crates/index-scheduler/src/error.rs

@@ -5,6 +5,7 @@ use meilisearch_types::error::{Code, ErrorCode};
 use meilisearch_types::milli::index::RollbackOutcome;
 use meilisearch_types::tasks::{Kind, Status};
 use meilisearch_types::{heed, milli};
+use reqwest::StatusCode;
 use thiserror::Error;
 
 use crate::TaskId;
@@ -127,6 +128,14 @@ pub enum Error {
     #[error("Aborted task")]
     AbortedTask,
 
+    #[error("S3 error: status: {status}, body: {body}")]
+    S3Error { status: StatusCode, body: String },
+    #[error("S3 HTTP error: {0}")]
+    S3HttpError(reqwest::Error),
+    #[error("S3 XML error: {0}")]
+    S3XmlError(Box<dyn std::error::Error + Send + Sync>),
+    #[error("S3 bucket error: {0}")]
+    S3BucketError(rusty_s3::BucketError),
     #[error(transparent)]
     Dump(#[from] dump::Error),
     #[error(transparent)]
@@ -226,6 +235,10 @@ impl Error {
             | Error::TaskCancelationWithEmptyQuery
             | Error::FromRemoteWhenExporting { .. }
             | Error::AbortedTask
+            | Error::S3Error { .. }
+            | Error::S3HttpError(_)
+            | Error::S3XmlError(_)
+            | Error::S3BucketError(_)
             | Error::Dump(_)
             | Error::Heed(_)
             | Error::Milli { .. }
@@ -293,8 +306,14 @@ impl ErrorCode for Error {
             Error::BatchNotFound(_) => Code::BatchNotFound,
             Error::TaskDeletionWithEmptyQuery => Code::MissingTaskFilters,
             Error::TaskCancelationWithEmptyQuery => Code::MissingTaskFilters,
-            // TODO: not sure of the Code to use
             Error::NoSpaceLeftInTaskQueue => Code::NoSpaceLeftOnDevice,
+            Error::S3Error { status, .. } if status.is_client_error() => {
+                Code::InvalidS3SnapshotRequest
+            }
+            Error::S3Error { .. } => Code::S3SnapshotServerError,
+            Error::S3HttpError(_) => Code::S3SnapshotServerError,
+            Error::S3XmlError(_) => Code::S3SnapshotServerError,
+            Error::S3BucketError(_) => Code::InvalidS3SnapshotParameters,
             Error::Dump(e) => e.error_code(),
             Error::Milli { error, .. } => error.error_code(),
             Error::ProcessBatchPanicked(_) => Code::Internal,

crates/index-scheduler/src/index_mapper/mod.rs

@@ -199,7 +199,7 @@ impl IndexMapper {
                 let uuid = Uuid::new_v4();
                 self.index_mapping.put(&mut wtxn, name, &uuid)?;
 
-                let index_path = self.base_path.join(uuid.to_string());
+                let index_path = self.index_path(uuid);
                 fs::create_dir_all(&index_path)?;
 
                 // Error if the UUIDv4 somehow already exists in the map, since it should be fresh.
@@ -286,7 +286,7 @@ impl IndexMapper {
         };
 
         let index_map = self.index_map.clone();
-        let index_path = self.base_path.join(uuid.to_string());
+        let index_path = self.index_path(uuid);
         let index_name = name.to_string();
         thread::Builder::new()
             .name(String::from("index_deleter"))
@@ -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)) =
@@ -388,7 +408,7 @@ impl IndexMapper {
                     } else {
                         continue;
                     };
-                    let index_path = self.base_path.join(uuid.to_string());
+                    let index_path = self.index_path(uuid);
                     // take the lock to reopen the environment.
                     reopen
                         .reopen(&mut self.index_map.write().unwrap(), &index_path)
@@ -405,7 +425,7 @@ impl IndexMapper {
                     // if it's not already there.
                     match index_map.get(&uuid) {
                         Missing => {
-                            let index_path = self.base_path.join(uuid.to_string());
+                            let index_path = self.index_path(uuid);
 
                             break index_map
                                 .create(
@@ -432,6 +452,14 @@ impl IndexMapper {
         Ok(index)
     }
 
+    /// Returns the path of the index.
+    ///
+    /// The folder located at this path is containing the data.mdb,
+    /// the lock.mdb and an optional data.mdb.cpy file.
+    pub fn index_path(&self, uuid: Uuid) -> PathBuf {
+        self.base_path.join(uuid.to_string())
+    }
+
     pub fn rollback_index(
         &self,
         rtxn: &RoTxn,
@@ -472,7 +500,7 @@ impl IndexMapper {
             };
         }
 
-        let index_path = self.base_path.join(uuid.to_string());
+        let index_path = self.index_path(uuid);
         Index::rollback(milli::heed::EnvOpenOptions::new().read_txn_without_tls(), index_path, to)
             .map_err(|err| crate::Error::from_milli(err, Some(name.to_string())))
     }

crates/index-scheduler/src/insta_snapshot.rs

@@ -36,6 +36,7 @@ pub fn snapshot_index_scheduler(scheduler: &IndexScheduler) -> String {
         run_loop_iteration: _,
         embedders: _,
         chat_settings: _,
+        runtime: _,
     } = scheduler;
 
     let rtxn = env.read_txn().unwrap();
@@ -231,6 +232,7 @@ pub fn snapshot_task(task: &Task) -> String {
         status,
         kind,
         network,
+        custom_metadata,
     } = task;
     snap.push('{');
     snap.push_str(&format!("uid: {uid}, "));
@@ -251,6 +253,9 @@ pub fn snapshot_task(task: &Task) -> String {
     if let Some(network) = network {
         snap.push_str(&format!("network: {network:?}, "))
     }
+    if let Some(custom_metadata) = custom_metadata {
+        snap.push_str(&format!("custom_metadata: {custom_metadata:?}"))
+    }
 
     snap.push('}');
     snap
@@ -317,6 +322,9 @@ 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:?} }}")
+        }
     }
 }
 

crates/index-scheduler/src/lib.rs

@@ -1,3 +1,6 @@
+// The main Error type is large and boxing the large variant make the pattern matching fails
+#![allow(clippy::result_large_err)]
+
 /*!
 This crate defines the index scheduler, which is responsible for:
 1. Keeping references to meilisearch's indexes and mapping them to their
@@ -213,6 +216,9 @@ pub struct IndexScheduler {
     /// A counter that is incremented before every call to [`tick`](IndexScheduler::tick)
     #[cfg(test)]
     run_loop_iteration: Arc<RwLock<usize>>,
+
+    /// The tokio runtime used for asynchronous tasks.
+    runtime: Option<tokio::runtime::Handle>,
 }
 
 impl IndexScheduler {
@@ -239,6 +245,7 @@ impl IndexScheduler {
             run_loop_iteration: self.run_loop_iteration.clone(),
             features: self.features.clone(),
             chat_settings: self.chat_settings,
+            runtime: self.runtime.clone(),
         }
     }
 
@@ -252,13 +259,23 @@ impl IndexScheduler {
     }
 
     /// Create an index scheduler and start its run loop.
-    #[allow(private_interfaces)] // because test_utils is private
     pub fn new(
         options: IndexSchedulerOptions,
         auth_env: Env<WithoutTls>,
         from_db_version: (u32, u32, u32),
-        #[cfg(test)] test_breakpoint_sdr: crossbeam_channel::Sender<(test_utils::Breakpoint, bool)>,
-        #[cfg(test)] planned_failures: Vec<(usize, test_utils::FailureLocation)>,
+        runtime: Option<tokio::runtime::Handle>,
+    ) -> Result<Self> {
+        let this = Self::new_without_run(options, auth_env, from_db_version, runtime)?;
+
+        this.run();
+        Ok(this)
+    }
+
+    fn new_without_run(
+        options: IndexSchedulerOptions,
+        auth_env: Env<WithoutTls>,
+        from_db_version: (u32, u32, u32),
+        runtime: Option<tokio::runtime::Handle>,
     ) -> Result<Self> {
         std::fs::create_dir_all(&options.tasks_path)?;
         std::fs::create_dir_all(&options.update_file_path)?;
@@ -313,8 +330,7 @@ impl IndexScheduler {
 
         wtxn.commit()?;
 
-        // allow unreachable_code to get rids of the warning in the case of a test build.
-        let this = Self {
+        Ok(Self {
             processing_tasks: Arc::new(RwLock::new(ProcessingTasks::new())),
             version,
             queue,
@@ -330,21 +346,38 @@ impl IndexScheduler {
             webhooks: Arc::new(webhooks),
             embedders: Default::default(),
 
-            #[cfg(test)]
-            test_breakpoint_sdr,
-            #[cfg(test)]
-            planned_failures,
+            #[cfg(test)] // Will be replaced in `new_tests` in test environments
+            test_breakpoint_sdr: crossbeam_channel::bounded(0).0,
+            #[cfg(test)] // Will be replaced in `new_tests` in test environments
+            planned_failures: Default::default(),
             #[cfg(test)]
             run_loop_iteration: Arc::new(RwLock::new(0)),
             features,
             chat_settings,
-        };
+            runtime,
+        })
+    }
+
+    /// Create an index scheduler and start its run loop.
+    #[cfg(test)]
+    fn new_test(
+        options: IndexSchedulerOptions,
+        auth_env: Env<WithoutTls>,
+        from_db_version: (u32, u32, u32),
+        runtime: Option<tokio::runtime::Handle>,
+        test_breakpoint_sdr: crossbeam_channel::Sender<(test_utils::Breakpoint, bool)>,
+        planned_failures: Vec<(usize, test_utils::FailureLocation)>,
+    ) -> Result<Self> {
+        let mut this = Self::new_without_run(options, auth_env, from_db_version, runtime)?;
+
+        this.test_breakpoint_sdr = test_breakpoint_sdr;
+        this.planned_failures = planned_failures;
 
         this.run();
         Ok(this)
     }
 
-    fn read_txn(&self) -> Result<RoTxn<WithoutTls>> {
+    fn read_txn(&self) -> Result<RoTxn<'_, WithoutTls>> {
         self.env.read_txn().map_err(|e| e.into())
     }
 
@@ -723,6 +756,19 @@ impl IndexScheduler {
         kind: KindWithContent,
         task_id: Option<TaskId>,
         dry_run: bool,
+    ) -> Result<Task> {
+        self.register_with_custom_metadata(kind, task_id, None, dry_run)
+    }
+
+    /// Register a new task in the scheduler, with metadata.
+    ///
+    /// If it fails and data was associated with the task, it tries to delete the associated data.
+    pub fn register_with_custom_metadata(
+        &self,
+        kind: KindWithContent,
+        task_id: Option<TaskId>,
+        custom_metadata: Option<String>,
+        dry_run: bool,
     ) -> Result<Task> {
         // if the task doesn't delete or cancel anything and 40% of the task queue is full, we must refuse to enqueue the incoming task
         if !matches!(&kind, KindWithContent::TaskDeletion { tasks, .. } | KindWithContent::TaskCancelation { tasks, .. } if !tasks.is_empty())
@@ -733,7 +779,7 @@ impl IndexScheduler {
         }
 
         let mut wtxn = self.env.write_txn()?;
-        let task = self.queue.register(&mut wtxn, &kind, task_id, dry_run)?;
+        let task = self.queue.register(&mut wtxn, &kind, task_id, custom_metadata, dry_run)?;
 
         // If the registered task is a task cancelation
         // we inform the processing tasks to stop (if necessary).
@@ -757,7 +803,7 @@ impl IndexScheduler {
 
     /// Register a new task coming from a dump in the scheduler.
     /// By taking a mutable ref we're pretty sure no one will ever import a dump while actix is running.
-    pub fn register_dumped_task(&mut self) -> Result<Dump> {
+    pub fn register_dumped_task(&mut self) -> Result<Dump<'_>> {
         Dump::new(self)
     }
 
@@ -806,10 +852,8 @@ impl IndexScheduler {
                                 .queue
                                 .tasks
                                 .get_task(self.rtxn, task_id)
-                                .map_err(|err| io::Error::new(io::ErrorKind::Other, err))?
-                                .ok_or_else(|| {
-                                    io::Error::new(io::ErrorKind::Other, Error::CorruptedTaskQueue)
-                                })?;
+                                .map_err(io::Error::other)?
+                                .ok_or_else(|| io::Error::other(Error::CorruptedTaskQueue))?;
 
                             serde_json::to_writer(&mut self.buffer, &TaskView::from_task(&task))?;
                             self.buffer.push(b'\n');

crates/index-scheduler/src/processing.rs

@@ -75,6 +75,7 @@ make_enum_progress! {
     pub enum TaskCancelationProgress {
         RetrievingTasks,
         CancelingUpgrade,
+        CleaningCompactionLeftover,
         UpdatingTasks,
     }
 }
@@ -138,6 +139,17 @@ make_enum_progress! {
     }
 }
 
+make_enum_progress! {
+    pub enum IndexCompaction {
+        RetrieveTheIndex,
+        CreateTemporaryFile,
+        CopyAndCompactTheIndex,
+        PersistTheCompactedIndex,
+        CloseTheIndex,
+        ReopenTheIndex,
+    }
+}
+
 make_enum_progress! {
     pub enum InnerSwappingTwoIndexes {
         RetrieveTheTasks,

crates/index-scheduler/src/queue/mod.rs

@@ -257,6 +257,7 @@ impl Queue {
         wtxn: &mut RwTxn,
         kind: &KindWithContent,
         task_id: Option<TaskId>,
+        custom_metadata: Option<String>,
         dry_run: bool,
     ) -> Result<Task> {
         let next_task_id = self.tasks.next_task_id(wtxn)?;
@@ -280,6 +281,7 @@ impl Queue {
             status: Status::Enqueued,
             kind: kind.clone(),
             network: None,
+            custom_metadata,
         };
         // For deletion and cancelation tasks, we want to make extra sure that they
         // don't attempt to delete/cancel tasks that are newer than themselves.
@@ -310,7 +312,8 @@ impl Queue {
             | self.tasks.status.get(wtxn, &Status::Failed)?.unwrap_or_default()
             | self.tasks.status.get(wtxn, &Status::Canceled)?.unwrap_or_default();
 
-        let to_delete = RoaringBitmap::from_iter(finished.into_iter().rev().take(100_000));
+        let to_delete =
+            RoaringBitmap::from_sorted_iter(finished.into_iter().take(100_000)).unwrap();
 
         // /!\ the len must be at least 2 or else we might enter an infinite loop where we only delete
         //     the deletion tasks we enqueued ourselves.
@@ -326,7 +329,7 @@ impl Queue {
         );
 
         // it's safe to unwrap here because we checked the len above
-        let newest_task_id = to_delete.iter().last().unwrap();
+        let newest_task_id = to_delete.iter().next_back().unwrap();
         let last_task_to_delete =
             self.tasks.get_task(wtxn, newest_task_id)?.ok_or(Error::CorruptedTaskQueue)?;
 
@@ -343,6 +346,7 @@ impl Queue {
                 tasks: to_delete,
             },
             None,
+            None,
             false,
         )?;
 

crates/index-scheduler/src/scheduler/autobatcher.rs

@@ -68,13 +68,14 @@ impl From<KindWithContent> for AutobatchKind {
             KindWithContent::IndexCreation { .. } => AutobatchKind::IndexCreation,
             KindWithContent::IndexUpdate { .. } => AutobatchKind::IndexUpdate,
             KindWithContent::IndexSwap { .. } => AutobatchKind::IndexSwap,
-            KindWithContent::TaskCancelation { .. }
+            KindWithContent::IndexCompaction { .. }
+            | KindWithContent::TaskCancelation { .. }
             | KindWithContent::TaskDeletion { .. }
             | KindWithContent::DumpCreation { .. }
             | KindWithContent::Export { .. }
             | KindWithContent::UpgradeDatabase { .. }
             | KindWithContent::SnapshotCreation => {
-                panic!("The autobatcher should never be called with tasks that don't apply to an index.")
+                panic!("The autobatcher should never be called with tasks with special priority or that don't apply to an index.")
             }
         }
     }
@@ -287,8 +288,10 @@ 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) => {
+                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 }))

crates/index-scheduler/src/scheduler/create_batch.rs

@@ -55,6 +55,10 @@ pub(crate) enum Batch {
     UpgradeDatabase {
         tasks: Vec<Task>,
     },
+    IndexCompaction {
+        index_uid: String,
+        task: Task,
+    },
 }
 
 #[derive(Debug)]
@@ -66,6 +70,7 @@ pub(crate) enum DocumentOperation {
 
 /// A [batch](Batch) that combines multiple tasks operating on an index.
 #[derive(Debug)]
+#[allow(clippy::large_enum_variant)]
 pub(crate) enum IndexOperation {
     DocumentOperation {
         index_uid: String,
@@ -109,7 +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)
@@ -154,7 +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),
         }
     }
 }
@@ -174,6 +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")?,
         };
@@ -511,17 +519,33 @@ impl IndexScheduler {
             return Ok(Some((Batch::TaskDeletions(tasks), current_batch)));
         }
 
-        // 3. we batch the export.
+        // 3. we get the next task to compact
+        let to_compact = self.queue.tasks.get_kind(rtxn, Kind::IndexCompaction)? & enqueued;
+        if let Some(task_id) = to_compact.min() {
+            let mut task =
+                self.queue.tasks.get_task(rtxn, task_id)?.ok_or(Error::CorruptedTaskQueue)?;
+            current_batch.processing(Some(&mut task));
+            current_batch.reason(BatchStopReason::TaskCannotBeBatched {
+                kind: Kind::IndexCompaction,
+                id: task_id,
+            });
+            let index_uid =
+                task.index_uid().expect("Compaction task must have an index uid").to_owned();
+            return Ok(Some((Batch::IndexCompaction { index_uid, task }, current_batch)));
+        }
+
+        // 4. we batch the export.
         let to_export = self.queue.tasks.get_kind(rtxn, Kind::Export)? & enqueued;
         if !to_export.is_empty() {
             let task_id = to_export.iter().next().expect("There must be at least one export task");
             let mut task = self.queue.tasks.get_task(rtxn, task_id)?.unwrap();
             current_batch.processing([&mut task]);
-            current_batch.reason(BatchStopReason::TaskKindCannotBeBatched { kind: Kind::Export });
+            current_batch
+                .reason(BatchStopReason::TaskCannotBeBatched { kind: Kind::Export, id: task_id });
             return Ok(Some((Batch::Export { task }, current_batch)));
         }
 
-        // 4. we batch the snapshot.
+        // 5. we batch the snapshot.
         let to_snapshot = self.queue.tasks.get_kind(rtxn, Kind::SnapshotCreation)? & enqueued;
         if !to_snapshot.is_empty() {
             let mut tasks = self.queue.tasks.get_existing_tasks(rtxn, to_snapshot)?;
@@ -531,7 +555,7 @@ impl IndexScheduler {
             return Ok(Some((Batch::SnapshotCreation(tasks), current_batch)));
         }
 
-        // 5. we batch the dumps.
+        // 6. we batch the dumps.
         let to_dump = self.queue.tasks.get_kind(rtxn, Kind::DumpCreation)? & enqueued;
         if let Some(to_dump) = to_dump.min() {
             let mut task =
@@ -544,7 +568,7 @@ impl IndexScheduler {
             return Ok(Some((Batch::Dump(task), current_batch)));
         }
 
-        // 6. We make a batch from the unprioritised tasks. Start by taking the next enqueued task.
+        // 7. We make a batch from the unprioritised tasks. Start by taking the next enqueued task.
         let task_id = if let Some(task_id) = enqueued.min() { task_id } else { return Ok(None) };
         let mut task =
             self.queue.tasks.get_task(rtxn, task_id)?.ok_or(Error::CorruptedTaskQueue)?;

crates/index-scheduler/src/scheduler/mod.rs

@@ -25,6 +25,7 @@ use convert_case::{Case, Casing as _};
 use meilisearch_types::error::ResponseError;
 use meilisearch_types::heed::{Env, WithoutTls};
 use meilisearch_types::milli;
+use meilisearch_types::milli::update::S3SnapshotOptions;
 use meilisearch_types::tasks::Status;
 use process_batch::ProcessBatchInfo;
 use rayon::current_num_threads;
@@ -87,11 +88,14 @@ pub struct Scheduler {
 
     /// Snapshot compaction status.
     pub(crate) experimental_no_snapshot_compaction: bool,
+
+    /// S3 Snapshot options.
+    pub(crate) s3_snapshot_options: Option<S3SnapshotOptions>,
 }
 
 impl Scheduler {
-    pub(crate) fn private_clone(&self) -> Scheduler {
-        Scheduler {
+    pub(crate) fn private_clone(&self) -> Self {
+        Self {
             must_stop_processing: self.must_stop_processing.clone(),
             wake_up: self.wake_up.clone(),
             autobatching_enabled: self.autobatching_enabled,
@@ -103,23 +107,52 @@ impl Scheduler {
             version_file_path: self.version_file_path.clone(),
             embedding_cache_cap: self.embedding_cache_cap,
             experimental_no_snapshot_compaction: self.experimental_no_snapshot_compaction,
+            s3_snapshot_options: self.s3_snapshot_options.clone(),
         }
     }
 
     pub fn new(options: &IndexSchedulerOptions, auth_env: Env<WithoutTls>) -> Scheduler {
+        let IndexSchedulerOptions {
+            version_file_path,
+            auth_path: _,
+            tasks_path: _,
+            update_file_path: _,
+            indexes_path: _,
+            snapshots_path,
+            dumps_path,
+            cli_webhook_url: _,
+            cli_webhook_authorization: _,
+            task_db_size: _,
+            index_base_map_size: _,
+            enable_mdb_writemap: _,
+            index_growth_amount: _,
+            index_count: _,
+            indexer_config,
+            autobatching_enabled,
+            cleanup_enabled: _,
+            max_number_of_tasks: _,
+            max_number_of_batched_tasks,
+            batched_tasks_size_limit,
+            instance_features: _,
+            auto_upgrade: _,
+            embedding_cache_cap,
+            experimental_no_snapshot_compaction,
+        } = options;
+
         Scheduler {
             must_stop_processing: MustStopProcessing::default(),
             // we want to start the loop right away in case meilisearch was ctrl+Ced while processing things
             wake_up: Arc::new(SignalEvent::auto(true)),
-            autobatching_enabled: options.autobatching_enabled,
-            max_number_of_batched_tasks: options.max_number_of_batched_tasks,
-            batched_tasks_size_limit: options.batched_tasks_size_limit,
-            dumps_path: options.dumps_path.clone(),
-            snapshots_path: options.snapshots_path.clone(),
+            autobatching_enabled: *autobatching_enabled,
+            max_number_of_batched_tasks: *max_number_of_batched_tasks,
+            batched_tasks_size_limit: *batched_tasks_size_limit,
+            dumps_path: dumps_path.clone(),
+            snapshots_path: snapshots_path.clone(),
             auth_env,
-            version_file_path: options.version_file_path.clone(),
-            embedding_cache_cap: options.embedding_cache_cap,
-            experimental_no_snapshot_compaction: options.experimental_no_snapshot_compaction,
+            version_file_path: version_file_path.clone(),
+            embedding_cache_cap: *embedding_cache_cap,
+            experimental_no_snapshot_compaction: *experimental_no_snapshot_compaction,
+            s3_snapshot_options: indexer_config.s3_snapshot_options.clone(),
         }
     }
 }

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

@@ -1,22 +1,27 @@
 use std::collections::{BTreeSet, HashMap, HashSet};
+use std::fs::{remove_file, File};
+use std::io::{ErrorKind, 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, TempPath};
 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,
@@ -24,6 +29,9 @@ use crate::utils::{
 };
 use crate::{Error, IndexScheduler, Result, TaskId};
 
+/// The name of the copy of the data.mdb file used during compaction.
+const DATA_MDB_COPY_NAME: &str = "data.mdb.cpy";
+
 #[derive(Debug, Default)]
 pub struct ProcessBatchInfo {
     /// The write channel congestion. None when unavailable: settings update.
@@ -418,6 +426,47 @@ 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 +542,92 @@ 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 src_path = index.path().join("data.mdb");
+        let pre_size = std::fs::metadata(&src_path)?.len();
+
+        let dst_path = TempPath::from_path(index.path().join(DATA_MDB_COPY_NAME));
+        let file = File::create(&dst_path)?;
+        let mut file = tempfile::NamedTempFile::from_parts(file, dst_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(src_path) {
+            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,
@@ -780,9 +915,10 @@ impl IndexScheduler {
 
         let enqueued_tasks = &self.queue.tasks.get_status(rtxn, Status::Enqueued)?;
 
-        // 0. Check if any upgrade task was matched.
+        // 0. Check if any upgrade or compaction tasks were matched.
         //    If so, we cancel all the failed or enqueued upgrade tasks.
         let upgrade_tasks = &self.queue.tasks.get_kind(rtxn, Kind::UpgradeDatabase)?;
+        let compaction_tasks = &self.queue.tasks.get_kind(rtxn, Kind::IndexCompaction)?;
         let is_canceling_upgrade = !matched_tasks.is_disjoint(upgrade_tasks);
         if is_canceling_upgrade {
             let failed_tasks = self.queue.tasks.get_status(rtxn, Status::Failed)?;
@@ -847,7 +983,33 @@ impl IndexScheduler {
             }
         }
 
-        // 3. We now have a list of tasks to cancel, cancel them
+        // 3. If we are cancelling a compaction task, remove the tempfiles after incomplete compactions
+        for compaction_task in &tasks_to_cancel & compaction_tasks {
+            progress.update_progress(TaskCancelationProgress::CleaningCompactionLeftover);
+            let task = self.queue.tasks.get_task(rtxn, compaction_task)?.unwrap();
+            let Some(Details::IndexCompaction {
+                index_uid,
+                pre_compaction_size: _,
+                post_compaction_size: _,
+            }) = task.details
+            else {
+                unreachable!("wrong details for compaction task {compaction_task}")
+            };
+
+            let index_path = match self.index_mapper.index_mapping.get(rtxn, &index_uid)? {
+                Some(index_uuid) => self.index_mapper.index_path(index_uuid),
+                None => continue,
+            };
+
+            if let Err(e) = remove_file(index_path.join(DATA_MDB_COPY_NAME)) {
+                match e.kind() {
+                    ErrorKind::NotFound => (),
+                    _ => return Err(Error::IoError(e)),
+                }
+            }
+        }
+
+        // 4. We now have a list of tasks to cancel, cancel them
         let (task_progress, progress_obj) = AtomicTaskStep::new(tasks_to_cancel.len() as u32);
         progress.update_progress(progress_obj);
 

crates/index-scheduler/src/scheduler/process_export.rs

@@ -370,7 +370,7 @@ fn ureq_error_into_error(error: ureq::Error) -> Error {
             }
             Err(e) => e.into(),
         },
-        ureq::Error::Transport(transport) => io::Error::new(io::ErrorKind::Other, transport).into(),
+        ureq::Error::Transport(transport) => io::Error::other(transport).into(),
     }
 }
 

crates/index-scheduler/src/scheduler/process_snapshot_creation.rs

@@ -12,6 +12,8 @@ use crate::processing::{AtomicUpdateFileStep, SnapshotCreationProgress};
 use crate::queue::TaskQueue;
 use crate::{Error, IndexScheduler, Result};
 
+const UPDATE_FILES_DIR_NAME: &str = "update_files";
+
 /// # Safety
 ///
 /// See [`EnvOpenOptions::open`].
@@ -78,10 +80,32 @@ impl IndexScheduler {
     pub(super) fn process_snapshot(
         &self,
         progress: Progress,
-        mut tasks: Vec<Task>,
+        tasks: Vec<Task>,
     ) -> Result<Vec<Task>> {
         progress.update_progress(SnapshotCreationProgress::StartTheSnapshotCreation);
 
+        match self.scheduler.s3_snapshot_options.clone() {
+            Some(options) => {
+                #[cfg(not(unix))]
+                {
+                    let _ = options;
+                    panic!("Non-unix platform does not support S3 snapshotting");
+                }
+                #[cfg(unix)]
+                self.runtime
+                    .as_ref()
+                    .expect("Runtime not initialized")
+                    .block_on(self.process_snapshot_to_s3(progress, options, tasks))
+            }
+            None => self.process_snapshots_to_disk(progress, tasks),
+        }
+    }
+
+    fn process_snapshots_to_disk(
+        &self,
+        progress: Progress,
+        mut tasks: Vec<Task>,
+    ) -> Result<Vec<Task>, Error> {
         fs::create_dir_all(&self.scheduler.snapshots_path)?;
         let temp_snapshot_dir = tempfile::tempdir()?;
 
@@ -128,7 +152,7 @@ impl IndexScheduler {
         let rtxn = self.env.read_txn()?;
 
         // 2.4 Create the update files directory
-        let update_files_dir = temp_snapshot_dir.path().join("update_files");
+        let update_files_dir = temp_snapshot_dir.path().join(UPDATE_FILES_DIR_NAME);
         fs::create_dir_all(&update_files_dir)?;
 
         // 2.5 Only copy the update files of the enqueued tasks
@@ -140,7 +164,7 @@ impl IndexScheduler {
             let task =
                 self.queue.tasks.get_task(&rtxn, task_id)?.ok_or(Error::CorruptedTaskQueue)?;
             if let Some(content_uuid) = task.content_uuid() {
-                let src = self.queue.file_store.get_update_path(content_uuid);
+                let src = self.queue.file_store.update_path(content_uuid);
                 let dst = update_files_dir.join(content_uuid.to_string());
                 fs::copy(src, dst)?;
             }
@@ -206,4 +230,403 @@ impl IndexScheduler {
 
         Ok(tasks)
     }
+
+    #[cfg(unix)]
+    pub(super) async fn process_snapshot_to_s3(
+        &self,
+        progress: Progress,
+        opts: meilisearch_types::milli::update::S3SnapshotOptions,
+        mut tasks: Vec<Task>,
+    ) -> Result<Vec<Task>> {
+        use meilisearch_types::milli::update::S3SnapshotOptions;
+
+        let S3SnapshotOptions {
+            s3_bucket_url,
+            s3_bucket_region,
+            s3_bucket_name,
+            s3_snapshot_prefix,
+            s3_access_key,
+            s3_secret_key,
+            s3_max_in_flight_parts,
+            s3_compression_level: level,
+            s3_signature_duration,
+            s3_multipart_part_size,
+        } = opts;
+
+        let must_stop_processing = self.scheduler.must_stop_processing.clone();
+        let retry_backoff = backoff::ExponentialBackoff::default();
+        let db_name = {
+            let mut base_path = self.env.path().to_owned();
+            base_path.pop();
+            base_path.file_name().and_then(OsStr::to_str).unwrap_or("data.ms").to_string()
+        };
+
+        let (reader, writer) = std::io::pipe()?;
+        let uploader_task = tokio::spawn(multipart_stream_to_s3(
+            s3_bucket_url,
+            s3_bucket_region,
+            s3_bucket_name,
+            s3_snapshot_prefix,
+            s3_access_key,
+            s3_secret_key,
+            s3_max_in_flight_parts,
+            s3_signature_duration,
+            s3_multipart_part_size,
+            must_stop_processing,
+            retry_backoff,
+            db_name,
+            reader,
+        ));
+
+        let index_scheduler = IndexScheduler::private_clone(self);
+        let builder_task = tokio::task::spawn_blocking(move || {
+            stream_tarball_into_pipe(progress, level, writer, index_scheduler)
+        });
+
+        let (uploader_result, builder_result) = tokio::join!(uploader_task, builder_task);
+
+        // Check uploader result first to early return on task abortion.
+        // safety: JoinHandle can return an error if the task was aborted, cancelled, or panicked.
+        uploader_result.unwrap()?;
+        builder_result.unwrap()?;
+
+        for task in &mut tasks {
+            task.status = Status::Succeeded;
+        }
+
+        Ok(tasks)
+    }
+}
+
+/// Streams a tarball of the database content into a pipe.
+#[cfg(unix)]
+fn stream_tarball_into_pipe(
+    progress: Progress,
+    level: u32,
+    writer: std::io::PipeWriter,
+    index_scheduler: IndexScheduler,
+) -> std::result::Result<(), Error> {
+    use std::io::Write as _;
+    use std::path::Path;
+
+    let writer = flate2::write::GzEncoder::new(writer, flate2::Compression::new(level));
+    let mut tarball = tar::Builder::new(writer);
+
+    // 1. Snapshot the version file
+    tarball
+        .append_path_with_name(&index_scheduler.scheduler.version_file_path, VERSION_FILE_NAME)?;
+
+    // 2. Snapshot the index scheduler LMDB env
+    progress.update_progress(SnapshotCreationProgress::SnapshotTheIndexScheduler);
+    let tasks_env_file = index_scheduler.env.try_clone_inner_file()?;
+    let path = Path::new("tasks").join("data.mdb");
+    append_file_to_tarball(&mut tarball, path, tasks_env_file)?;
+
+    // 2.3 Create a read transaction on the index-scheduler
+    let rtxn = index_scheduler.env.read_txn()?;
+
+    // 2.4 Create the update files directory
+    //     And only copy the update files of the enqueued tasks
+    progress.update_progress(SnapshotCreationProgress::SnapshotTheUpdateFiles);
+    let enqueued = index_scheduler.queue.tasks.get_status(&rtxn, Status::Enqueued)?;
+    let (atomic, update_file_progress) = AtomicUpdateFileStep::new(enqueued.len() as u32);
+    progress.update_progress(update_file_progress);
+
+    // We create the update_files directory so that it
+    // always exists even if there are no update files
+    let update_files_dir = Path::new(UPDATE_FILES_DIR_NAME);
+    let src_update_files_dir = {
+        let mut path = index_scheduler.env.path().to_path_buf();
+        path.pop();
+        path.join(UPDATE_FILES_DIR_NAME)
+    };
+    tarball.append_dir(update_files_dir, src_update_files_dir)?;
+
+    for task_id in enqueued {
+        let task = index_scheduler
+            .queue
+            .tasks
+            .get_task(&rtxn, task_id)?
+            .ok_or(Error::CorruptedTaskQueue)?;
+        if let Some(content_uuid) = task.content_uuid() {
+            use std::fs::File;
+
+            let src = index_scheduler.queue.file_store.update_path(content_uuid);
+            let mut update_file = File::open(src)?;
+            let path = update_files_dir.join(content_uuid.to_string());
+            tarball.append_file(path, &mut update_file)?;
+        }
+        atomic.fetch_add(1, Ordering::Relaxed);
+    }
+
+    // 3. Snapshot every indexes
+    progress.update_progress(SnapshotCreationProgress::SnapshotTheIndexes);
+    let index_mapping = index_scheduler.index_mapper.index_mapping;
+    let nb_indexes = index_mapping.len(&rtxn)? as u32;
+    let indexes_dir = Path::new("indexes");
+    let indexes_references: Vec<_> = index_scheduler
+        .index_mapper
+        .index_mapping
+        .iter(&rtxn)?
+        .map(|res| res.map_err(Error::from).map(|(name, uuid)| (name.to_string(), uuid)))
+        .collect::<Result<_, Error>>()?;
+
+    // It's prettier to use a for loop instead of the IndexMapper::try_for_each_index
+    // method, especially when we need to access the UUID, local path and index number.
+    for (i, (name, uuid)) in indexes_references.into_iter().enumerate() {
+        progress.update_progress(VariableNameStep::<SnapshotCreationProgress>::new(
+            &name, i as u32, nb_indexes,
+        ));
+        let path = indexes_dir.join(uuid.to_string()).join("data.mdb");
+        let index = index_scheduler.index_mapper.index(&rtxn, &name)?;
+        let index_file = index.try_clone_inner_file()?;
+        tracing::trace!("Appending index file for {name} in {}", path.display());
+        append_file_to_tarball(&mut tarball, path, index_file)?;
+    }
+
+    drop(rtxn);
+
+    // 4. Snapshot the auth LMDB env
+    progress.update_progress(SnapshotCreationProgress::SnapshotTheApiKeys);
+    let auth_env_file = index_scheduler.scheduler.auth_env.try_clone_inner_file()?;
+    let path = Path::new("auth").join("data.mdb");
+    append_file_to_tarball(&mut tarball, path, auth_env_file)?;
+
+    let mut gzencoder = tarball.into_inner()?;
+    gzencoder.flush()?;
+    gzencoder.try_finish()?;
+    let mut writer = gzencoder.finish()?;
+    writer.flush()?;
+
+    Result::<_, Error>::Ok(())
+}
+
+#[cfg(unix)]
+fn append_file_to_tarball<W, P>(
+    tarball: &mut tar::Builder<W>,
+    path: P,
+    mut auth_env_file: fs::File,
+) -> Result<(), Error>
+where
+    W: std::io::Write,
+    P: AsRef<std::path::Path>,
+{
+    use std::io::{Seek as _, SeekFrom};
+
+    // Note: A previous snapshot operation may have left the cursor
+    //       at the end of the file so we need to seek to the start.
+    auth_env_file.seek(SeekFrom::Start(0))?;
+    tarball.append_file(path, &mut auth_env_file)?;
+    Ok(())
+}
+
+/// Streams the content read from the given reader to S3.
+#[cfg(unix)]
+#[allow(clippy::too_many_arguments)]
+async fn multipart_stream_to_s3(
+    s3_bucket_url: String,
+    s3_bucket_region: String,
+    s3_bucket_name: String,
+    s3_snapshot_prefix: String,
+    s3_access_key: String,
+    s3_secret_key: String,
+    s3_max_in_flight_parts: std::num::NonZero<usize>,
+    s3_signature_duration: std::time::Duration,
+    s3_multipart_part_size: u64,
+    must_stop_processing: super::MustStopProcessing,
+    retry_backoff: backoff::exponential::ExponentialBackoff<backoff::SystemClock>,
+    db_name: String,
+    reader: std::io::PipeReader,
+) -> Result<(), Error> {
+    use std::{collections::VecDeque, os::fd::OwnedFd, path::PathBuf};
+
+    use bytes::{Bytes, BytesMut};
+    use reqwest::{Client, Response};
+    use rusty_s3::S3Action as _;
+    use rusty_s3::{actions::CreateMultipartUpload, Bucket, BucketError, Credentials, UrlStyle};
+    use tokio::task::JoinHandle;
+
+    let reader = OwnedFd::from(reader);
+    let reader = tokio::net::unix::pipe::Receiver::from_owned_fd(reader)?;
+    let s3_snapshot_prefix = PathBuf::from(s3_snapshot_prefix);
+    let url =
+        s3_bucket_url.parse().map_err(BucketError::ParseError).map_err(Error::S3BucketError)?;
+    let bucket = Bucket::new(url, UrlStyle::Path, s3_bucket_name, s3_bucket_region)
+        .map_err(Error::S3BucketError)?;
+    let credential = Credentials::new(s3_access_key, s3_secret_key);
+
+    // Note for the future (rust 1.91+): use with_added_extension, it's prettier
+    let object_path = s3_snapshot_prefix.join(format!("{db_name}.snapshot"));
+    // Note: It doesn't work on Windows and if a port to this platform is needed,
+    //       use the slash-path crate or similar to get the correct path separator.
+    let object = object_path.display().to_string();
+
+    let action = bucket.create_multipart_upload(Some(&credential), &object);
+    let url = action.sign(s3_signature_duration);
+
+    let client = Client::new();
+    let resp = client.post(url).send().await.map_err(Error::S3HttpError)?;
+    let status = resp.status();
+
+    let body = match resp.error_for_status_ref() {
+        Ok(_) => resp.text().await.map_err(Error::S3HttpError)?,
+        Err(_) => {
+            return Err(Error::S3Error { status, body: resp.text().await.unwrap_or_default() })
+        }
+    };
+
+    let multipart =
+        CreateMultipartUpload::parse_response(&body).map_err(|e| Error::S3XmlError(Box::new(e)))?;
+    tracing::debug!("Starting the upload of the snapshot to {object}");
+
+    // We use this bumpalo for etags strings.
+    let bump = bumpalo::Bump::new();
+    let mut etags = Vec::<&str>::new();
+    let mut in_flight = VecDeque::<(JoinHandle<reqwest::Result<Response>>, Bytes)>::with_capacity(
+        s3_max_in_flight_parts.get(),
+    );
+
+    // Part numbers start at 1 and cannot be larger than 10k
+    for part_number in 1u16.. {
+        if must_stop_processing.get() {
+            return Err(Error::AbortedTask);
+        }
+
+        let part_upload =
+            bucket.upload_part(Some(&credential), &object, part_number, multipart.upload_id());
+        let url = part_upload.sign(s3_signature_duration);
+
+        // Wait for a buffer to be ready if there are in-flight parts that landed
+        let mut buffer = if in_flight.len() >= s3_max_in_flight_parts.get() {
+            let (handle, buffer) = in_flight.pop_front().expect("At least one in flight request");
+            let resp = join_and_map_error(handle).await?;
+            extract_and_append_etag(&bump, &mut etags, resp.headers())?;
+
+            let mut buffer = match buffer.try_into_mut() {
+                Ok(buffer) => buffer,
+                Err(_) => unreachable!("All bytes references were consumed in the task"),
+            };
+            buffer.clear();
+            buffer
+        } else {
+            BytesMut::with_capacity(s3_multipart_part_size as usize)
+        };
+
+        // If we successfully read enough bytes,
+        // we can continue and send the buffer/part
+        while buffer.len() < (s3_multipart_part_size as usize / 2) {
+            // Wait for the pipe to be readable
+
+            use std::io;
+            reader.readable().await?;
+
+            match reader.try_read_buf(&mut buffer) {
+                Ok(0) => break,
+                // We read some bytes but maybe not enough
+                Ok(_) => continue,
+                // The readiness event is a false positive.
+                Err(ref e) if e.kind() == io::ErrorKind::WouldBlock => continue,
+                Err(e) => return Err(e.into()),
+            }
+        }
+
+        if buffer.is_empty() {
+            // Break the loop if the buffer is
+            // empty after we tried to read bytes
+            break;
+        }
+
+        let body = buffer.freeze();
+        tracing::trace!("Sending part {part_number}");
+        let task = tokio::spawn({
+            let client = client.clone();
+            let body = body.clone();
+            backoff::future::retry(retry_backoff.clone(), move || {
+                let client = client.clone();
+                let url = url.clone();
+                let body = body.clone();
+                async move {
+                    match client.put(url).body(body).send().await {
+                        Ok(resp) if resp.status().is_client_error() => {
+                            resp.error_for_status().map_err(backoff::Error::Permanent)
+                        }
+                        Ok(resp) => Ok(resp),
+                        Err(e) => Err(backoff::Error::transient(e)),
+                    }
+                }
+            })
+        });
+        in_flight.push_back((task, body));
+    }
+
+    for (handle, _buffer) in in_flight {
+        let resp = join_and_map_error(handle).await?;
+        extract_and_append_etag(&bump, &mut etags, resp.headers())?;
+    }
+
+    tracing::debug!("Finalizing the multipart upload");
+
+    let action = bucket.complete_multipart_upload(
+        Some(&credential),
+        &object,
+        multipart.upload_id(),
+        etags.iter().map(AsRef::as_ref),
+    );
+    let url = action.sign(s3_signature_duration);
+    let body = action.body();
+    let resp = backoff::future::retry(retry_backoff, move || {
+        let client = client.clone();
+        let url = url.clone();
+        let body = body.clone();
+        async move {
+            match client.post(url).body(body).send().await {
+                Ok(resp) if resp.status().is_client_error() => {
+                    resp.error_for_status().map_err(backoff::Error::Permanent)
+                }
+                Ok(resp) => Ok(resp),
+                Err(e) => Err(backoff::Error::transient(e)),
+            }
+        }
+    })
+    .await
+    .map_err(Error::S3HttpError)?;
+
+    let status = resp.status();
+    let body = resp.text().await.map_err(|e| Error::S3Error { status, body: e.to_string() })?;
+    if status.is_success() {
+        Ok(())
+    } else {
+        Err(Error::S3Error { status, body })
+    }
+}
+
+#[cfg(unix)]
+async fn join_and_map_error(
+    join_handle: tokio::task::JoinHandle<Result<reqwest::Response, reqwest::Error>>,
+) -> Result<reqwest::Response> {
+    // safety: Panic happens if the task (JoinHandle) was aborted, cancelled, or panicked
+    let request = join_handle.await.unwrap();
+    let resp = request.map_err(Error::S3HttpError)?;
+    match resp.error_for_status_ref() {
+        Ok(_) => Ok(resp),
+        Err(_) => Err(Error::S3Error {
+            status: resp.status(),
+            body: resp.text().await.unwrap_or_default(),
+        }),
+    }
+}
+
+#[cfg(unix)]
+fn extract_and_append_etag<'b>(
+    bump: &'b bumpalo::Bump,
+    etags: &mut Vec<&'b str>,
+    headers: &reqwest::header::HeaderMap,
+) -> Result<()> {
+    use reqwest::header::ETAG;
+
+    let etag = headers.get(ETAG).ok_or_else(|| Error::S3XmlError("Missing ETag header".into()))?;
+    let etag = etag.to_str().map_err(|e| Error::S3XmlError(Box::new(e)))?;
+    etags.push(bump.alloc_str(etag));
+
+    Ok(())
 }

crates/index-scheduler/src/scheduler/snapshots/test_failure.rs/upgrade_failure/after_processing_everything.snap

@@ -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, 21, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 26, 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":"v1.21.0"}, 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.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.26.0"}, 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.", }

crates/index-scheduler/src/scheduler/snapshots/test_failure.rs/upgrade_failure/register_automatic_upgrade_task.snap

@@ -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, 21, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 26, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
 ----------------------------------------------------------------------
 ### Status:
 enqueued [0,]

crates/index-scheduler/src/scheduler/snapshots/test_failure.rs/upgrade_failure/registered_a_task_while_the_upgrade_task_is_enqueued.snap

@@ -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, 21, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 26, 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:

crates/index-scheduler/src/scheduler/snapshots/test_failure.rs/upgrade_failure/upgrade_task_failed.snap

@@ -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, 21, 0) }, 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, 26, 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":"v1.21.0"}, 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.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.26.0"}, 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,]

crates/index-scheduler/src/scheduler/snapshots/test_failure.rs/upgrade_failure/upgrade_task_failed_again.snap

@@ -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, 21, 0) }, 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, 26, 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":"v1.21.0"}, 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.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.26.0"}, 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,]

crates/index-scheduler/src/scheduler/snapshots/test_failure.rs/upgrade_failure/upgrade_task_succeeded.snap

@@ -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, 21, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 26, 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":"v1.21.0"}, 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.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.26.0"}, 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,]

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

@@ -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]

crates/index-scheduler/src/test_utils.rs

@@ -126,7 +126,7 @@ impl IndexScheduler {
         std::fs::create_dir_all(&options.auth_path).unwrap();
         let auth_env = open_auth_store_env(&options.auth_path).unwrap();
         let index_scheduler =
-            Self::new(options, auth_env, version, sender, planned_failures).unwrap();
+            Self::new_test(options, auth_env, version, None, sender, planned_failures).unwrap();
 
         // To be 100% consistent between all test we're going to start the scheduler right now
         // and ensure it's in the expected starting state.

crates/index-scheduler/src/upgrade/mod.rs

@@ -45,6 +45,11 @@ pub fn upgrade_index_scheduler(
         (1, 19, _) => 0,
         (1, 20, _) => 0,
         (1, 21, _) => 0,
+        (1, 22, _) => 0,
+        (1, 23, _) => 0,
+        (1, 24, _) => 0,
+        (1, 25, _) => 0,
+        (1, 26, _) => 0,
         (major, minor, patch) => {
             if major > current_major
                 || (major == current_major && minor > current_minor)
@@ -95,6 +100,7 @@ pub fn upgrade_index_scheduler(
             status: Status::Enqueued,
             kind: KindWithContent::UpgradeDatabase { from },
             network: None,
+            custom_metadata: None,
         },
     )?;
     wtxn.commit()?;

crates/index-scheduler/src/utils.rs

@@ -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 {
@@ -378,6 +379,7 @@ impl crate::IndexScheduler {
                 status,
                 kind,
                 network: _,
+                custom_metadata: _,
             } = task;
             assert_eq!(uid, task.uid);
             if task.status != Status::Enqueued {
@@ -618,6 +620,13 @@ 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);
+                    }
                 }
             }
 

crates/index-scheduler/src/uuid_codec.rs

@@ -17,7 +17,7 @@ impl<'a> BytesDecode<'a> for UuidCodec {
 impl BytesEncode<'_> for UuidCodec {
     type EItem = Uuid;
 
-    fn bytes_encode(item: &Self::EItem) -> Result<Cow<[u8]>, BoxedError> {
+    fn bytes_encode(item: &Self::EItem) -> Result<Cow<'_, [u8]>, BoxedError> {
         Ok(Cow::Borrowed(item.as_bytes()))
     }
 }

crates/meili-snap/src/lib.rs

@@ -271,9 +271,10 @@ macro_rules! json_string {
 
 #[cfg(test)]
 mod tests {
+    use uuid::Uuid;
+
     use crate as meili_snap;
     use crate::UUID_IN_MESSAGE_RE;
-    use uuid::Uuid;
 
     #[test]
     fn snap() {

crates/meilisearch-auth/src/store.rs

@@ -109,6 +109,7 @@ impl HeedAuthStore {
                             Action::IndexesGet,
                             Action::IndexesUpdate,
                             Action::IndexesSwap,
+                            Action::IndexesCompact,
                         ]
                         .iter(),
                     );
@@ -315,7 +316,9 @@ impl<'a> heed::BytesDecode<'a> for KeyIdActionCodec {
 impl<'a> heed::BytesEncode<'a> for KeyIdActionCodec {
     type EItem = (&'a KeyId, &'a Action, Option<&'a [u8]>);
 
-    fn bytes_encode((key_id, action, index): &Self::EItem) -> StdResult<Cow<[u8]>, BoxedError> {
+    fn bytes_encode(
+        (key_id, action, index): &'_ Self::EItem,
+    ) -> StdResult<Cow<'_, [u8]>, BoxedError> {
         let mut bytes = Vec::new();
 
         bytes.extend_from_slice(key_id.as_bytes());

crates/meilisearch-types/src/error.rs

@@ -5,6 +5,7 @@ use actix_web::{self as aweb, HttpResponseBuilder};
 use aweb::http::header;
 use aweb::rt::task::JoinError;
 use convert_case::Casing;
+use milli::cellulite;
 use milli::heed::{Error as HeedError, MdbError};
 use serde::{Deserialize, Serialize};
 use utoipa::ToSchema;
@@ -239,6 +240,7 @@ InconsistentDocumentChangeHeaders              , InvalidRequest       , BAD_REQU
 InvalidDocumentFilter                          , InvalidRequest       , BAD_REQUEST ;
 InvalidDocumentSort                            , InvalidRequest       , BAD_REQUEST ;
 InvalidDocumentGeoField                        , InvalidRequest       , BAD_REQUEST ;
+InvalidDocumentGeojsonField                    , InvalidRequest       , BAD_REQUEST ;
 InvalidHeaderValue                             , InvalidRequest       , BAD_REQUEST ;
 InvalidVectorDimensions                        , InvalidRequest       , BAD_REQUEST ;
 InvalidVectorsType                             , InvalidRequest       , BAD_REQUEST ;
@@ -252,10 +254,12 @@ InvalidSearchHybridQuery                       , InvalidRequest       , BAD_REQU
 InvalidIndexLimit                              , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexOffset                             , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexPrimaryKey                         , InvalidRequest       , BAD_REQUEST ;
+InvalidIndexCustomMetadata                     , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexUid                                , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFacets                       , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFacetsByIndex                , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFacetOrder                   , InvalidRequest       , BAD_REQUEST ;
+InvalidMultiSearchQueryPersonalization         , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFederated                    , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFederationOptions            , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchMaxValuesPerFacet            , InvalidRequest       , BAD_REQUEST ;
@@ -313,6 +317,8 @@ InvalidSearchShowRankingScoreDetails           , InvalidRequest       , BAD_REQU
 InvalidSimilarShowRankingScoreDetails          , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchSort                              , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchDistinct                          , InvalidRequest       , BAD_REQUEST ;
+InvalidSearchPersonalize                       , InvalidRequest       , BAD_REQUEST ;
+InvalidSearchPersonalizeUserContext            , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchMediaAndVector                    , InvalidRequest       , BAD_REQUEST ;
 InvalidSettingsDisplayedAttributes             , InvalidRequest       , BAD_REQUEST ;
 InvalidSettingsDistinctAttribute               , InvalidRequest       , BAD_REQUEST ;
@@ -388,6 +394,9 @@ TooManyVectors                                 , InvalidRequest       , BAD_REQU
 UnretrievableDocument                          , Internal             , BAD_REQUEST ;
 UnretrievableErrorCode                         , InvalidRequest       , BAD_REQUEST ;
 UnsupportedMediaType                           , InvalidRequest       , UNSUPPORTED_MEDIA_TYPE ;
+InvalidS3SnapshotRequest                       , Internal             , BAD_REQUEST ;
+InvalidS3SnapshotParameters                    , Internal             , BAD_REQUEST ;
+S3SnapshotServerError                          , Internal             , BAD_GATEWAY ;
 
 // Experimental features
 VectorEmbeddingError                           , InvalidRequest       , BAD_REQUEST ;
@@ -501,7 +510,9 @@ impl ErrorCode for milli::Error {
                         Code::InvalidFacetSearchFacetName
                     }
                     UserError::CriterionError(_) => Code::InvalidSettingsRankingRules,
-                    UserError::InvalidGeoField { .. } => Code::InvalidDocumentGeoField,
+                    UserError::InvalidGeoField { .. } | UserError::GeoJsonError(_) => {
+                        Code::InvalidDocumentGeoField
+                    }
                     UserError::InvalidVectorDimensions { .. }
                     | UserError::InvalidIndexingVectorDimensions { .. } => {
                         Code::InvalidVectorDimensions
@@ -525,6 +536,17 @@ impl ErrorCode for milli::Error {
                     | UserError::DocumentEditionCompilationError(_) => {
                         Code::EditDocumentsByFunctionError
                     }
+                    UserError::CelluliteError(err) => match err {
+                        cellulite::Error::BuildCanceled
+                        | cellulite::Error::VersionMismatchOnBuild(_)
+                        | cellulite::Error::DatabaseDoesntExists
+                        | cellulite::Error::Heed(_)
+                        | cellulite::Error::InvalidGeometry(_)
+                        | cellulite::Error::InternalDocIdMissing(_, _)
+                        | cellulite::Error::CannotConvertLineToCell(_, _, _) => Code::Internal,
+                        cellulite::Error::InvalidGeoJson(_) => Code::InvalidDocumentGeojsonField,
+                    },
+                    UserError::MalformedGeojson(_) => Code::InvalidDocumentGeojsonField,
                 }
             }
         }
@@ -664,6 +686,18 @@ impl fmt::Display for deserr_codes::InvalidNetworkSearchApiKey {
     }
 }
 
+impl fmt::Display for deserr_codes::InvalidSearchPersonalize {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        write!(f, "the value of `personalize` is invalid, expected a JSON object with `userContext` string.")
+    }
+}
+
+impl fmt::Display for deserr_codes::InvalidSearchPersonalizeUserContext {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        write!(f, "the value of `userContext` is invalid, expected a string.")
+    }
+}
+
 #[macro_export]
 macro_rules! internal_error {
     ($target:ty : $($other:path), *) => {

crates/meilisearch-types/src/keys.rs

@@ -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]

crates/meilisearch-types/src/lib.rs

@@ -1,3 +1,5 @@
+#![allow(clippy::result_large_err)]
+
 pub mod batch_view;
 pub mod batches;
 pub mod compression;

crates/meilisearch-types/src/settings.rs

@@ -346,24 +346,26 @@ impl<T> Settings<T> {
                 continue;
             };
 
-            Self::hide_secret(api_key);
+            hide_secret(api_key, 0);
         }
     }
+}
 
-    fn hide_secret(secret: &mut String) {
-        match secret.len() {
-            x if x < 10 => {
-                secret.replace_range(.., "XXX...");
-            }
-            x if x < 20 => {
-                secret.replace_range(2.., "XXXX...");
-            }
-            x if x < 30 => {
-                secret.replace_range(3.., "XXXXX...");
-            }
-            _x => {
-                secret.replace_range(5.., "XXXXXX...");
-            }
+/// Redact a secret string, starting from the `secret_offset`th byte.
+pub fn hide_secret(secret: &mut String, secret_offset: usize) {
+    match secret.len().checked_sub(secret_offset) {
+        None => (),
+        Some(x) if x < 10 => {
+            secret.replace_range(secret_offset.., "XXX...");
+        }
+        Some(x) if x < 20 => {
+            secret.replace_range((secret_offset + 2).., "XXXX...");
+        }
+        Some(x) if x < 30 => {
+            secret.replace_range((secret_offset + 3).., "XXXXX...");
+        }
+        Some(_x) => {
+            secret.replace_range((secret_offset + 5).., "XXXXXX...");
         }
     }
 }

crates/meilisearch-types/src/task_view.rs

@@ -55,6 +55,9 @@ pub struct TaskView {
 
     #[serde(default, skip_serializing_if = "Option::is_none")]
     pub network: Option<TaskNetwork>,
+
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub custom_metadata: Option<String>,
 }
 
 impl TaskView {
@@ -73,6 +76,7 @@ impl TaskView {
             started_at: task.started_at,
             finished_at: task.finished_at,
             network: task.network.clone(),
+            custom_metadata: task.custom_metadata.clone(),
         }
     }
 }
@@ -142,6 +146,11 @@ 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 +323,24 @@ 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 +442,15 @@ 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()
+                }
+            }
         }
     }
 }

crates/meilisearch-types/src/tasks.rs

@@ -45,6 +45,9 @@ pub struct Task {
 
     #[serde(default, skip_serializing_if = "Option::is_none")]
     pub network: Option<TaskNetwork>,
+
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub custom_metadata: Option<String>,
 }
 
 impl Task {
@@ -67,7 +70,8 @@ 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 +98,8 @@ impl Task {
             | KindWithContent::DumpCreation { .. }
             | KindWithContent::SnapshotCreation
             | KindWithContent::Export { .. }
-            | KindWithContent::UpgradeDatabase { .. } => None,
+            | KindWithContent::UpgradeDatabase { .. }
+            | KindWithContent::IndexCompaction { .. } => None,
         }
     }
 }
@@ -170,6 +175,9 @@ pub enum KindWithContent {
     UpgradeDatabase {
         from: (u32, u32, u32),
     },
+    IndexCompaction {
+        index_uid: String,
+    },
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize, ToSchema)]
@@ -206,6 +214,7 @@ impl KindWithContent {
             KindWithContent::SnapshotCreation => Kind::SnapshotCreation,
             KindWithContent::Export { .. } => Kind::Export,
             KindWithContent::UpgradeDatabase { .. } => Kind::UpgradeDatabase,
+            KindWithContent::IndexCompaction { .. } => Kind::IndexCompaction,
         }
     }
 
@@ -226,7 +235,8 @@ 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 +335,11 @@ 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 +422,11 @@ 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 +489,11 @@ 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 +604,7 @@ pub enum Kind {
     SnapshotCreation,
     Export,
     UpgradeDatabase,
+    IndexCompaction,
 }
 
 impl Kind {
@@ -590,7 +616,8 @@ impl Kind {
             | Kind::SettingsUpdate
             | Kind::IndexCreation
             | Kind::IndexDeletion
-            | Kind::IndexUpdate => true,
+            | Kind::IndexUpdate
+            | Kind::IndexCompaction => true,
             Kind::IndexSwap
             | Kind::TaskCancelation
             | Kind::TaskDeletion
@@ -618,6 +645,7 @@ 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 +681,8 @@ 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 +768,11 @@ 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 +835,10 @@ 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 { .. }

crates/meilisearch-types/src/webhooks.rs

@@ -11,6 +11,24 @@ pub struct Webhook {
     pub headers: BTreeMap<String, String>,
 }
 
+impl Webhook {
+    pub fn redact_authorization_header(&mut self) {
+        // headers are case insensitive, so to make the redaction robust we iterate over qualifying headers
+        // rather than getting one canonical `Authorization` header.
+        for value in self
+            .headers
+            .iter_mut()
+            .filter_map(|(name, value)| name.eq_ignore_ascii_case("authorization").then_some(value))
+        {
+            if value.starts_with("Bearer ") {
+                crate::settings::hide_secret(value, "Bearer ".len());
+            } else {
+                crate::settings::hide_secret(value, 0);
+            }
+        }
+    }
+}
+
 #[derive(Debug, Serialize, Default, Clone, PartialEq)]
 #[serde(rename_all = "camelCase")]
 pub struct WebhooksView {

crates/meilisearch/Cargo.toml

@@ -91,7 +91,7 @@ time = { version = "0.3.41", features = [
 ] }
 tokio = { version = "1.45.1", features = ["full"] }
 toml = "0.8.23"
-uuid = { version = "1.17.0", features = ["serde", "v4"] }
+uuid = { version = "1.18.0", features = ["serde", "v4", "v7"] }
 serde_urlencoded = "0.7.1"
 termcolor = "1.4.1"
 url = { version = "2.5.4", features = ["serde"] }

crates/meilisearch/src/analytics/segment_analytics.rs

@@ -205,7 +205,10 @@ struct Infos {
     experimental_no_snapshot_compaction: bool,
     experimental_no_edition_2024_for_dumps: bool,
     experimental_no_edition_2024_for_settings: bool,
+    experimental_no_edition_2024_for_prefix_post_processing: bool,
+    experimental_no_edition_2024_for_facet_post_processing: bool,
     experimental_vector_store_setting: bool,
+    experimental_personalization: bool,
     gpu_enabled: bool,
     db_path: bool,
     import_dump: bool,
@@ -215,6 +218,7 @@ struct Infos {
     import_snapshot: bool,
     schedule_snapshot: Option<u64>,
     snapshot_dir: bool,
+    uses_s3_snapshots: bool,
     ignore_missing_snapshot: bool,
     ignore_snapshot_if_db_exists: bool,
     http_addr: bool,
@@ -283,6 +287,8 @@ impl Infos {
             indexer_options,
             config_file_path,
             no_analytics: _,
+            experimental_personalization_api_key,
+            s3_snapshot_options,
         } = options;
 
         let schedule_snapshot = match schedule_snapshot {
@@ -296,6 +302,8 @@ impl Infos {
             skip_index_budget: _,
             experimental_no_edition_2024_for_settings,
             experimental_no_edition_2024_for_dumps,
+            experimental_no_edition_2024_for_prefix_post_processing,
+            experimental_no_edition_2024_for_facet_post_processing,
         } = indexer_options;
 
         let RuntimeTogglableFeatures {
@@ -344,6 +352,7 @@ impl Infos {
             import_snapshot: import_snapshot.is_some(),
             schedule_snapshot,
             snapshot_dir: snapshot_dir != PathBuf::from("snapshots/"),
+            uses_s3_snapshots: s3_snapshot_options.is_some(),
             ignore_missing_snapshot,
             ignore_snapshot_if_db_exists,
             http_addr: http_addr != default_http_addr(),
@@ -365,6 +374,9 @@ impl Infos {
             ssl_resumption,
             ssl_tickets,
             experimental_no_edition_2024_for_settings,
+            experimental_no_edition_2024_for_prefix_post_processing,
+            experimental_no_edition_2024_for_facet_post_processing,
+            experimental_personalization: experimental_personalization_api_key.is_some(),
         }
     }
 }

crates/meilisearch/src/error.rs

@@ -12,6 +12,7 @@ use tokio::task::JoinError;
 use crate::routes::indexes::{PROXY_ORIGIN_REMOTE_HEADER, PROXY_ORIGIN_TASK_UID_HEADER};
 
 #[derive(Debug, thiserror::Error)]
+#[allow(clippy::large_enum_variant)]
 pub enum MeilisearchHttpError {
     #[error("A Content-Type header is missing. Accepted values for the Content-Type header are: {}",
             .0.iter().map(|s| format!("`{}`", s)).collect::<Vec<_>>().join(", "))]
@@ -37,6 +38,8 @@ pub enum MeilisearchHttpError {
     PaginationInFederatedQuery(usize, &'static str),
     #[error("Inside `.queries[{0}]`: Using facet options is not allowed in federated queries.\n - Hint: remove `facets` from query #{0} or remove `federation` from the request\n - Hint: pass `federation.facetsByIndex.{1}: {2:?}` for facets in federated search")]
     FacetsInFederatedQuery(usize, String, Vec<String>),
+    #[error("Inside `.queries[{0}]`: Using `.personalize` is not allowed in federated queries.\n - Hint: remove `personalize` from query #{0} or remove `federation` from the request")]
+    PersonalizationInFederatedQuery(usize),
     #[error("Inconsistent order for values in facet `{facet}`: index `{previous_uid}` orders {previous_facet_order}, but index `{current_uid}` orders {index_facet_order}.\n - Hint: Remove `federation.mergeFacets` or change `faceting.sortFacetValuesBy` to be consistent in settings.")]
     InconsistentFacetOrder {
         facet: String,
@@ -136,6 +139,9 @@ impl ErrorCode for MeilisearchHttpError {
             MeilisearchHttpError::InconsistentFacetOrder { .. } => {
                 Code::InvalidMultiSearchFacetOrder
             }
+            MeilisearchHttpError::PersonalizationInFederatedQuery(_) => {
+                Code::InvalidMultiSearchQueryPersonalization
+            }
             MeilisearchHttpError::InconsistentOriginHeaders { .. } => {
                 Code::InconsistentDocumentChangeHeaders
             }

crates/meilisearch/src/lib.rs

@@ -1,4 +1,6 @@
+#![allow(clippy::result_large_err)]
 #![allow(rustdoc::private_intra_doc_links)]
+
 #[macro_use]
 pub mod error;
 pub mod analytics;
@@ -9,6 +11,7 @@ pub mod middleware;
 pub mod option;
 #[cfg(test)]
 mod option_test;
+pub mod personalization;
 pub mod routes;
 pub mod search;
 pub mod search_queue;
@@ -56,6 +59,7 @@ use tracing::{error, info_span};
 use tracing_subscriber::filter::Targets;
 
 use crate::error::MeilisearchHttpError;
+use crate::personalization::PersonalizationService;
 
 /// Default number of simultaneously opened indexes.
 ///
@@ -126,12 +130,8 @@ pub type LogStderrType = tracing_subscriber::filter::Filtered<
 >;
 
 pub fn create_app(
-    index_scheduler: Data<IndexScheduler>,
-    auth_controller: Data<AuthController>,
-    search_queue: Data<SearchQueue>,
+    services: ServicesData,
     opt: Opt,
-    logs: (LogRouteHandle, LogStderrHandle),
-    analytics: Data<Analytics>,
     enable_dashboard: bool,
 ) -> actix_web::App<
     impl ServiceFactory<
@@ -143,17 +143,7 @@ pub fn create_app(
     >,
 > {
     let app = actix_web::App::new()
-        .configure(|s| {
-            configure_data(
-                s,
-                index_scheduler.clone(),
-                auth_controller.clone(),
-                search_queue.clone(),
-                &opt,
-                logs,
-                analytics.clone(),
-            )
-        })
+        .configure(|s| configure_data(s, services, &opt))
         .configure(routes::configure)
         .configure(|s| dashboard(s, enable_dashboard));
 
@@ -214,7 +204,10 @@ enum OnFailure {
     KeepDb,
 }
 
-pub fn setup_meilisearch(opt: &Opt) -> anyhow::Result<(Arc<IndexScheduler>, Arc<AuthController>)> {
+pub fn setup_meilisearch(
+    opt: &Opt,
+    handle: tokio::runtime::Handle,
+) -> anyhow::Result<(Arc<IndexScheduler>, Arc<AuthController>)> {
     let index_scheduler_opt = IndexSchedulerOptions {
         version_file_path: opt.db_path.join(VERSION_FILE_NAME),
         auth_path: opt.db_path.join("auth"),
@@ -228,7 +221,11 @@ pub fn setup_meilisearch(opt: &Opt) -> anyhow::Result<(Arc<IndexScheduler>, Arc<
         task_db_size: opt.max_task_db_size.as_u64() as usize,
         index_base_map_size: opt.max_index_size.as_u64() as usize,
         enable_mdb_writemap: opt.experimental_reduce_indexing_memory_usage,
-        indexer_config: Arc::new((&opt.indexer_options).try_into()?),
+        indexer_config: Arc::new({
+            let s3_snapshot_options =
+                opt.s3_snapshot_options.clone().map(|opt| opt.try_into()).transpose()?;
+            IndexerConfig { s3_snapshot_options, ..(&opt.indexer_options).try_into()? }
+        }),
         autobatching_enabled: true,
         cleanup_enabled: !opt.experimental_replication_parameters,
         max_number_of_tasks: 1_000_000,
@@ -254,6 +251,7 @@ pub fn setup_meilisearch(opt: &Opt) -> anyhow::Result<(Arc<IndexScheduler>, Arc<
                     index_scheduler_opt,
                     OnFailure::RemoveDb,
                     binary_version, // the db is empty
+                    handle,
                 )?,
                 Err(e) => {
                     std::fs::remove_dir_all(&opt.db_path)?;
@@ -271,7 +269,7 @@ pub fn setup_meilisearch(opt: &Opt) -> anyhow::Result<(Arc<IndexScheduler>, Arc<
             bail!("snapshot doesn't exist at {}", snapshot_path.display())
         // the snapshot and the db exist, and we can ignore the snapshot because of the ignore_snapshot_if_db_exists flag
         } else {
-            open_or_create_database(opt, index_scheduler_opt, empty_db, binary_version)?
+            open_or_create_database(opt, index_scheduler_opt, empty_db, binary_version, handle)?
         }
     } else if let Some(ref path) = opt.import_dump {
         let src_path_exists = path.exists();
@@ -282,6 +280,7 @@ pub fn setup_meilisearch(opt: &Opt) -> anyhow::Result<(Arc<IndexScheduler>, Arc<
                 index_scheduler_opt,
                 OnFailure::RemoveDb,
                 binary_version, // the db is empty
+                handle,
             )?;
             match import_dump(&opt.db_path, path, &mut index_scheduler, &mut auth_controller) {
                 Ok(()) => (index_scheduler, auth_controller),
@@ -302,10 +301,10 @@ pub fn setup_meilisearch(opt: &Opt) -> anyhow::Result<(Arc<IndexScheduler>, Arc<
         // the dump and the db exist and we can ignore the dump because of the ignore_dump_if_db_exists flag
         // or, the dump is missing but we can ignore that because of the ignore_missing_dump flag
         } else {
-            open_or_create_database(opt, index_scheduler_opt, empty_db, binary_version)?
+            open_or_create_database(opt, index_scheduler_opt, empty_db, binary_version, handle)?
         }
     } else {
-        open_or_create_database(opt, index_scheduler_opt, empty_db, binary_version)?
+        open_or_create_database(opt, index_scheduler_opt, empty_db, binary_version, handle)?
     };
 
     // We create a loop in a thread that registers snapshotCreation tasks
@@ -336,6 +335,7 @@ fn open_or_create_database_unchecked(
     index_scheduler_opt: IndexSchedulerOptions,
     on_failure: OnFailure,
     version: (u32, u32, u32),
+    handle: tokio::runtime::Handle,
 ) -> anyhow::Result<(IndexScheduler, AuthController)> {
     // we don't want to create anything in the data.ms yet, thus we
     // wrap our two builders in a closure that'll be executed later.
@@ -343,7 +343,7 @@ fn open_or_create_database_unchecked(
     let auth_env = open_auth_store_env(&index_scheduler_opt.auth_path).unwrap();
     let auth_controller = AuthController::new(auth_env.clone(), &opt.master_key);
     let index_scheduler_builder = || -> anyhow::Result<_> {
-        Ok(IndexScheduler::new(index_scheduler_opt, auth_env, version)?)
+        Ok(IndexScheduler::new(index_scheduler_opt, auth_env, version, Some(handle))?)
     };
 
     match (
@@ -450,6 +450,7 @@ fn open_or_create_database(
     index_scheduler_opt: IndexSchedulerOptions,
     empty_db: bool,
     binary_version: (u32, u32, u32),
+    handle: tokio::runtime::Handle,
 ) -> anyhow::Result<(IndexScheduler, AuthController)> {
     let version = if !empty_db {
         check_version(opt, &index_scheduler_opt, binary_version)?
@@ -457,7 +458,7 @@ fn open_or_create_database(
         binary_version
     };
 
-    open_or_create_database_unchecked(opt, index_scheduler_opt, OnFailure::KeepDb, version)
+    open_or_create_database_unchecked(opt, index_scheduler_opt, OnFailure::KeepDb, version, handle)
 }
 
 fn import_dump(
@@ -525,7 +526,11 @@ fn import_dump(
     let indexer_config = if base_config.max_threads.is_none() {
         let (thread_pool, _) = default_thread_pool_and_threads();
 
-        let _config = IndexerConfig { thread_pool, ..*base_config };
+        let _config = IndexerConfig {
+            thread_pool,
+            s3_snapshot_options: base_config.s3_snapshot_options.clone(),
+            ..*base_config
+        };
         backup_config = _config;
         &backup_config
     } else {
@@ -673,23 +678,26 @@ fn import_dump(
     Ok(index_scheduler_dump.finish()?)
 }
 
-pub fn configure_data(
-    config: &mut web::ServiceConfig,
-    index_scheduler: Data<IndexScheduler>,
-    auth: Data<AuthController>,
-    search_queue: Data<SearchQueue>,
-    opt: &Opt,
-    (logs_route, logs_stderr): (LogRouteHandle, LogStderrHandle),
-    analytics: Data<Analytics>,
-) {
+pub fn configure_data(config: &mut web::ServiceConfig, services: ServicesData, opt: &Opt) {
+    let ServicesData {
+        index_scheduler,
+        auth,
+        search_queue,
+        personalization_service,
+        logs_route_handle,
+        logs_stderr_handle,
+        analytics,
+    } = services;
+
     let http_payload_size_limit = opt.http_payload_size_limit.as_u64() as usize;
     config
         .app_data(index_scheduler)
         .app_data(auth)
         .app_data(search_queue)
         .app_data(analytics)
-        .app_data(web::Data::new(logs_route))
-        .app_data(web::Data::new(logs_stderr))
+        .app_data(personalization_service)
+        .app_data(logs_route_handle)
+        .app_data(logs_stderr_handle)
         .app_data(web::Data::new(opt.clone()))
         .app_data(
             web::JsonConfig::default()
@@ -750,3 +758,14 @@ pub fn dashboard(config: &mut web::ServiceConfig, enable_frontend: bool) {
 pub fn dashboard(config: &mut web::ServiceConfig, _enable_frontend: bool) {
     config.service(web::resource("/").route(web::get().to(routes::running)));
 }
+
+#[derive(Clone)]
+pub struct ServicesData {
+    pub index_scheduler: Data<IndexScheduler>,
+    pub auth: Data<AuthController>,
+    pub search_queue: Data<SearchQueue>,
+    pub personalization_service: Data<PersonalizationService>,
+    pub logs_route_handle: Data<LogRouteHandle>,
+    pub logs_stderr_handle: Data<LogStderrHandle>,
+    pub analytics: Data<Analytics>,
+}

crates/meilisearch/src/main.rs

@@ -14,10 +14,11 @@ use index_scheduler::IndexScheduler;
 use is_terminal::IsTerminal;
 use meilisearch::analytics::Analytics;
 use meilisearch::option::LogMode;
+use meilisearch::personalization::PersonalizationService;
 use meilisearch::search_queue::SearchQueue;
 use meilisearch::{
     analytics, create_app, setup_meilisearch, LogRouteHandle, LogRouteType, LogStderrHandle,
-    LogStderrType, Opt, SubscriberForSecondLayer,
+    LogStderrType, Opt, ServicesData, SubscriberForSecondLayer,
 };
 use meilisearch_auth::{generate_master_key, AuthController, MASTER_KEY_MIN_SIZE};
 use termcolor::{Color, ColorChoice, ColorSpec, StandardStream, WriteColor};
@@ -76,7 +77,10 @@ fn on_panic(info: &std::panic::PanicHookInfo) {
 
 #[actix_web::main]
 async fn main() -> anyhow::Result<()> {
-    try_main().await.inspect_err(|error| {
+    // won't panic inside of tokio::main
+    let runtime = tokio::runtime::Handle::current();
+
+    try_main(runtime).await.inspect_err(|error| {
         tracing::error!(%error);
         let mut current = error.source();
         let mut depth = 0;
@@ -88,7 +92,7 @@ async fn main() -> anyhow::Result<()> {
     })
 }
 
-async fn try_main() -> anyhow::Result<()> {
+async fn try_main(runtime: tokio::runtime::Handle) -> anyhow::Result<()> {
     let (opt, config_read_from) = Opt::try_build()?;
 
     std::panic::set_hook(Box::new(on_panic));
@@ -122,7 +126,7 @@ async fn try_main() -> anyhow::Result<()> {
         _ => (),
     }
 
-    let (index_scheduler, auth_controller) = setup_meilisearch(&opt)?;
+    let (index_scheduler, auth_controller) = setup_meilisearch(&opt, runtime)?;
 
     let analytics =
         analytics::Analytics::new(&opt, index_scheduler.clone(), auth_controller.clone()).await;
@@ -149,8 +153,15 @@ async fn run_http(
     let enable_dashboard = &opt.env == "development";
     let opt_clone = opt.clone();
     let index_scheduler = Data::from(index_scheduler);
-    let auth_controller = Data::from(auth_controller);
+    let auth = Data::from(auth_controller);
     let analytics = Data::from(analytics);
+    // Create personalization service with API key from options
+    let personalization_service = Data::new(
+        opt.experimental_personalization_api_key
+            .clone()
+            .map(PersonalizationService::cohere)
+            .unwrap_or_else(PersonalizationService::disabled),
+    );
     let search_queue = SearchQueue::new(
         opt.experimental_search_queue_size,
         available_parallelism()
@@ -162,21 +173,25 @@ async fn run_http(
         usize::from(opt.experimental_drop_search_after) as u64
     ));
     let search_queue = Data::new(search_queue);
+    let (logs_route_handle, logs_stderr_handle) = logs;
+    let logs_route_handle = Data::new(logs_route_handle);
+    let logs_stderr_handle = Data::new(logs_stderr_handle);
 
-    let http_server = HttpServer::new(move || {
-        create_app(
-            index_scheduler.clone(),
-            auth_controller.clone(),
-            search_queue.clone(),
-            opt.clone(),
-            logs.clone(),
-            analytics.clone(),
-            enable_dashboard,
-        )
-    })
-    // Disable signals allows the server to terminate immediately when a user enter CTRL-C
-    .disable_signals()
-    .keep_alive(KeepAlive::Os);
+    let services = ServicesData {
+        index_scheduler,
+        auth,
+        search_queue,
+        personalization_service,
+        logs_route_handle,
+        logs_stderr_handle,
+        analytics,
+    };
+
+    let http_server =
+        HttpServer::new(move || create_app(services.clone(), opt.clone(), enable_dashboard))
+            // Disable signals allows the server to terminate immediately when a user enter CTRL-C
+            .disable_signals()
+            .keep_alive(KeepAlive::Os);
 
     if let Some(config) = opt_clone.get_ssl_config()? {
         http_server.bind_rustls_0_23(opt_clone.http_addr, config)?.run().await?;

crates/meilisearch/src/metrics.rs

@@ -114,4 +114,9 @@ lazy_static! {
             "Meilisearch Task Queue Size Until Stop Registering",
         ))
         .expect("Can't create a metric");
+    pub static ref MEILISEARCH_PERSONALIZED_SEARCH_REQUESTS: IntGauge = register_int_gauge!(opts!(
+        "meilisearch_personalized_search_requests",
+        "Meilisearch number of search requests with personalization"
+    ))
+    .expect("Can't create a metric");
 }

crates/meilisearch/src/option.rs

@@ -7,12 +7,13 @@ use std::ops::Deref;
 use std::path::PathBuf;
 use std::str::FromStr;
 use std::sync::Arc;
+use std::time::Duration;
 use std::{env, fmt, fs};
 
 use byte_unit::{Byte, ParseError, UnitType};
 use clap::Parser;
 use meilisearch_types::features::InstanceTogglableFeatures;
-use meilisearch_types::milli::update::IndexerConfig;
+use meilisearch_types::milli::update::{IndexerConfig, S3SnapshotOptions};
 use meilisearch_types::milli::ThreadPoolNoAbortBuilder;
 use rustls::server::{ServerSessionMemoryCache, WebPkiClientVerifier};
 use rustls::RootCertStore;
@@ -55,6 +56,10 @@ const MEILI_EXPERIMENTAL_ENABLE_LOGS_ROUTE: &str = "MEILI_EXPERIMENTAL_ENABLE_LO
 const MEILI_EXPERIMENTAL_CONTAINS_FILTER: &str = "MEILI_EXPERIMENTAL_CONTAINS_FILTER";
 const MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_SETTINGS: &str =
     "MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_SETTINGS";
+const MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_FACET_POST_PROCESSING: &str =
+    "MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_FACET_POST_PROCESSING";
+const MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_PREFIX_POST_PROCESSING: &str =
+    "MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_PREFIX_POST_PROCESSING";
 const MEILI_EXPERIMENTAL_ENABLE_METRICS: &str = "MEILI_EXPERIMENTAL_ENABLE_METRICS";
 const MEILI_EXPERIMENTAL_SEARCH_QUEUE_SIZE: &str = "MEILI_EXPERIMENTAL_SEARCH_QUEUE_SIZE";
 const MEILI_EXPERIMENTAL_DROP_SEARCH_AFTER: &str = "MEILI_EXPERIMENTAL_DROP_SEARCH_AFTER";
@@ -70,6 +75,22 @@ const MEILI_EXPERIMENTAL_EMBEDDING_CACHE_ENTRIES: &str =
 const MEILI_EXPERIMENTAL_NO_SNAPSHOT_COMPACTION: &str = "MEILI_EXPERIMENTAL_NO_SNAPSHOT_COMPACTION";
 const MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_DUMPS: &str =
     "MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_DUMPS";
+const MEILI_EXPERIMENTAL_PERSONALIZATION_API_KEY: &str =
+    "MEILI_EXPERIMENTAL_PERSONALIZATION_API_KEY";
+
+// Related to S3 snapshots
+const MEILI_S3_BUCKET_URL: &str = "MEILI_S3_BUCKET_URL";
+const MEILI_S3_BUCKET_REGION: &str = "MEILI_S3_BUCKET_REGION";
+const MEILI_S3_BUCKET_NAME: &str = "MEILI_S3_BUCKET_NAME";
+const MEILI_S3_SNAPSHOT_PREFIX: &str = "MEILI_S3_SNAPSHOT_PREFIX";
+const MEILI_S3_ACCESS_KEY: &str = "MEILI_S3_ACCESS_KEY";
+const MEILI_S3_SECRET_KEY: &str = "MEILI_S3_SECRET_KEY";
+const MEILI_EXPERIMENTAL_S3_MAX_IN_FLIGHT_PARTS: &str = "MEILI_EXPERIMENTAL_S3_MAX_IN_FLIGHT_PARTS";
+const MEILI_EXPERIMENTAL_S3_COMPRESSION_LEVEL: &str = "MEILI_EXPERIMENTAL_S3_COMPRESSION_LEVEL";
+const MEILI_EXPERIMENTAL_S3_SIGNATURE_DURATION_SECONDS: &str =
+    "MEILI_EXPERIMENTAL_S3_SIGNATURE_DURATION_SECONDS";
+const MEILI_EXPERIMENTAL_S3_MULTIPART_PART_SIZE: &str = "MEILI_EXPERIMENTAL_S3_MULTIPART_PART_SIZE";
+
 const DEFAULT_CONFIG_FILE_PATH: &str = "./config.toml";
 const DEFAULT_DB_PATH: &str = "./data.ms";
 const DEFAULT_HTTP_ADDR: &str = "localhost:7700";
@@ -79,6 +100,10 @@ const DEFAULT_SNAPSHOT_DIR: &str = "snapshots/";
 const DEFAULT_SNAPSHOT_INTERVAL_SEC: u64 = 86400;
 const DEFAULT_SNAPSHOT_INTERVAL_SEC_STR: &str = "86400";
 const DEFAULT_DUMP_DIR: &str = "dumps/";
+const DEFAULT_S3_SNAPSHOT_MAX_IN_FLIGHT_PARTS: NonZeroUsize = NonZeroUsize::new(10).unwrap();
+const DEFAULT_S3_SNAPSHOT_COMPRESSION_LEVEL: u32 = 0;
+const DEFAULT_S3_SNAPSHOT_SIGNATURE_DURATION_SECONDS: u64 = 8 * 3600; // 8 hours
+const DEFAULT_S3_SNAPSHOT_MULTIPART_PART_SIZE: Byte = Byte::from_u64(375 * 1024 * 1024); // 375 MiB
 
 const MEILI_MAX_INDEXING_MEMORY: &str = "MEILI_MAX_INDEXING_MEMORY";
 const MEILI_MAX_INDEXING_THREADS: &str = "MEILI_MAX_INDEXING_THREADS";
@@ -471,10 +496,20 @@ pub struct Opt {
     #[serde(default)]
     pub experimental_no_snapshot_compaction: bool,
 
+    /// Experimental personalization API key feature.
+    ///
+    /// Sets the API key for personalization features.
+    #[clap(long, env = MEILI_EXPERIMENTAL_PERSONALIZATION_API_KEY)]
+    pub experimental_personalization_api_key: Option<String>,
+
     #[serde(flatten)]
     #[clap(flatten)]
     pub indexer_options: IndexerOpts,
 
+    #[serde(flatten)]
+    #[clap(flatten)]
+    pub s3_snapshot_options: Option<S3SnapshotOpts>,
+
     /// Set the path to a configuration file that should be used to setup the engine.
     /// Format must be TOML.
     #[clap(long)]
@@ -576,6 +611,8 @@ impl Opt {
             experimental_limit_batched_tasks_total_size,
             experimental_embedding_cache_entries,
             experimental_no_snapshot_compaction,
+            experimental_personalization_api_key,
+            s3_snapshot_options,
         } = self;
         export_to_env_if_not_present(MEILI_DB_PATH, db_path);
         export_to_env_if_not_present(MEILI_HTTP_ADDR, http_addr);
@@ -676,7 +713,22 @@ impl Opt {
             MEILI_EXPERIMENTAL_NO_SNAPSHOT_COMPACTION,
             experimental_no_snapshot_compaction.to_string(),
         );
+        if let Some(experimental_personalization_api_key) = experimental_personalization_api_key {
+            export_to_env_if_not_present(
+                MEILI_EXPERIMENTAL_PERSONALIZATION_API_KEY,
+                experimental_personalization_api_key,
+            );
+        }
         indexer_options.export_to_env();
+        if let Some(s3_snapshot_options) = s3_snapshot_options {
+            #[cfg(not(unix))]
+            {
+                let _ = s3_snapshot_options;
+                panic!("S3 snapshot options are not supported on Windows");
+            }
+            #[cfg(unix)]
+            s3_snapshot_options.export_to_env();
+        }
     }
 
     pub fn get_ssl_config(&self) -> anyhow::Result<Option<rustls::ServerConfig>> {
@@ -772,6 +824,22 @@ pub struct IndexerOpts {
     #[clap(long, env = MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_DUMPS)]
     #[serde(default)]
     pub experimental_no_edition_2024_for_dumps: bool,
+
+    /// Experimental no edition 2024 to compute prefixes. For more information,
+    /// see: <https://github.com/orgs/meilisearch/discussions/862>
+    ///
+    /// Enables the experimental no edition 2024 to compute prefixes.
+    #[clap(long, env = MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_PREFIX_POST_PROCESSING)]
+    #[serde(default)]
+    pub experimental_no_edition_2024_for_prefix_post_processing: bool,
+
+    /// Experimental no edition 2024 to compute facets. For more information,
+    /// see: <https://github.com/orgs/meilisearch/discussions/862>
+    ///
+    /// Enables the experimental no edition 2024 to compute facets.
+    #[clap(long, env = MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_FACET_POST_PROCESSING)]
+    #[serde(default)]
+    pub experimental_no_edition_2024_for_facet_post_processing: bool,
 }
 
 impl IndexerOpts {
@@ -783,6 +851,8 @@ impl IndexerOpts {
             skip_index_budget: _,
             experimental_no_edition_2024_for_settings,
             experimental_no_edition_2024_for_dumps,
+            experimental_no_edition_2024_for_prefix_post_processing,
+            experimental_no_edition_2024_for_facet_post_processing,
         } = self;
         if let Some(max_indexing_memory) = max_indexing_memory.0 {
             export_to_env_if_not_present(
@@ -808,6 +878,18 @@ impl IndexerOpts {
                 experimental_no_edition_2024_for_dumps.to_string(),
             );
         }
+        if experimental_no_edition_2024_for_prefix_post_processing {
+            export_to_env_if_not_present(
+                MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_PREFIX_POST_PROCESSING,
+                experimental_no_edition_2024_for_prefix_post_processing.to_string(),
+            );
+        }
+        if experimental_no_edition_2024_for_facet_post_processing {
+            export_to_env_if_not_present(
+                MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_FACET_POST_PROCESSING,
+                experimental_no_edition_2024_for_facet_post_processing.to_string(),
+            );
+        }
     }
 }
 
@@ -815,6 +897,16 @@ impl TryFrom<&IndexerOpts> for IndexerConfig {
     type Error = anyhow::Error;
 
     fn try_from(other: &IndexerOpts) -> Result<Self, Self::Error> {
+        let IndexerOpts {
+            max_indexing_memory,
+            max_indexing_threads,
+            skip_index_budget,
+            experimental_no_edition_2024_for_settings,
+            experimental_no_edition_2024_for_dumps,
+            experimental_no_edition_2024_for_prefix_post_processing,
+            experimental_no_edition_2024_for_facet_post_processing,
+        } = other;
+
         let thread_pool = ThreadPoolNoAbortBuilder::new_for_indexing()
             .num_threads(other.max_indexing_threads.unwrap_or_else(|| num_cpus::get() / 2))
             .build()?;
@@ -822,17 +914,163 @@ impl TryFrom<&IndexerOpts> for IndexerConfig {
         Ok(Self {
             thread_pool,
             log_every_n: Some(DEFAULT_LOG_EVERY_N),
-            max_memory: other.max_indexing_memory.map(|b| b.as_u64() as usize),
-            max_threads: *other.max_indexing_threads,
+            max_memory: max_indexing_memory.map(|b| b.as_u64() as usize),
+            max_threads: max_indexing_threads.0,
             max_positions_per_attributes: None,
-            skip_index_budget: other.skip_index_budget,
-            experimental_no_edition_2024_for_settings: other
-                .experimental_no_edition_2024_for_settings,
-            experimental_no_edition_2024_for_dumps: other.experimental_no_edition_2024_for_dumps,
+            skip_index_budget: *skip_index_budget,
+            experimental_no_edition_2024_for_settings: *experimental_no_edition_2024_for_settings,
+            experimental_no_edition_2024_for_dumps: *experimental_no_edition_2024_for_dumps,
             chunk_compression_type: Default::default(),
             chunk_compression_level: Default::default(),
             documents_chunk_size: Default::default(),
             max_nb_chunks: Default::default(),
+            experimental_no_edition_2024_for_prefix_post_processing:
+                *experimental_no_edition_2024_for_prefix_post_processing,
+            experimental_no_edition_2024_for_facet_post_processing:
+                *experimental_no_edition_2024_for_facet_post_processing,
+            s3_snapshot_options: None,
+        })
+    }
+}
+
+#[derive(Debug, Clone, Parser, Deserialize)]
+// This group is a bit tricky but makes it possible to require all listed fields if one of them
+// is specified. It lets us keep an Option for the S3SnapshotOpts configuration.
+// <https://github.com/clap-rs/clap/issues/5092#issuecomment-2616986075>
+#[group(requires_all = ["s3_bucket_url", "s3_bucket_region", "s3_bucket_name", "s3_snapshot_prefix", "s3_access_key", "s3_secret_key"])]
+pub struct S3SnapshotOpts {
+    /// The S3 bucket URL in the format https://s3.<region>.amazonaws.com.
+    #[clap(long, env = MEILI_S3_BUCKET_URL, required = false)]
+    #[serde(default)]
+    pub s3_bucket_url: String,
+
+    /// The region in the format us-east-1.
+    #[clap(long, env = MEILI_S3_BUCKET_REGION, required = false)]
+    #[serde(default)]
+    pub s3_bucket_region: String,
+
+    /// The bucket name.
+    #[clap(long, env = MEILI_S3_BUCKET_NAME, required = false)]
+    #[serde(default)]
+    pub s3_bucket_name: String,
+
+    /// The prefix path where to put the snapshot, uses normal slashes (/).
+    #[clap(long, env = MEILI_S3_SNAPSHOT_PREFIX, required = false)]
+    #[serde(default)]
+    pub s3_snapshot_prefix: String,
+
+    /// The S3 access key.
+    #[clap(long, env = MEILI_S3_ACCESS_KEY, required = false)]
+    #[serde(default)]
+    pub s3_access_key: String,
+
+    /// The S3 secret key.
+    #[clap(long, env = MEILI_S3_SECRET_KEY, required = false)]
+    #[serde(default)]
+    pub s3_secret_key: String,
+
+    /// The maximum number of parts that can be uploaded in parallel.
+    ///
+    /// For more information, see <https://github.com/orgs/meilisearch/discussions/869>.
+    #[clap(long, env = MEILI_EXPERIMENTAL_S3_MAX_IN_FLIGHT_PARTS, default_value_t = default_experimental_s3_snapshot_max_in_flight_parts())]
+    #[serde(default = "default_experimental_s3_snapshot_max_in_flight_parts")]
+    pub experimental_s3_max_in_flight_parts: NonZeroUsize,
+
+    /// The compression level. Defaults to no compression (0).
+    ///
+    /// For more information, see <https://github.com/orgs/meilisearch/discussions/869>.
+    #[clap(long, env = MEILI_EXPERIMENTAL_S3_COMPRESSION_LEVEL, default_value_t = default_experimental_s3_snapshot_compression_level())]
+    #[serde(default = "default_experimental_s3_snapshot_compression_level")]
+    pub experimental_s3_compression_level: u32,
+
+    /// The signature duration for the multipart upload.
+    ///
+    /// For more information, see <https://github.com/orgs/meilisearch/discussions/869>.
+    #[clap(long, env = MEILI_EXPERIMENTAL_S3_SIGNATURE_DURATION_SECONDS, default_value_t = default_experimental_s3_snapshot_signature_duration_seconds())]
+    #[serde(default = "default_experimental_s3_snapshot_signature_duration_seconds")]
+    pub experimental_s3_signature_duration_seconds: u64,
+
+    /// The size of the the multipart parts.
+    ///
+    /// Must not be less than 10MiB and larger than 8GiB. Yes,
+    /// twice the boundaries of the AWS S3 multipart upload
+    /// because we use it a bit differently internally.
+    ///
+    /// For more information, see <https://github.com/orgs/meilisearch/discussions/869>.
+    #[clap(long, env = MEILI_EXPERIMENTAL_S3_MULTIPART_PART_SIZE, default_value_t = default_experimental_s3_snapshot_multipart_part_size())]
+    #[serde(default = "default_experimental_s3_snapshot_multipart_part_size")]
+    pub experimental_s3_multipart_part_size: Byte,
+}
+
+impl S3SnapshotOpts {
+    /// Exports the values to their corresponding env vars if they are not set.
+    pub fn export_to_env(self) {
+        let S3SnapshotOpts {
+            s3_bucket_url,
+            s3_bucket_region,
+            s3_bucket_name,
+            s3_snapshot_prefix,
+            s3_access_key,
+            s3_secret_key,
+            experimental_s3_max_in_flight_parts,
+            experimental_s3_compression_level,
+            experimental_s3_signature_duration_seconds,
+            experimental_s3_multipart_part_size,
+        } = self;
+
+        export_to_env_if_not_present(MEILI_S3_BUCKET_URL, s3_bucket_url);
+        export_to_env_if_not_present(MEILI_S3_BUCKET_REGION, s3_bucket_region);
+        export_to_env_if_not_present(MEILI_S3_BUCKET_NAME, s3_bucket_name);
+        export_to_env_if_not_present(MEILI_S3_SNAPSHOT_PREFIX, s3_snapshot_prefix);
+        export_to_env_if_not_present(MEILI_S3_ACCESS_KEY, s3_access_key);
+        export_to_env_if_not_present(MEILI_S3_SECRET_KEY, s3_secret_key);
+        export_to_env_if_not_present(
+            MEILI_EXPERIMENTAL_S3_MAX_IN_FLIGHT_PARTS,
+            experimental_s3_max_in_flight_parts.to_string(),
+        );
+        export_to_env_if_not_present(
+            MEILI_EXPERIMENTAL_S3_COMPRESSION_LEVEL,
+            experimental_s3_compression_level.to_string(),
+        );
+        export_to_env_if_not_present(
+            MEILI_EXPERIMENTAL_S3_SIGNATURE_DURATION_SECONDS,
+            experimental_s3_signature_duration_seconds.to_string(),
+        );
+        export_to_env_if_not_present(
+            MEILI_EXPERIMENTAL_S3_MULTIPART_PART_SIZE,
+            experimental_s3_multipart_part_size.to_string(),
+        );
+    }
+}
+
+impl TryFrom<S3SnapshotOpts> for S3SnapshotOptions {
+    type Error = anyhow::Error;
+
+    fn try_from(other: S3SnapshotOpts) -> Result<Self, Self::Error> {
+        let S3SnapshotOpts {
+            s3_bucket_url,
+            s3_bucket_region,
+            s3_bucket_name,
+            s3_snapshot_prefix,
+            s3_access_key,
+            s3_secret_key,
+            experimental_s3_max_in_flight_parts,
+            experimental_s3_compression_level,
+            experimental_s3_signature_duration_seconds,
+            experimental_s3_multipart_part_size,
+        } = other;
+
+        Ok(S3SnapshotOptions {
+            s3_bucket_url,
+            s3_bucket_region,
+            s3_bucket_name,
+            s3_snapshot_prefix,
+            s3_access_key,
+            s3_secret_key,
+            s3_max_in_flight_parts: experimental_s3_max_in_flight_parts,
+            s3_compression_level: experimental_s3_compression_level,
+            s3_signature_duration: Duration::from_secs(experimental_s3_signature_duration_seconds),
+            s3_multipart_part_size: experimental_s3_multipart_part_size.as_u64(),
         })
     }
 }
@@ -1051,6 +1289,22 @@ fn default_snapshot_interval_sec() -> &'static str {
     DEFAULT_SNAPSHOT_INTERVAL_SEC_STR
 }
 
+fn default_experimental_s3_snapshot_max_in_flight_parts() -> NonZeroUsize {
+    DEFAULT_S3_SNAPSHOT_MAX_IN_FLIGHT_PARTS
+}
+
+fn default_experimental_s3_snapshot_compression_level() -> u32 {
+    DEFAULT_S3_SNAPSHOT_COMPRESSION_LEVEL
+}
+
+fn default_experimental_s3_snapshot_signature_duration_seconds() -> u64 {
+    DEFAULT_S3_SNAPSHOT_SIGNATURE_DURATION_SECONDS
+}
+
+fn default_experimental_s3_snapshot_multipart_part_size() -> Byte {
+    DEFAULT_S3_SNAPSHOT_MULTIPART_PART_SIZE
+}
+
 fn default_dump_dir() -> PathBuf {
     PathBuf::from(DEFAULT_DUMP_DIR)
 }

crates/meilisearch/src/personalization/mod.rs

@@ -0,0 +1,366 @@
+use crate::search::{Personalize, SearchResult};
+use meilisearch_types::{
+    error::{Code, ErrorCode, ResponseError},
+    milli::TimeBudget,
+};
+use rand::Rng;
+use reqwest::Client;
+use serde::{Deserialize, Serialize};
+use std::time::Duration;
+use tracing::{debug, info, warn};
+
+const COHERE_API_URL: &str = "https://api.cohere.ai/v1/rerank";
+const MAX_RETRIES: u32 = 10;
+
+#[derive(Debug, thiserror::Error)]
+enum PersonalizationError {
+    #[error("Personalization service: HTTP request failed: {0}")]
+    Request(#[from] reqwest::Error),
+    #[error("Personalization service: Failed to parse response: {0}")]
+    Parse(String),
+    #[error("Personalization service: Cohere API error: {0}")]
+    Api(String),
+    #[error("Personalization service: Unauthorized: invalid API key")]
+    Unauthorized,
+    #[error("Personalization service: Rate limited: too many requests")]
+    RateLimited,
+    #[error("Personalization service: Bad request: {0}")]
+    BadRequest(String),
+    #[error("Personalization service: Internal server error: {0}")]
+    InternalServerError(String),
+    #[error("Personalization service: Network error: {0}")]
+    Network(String),
+    #[error("Personalization service: Deadline exceeded")]
+    DeadlineExceeded,
+    #[error(transparent)]
+    FeatureNotEnabled(#[from] index_scheduler::error::FeatureNotEnabledError),
+}
+
+impl ErrorCode for PersonalizationError {
+    fn error_code(&self) -> Code {
+        match self {
+            PersonalizationError::FeatureNotEnabled { .. } => Code::FeatureNotEnabled,
+            PersonalizationError::Unauthorized => Code::RemoteInvalidApiKey,
+            PersonalizationError::RateLimited => Code::TooManySearchRequests,
+            PersonalizationError::BadRequest(_) => Code::RemoteBadRequest,
+            PersonalizationError::InternalServerError(_) => Code::RemoteRemoteError,
+            PersonalizationError::Network(_) | PersonalizationError::Request(_) => {
+                Code::RemoteCouldNotSendRequest
+            }
+            PersonalizationError::Parse(_) | PersonalizationError::Api(_) => {
+                Code::RemoteBadResponse
+            }
+            PersonalizationError::DeadlineExceeded => Code::Internal, // should not be returned to the client
+        }
+    }
+}
+
+pub struct CohereService {
+    client: Client,
+    api_key: String,
+}
+
+impl CohereService {
+    pub fn new(api_key: String) -> Self {
+        info!("Personalization service initialized with Cohere API");
+        let client = Client::builder()
+            .timeout(Duration::from_secs(30))
+            .build()
+            .expect("Failed to create HTTP client");
+        Self { client, api_key }
+    }
+
+    pub async fn rerank_search_results(
+        &self,
+        search_result: SearchResult,
+        personalize: &Personalize,
+        query: Option<&str>,
+        time_budget: TimeBudget,
+    ) -> Result<SearchResult, ResponseError> {
+        if time_budget.exceeded() {
+            warn!("Could not rerank due to deadline");
+            // If the deadline is exceeded, return the original search result instead of an error
+            return Ok(search_result);
+        }
+
+        // Extract user context from personalization
+        let user_context = personalize.user_context.as_str();
+
+        // Build the prompt by merging query and user context
+        let prompt = match query {
+            Some(q) => format!("User Context: {user_context}\nQuery: {q}"),
+            None => format!("User Context: {user_context}"),
+        };
+
+        // Extract documents for reranking
+        let documents: Vec<String> = search_result
+            .hits
+            .iter()
+            .map(|hit| {
+                // Convert the document to a string representation for reranking
+                serde_json::to_string(&hit.document).unwrap_or_else(|_| "{}".to_string())
+            })
+            .collect();
+
+        if documents.is_empty() {
+            return Ok(search_result);
+        }
+
+        // Call Cohere's rerank API with retry logic
+        let reranked_indices =
+            match self.call_rerank_with_retry(&prompt, &documents, time_budget).await {
+                Ok(indices) => indices,
+                Err(PersonalizationError::DeadlineExceeded) => {
+                    // If the deadline is exceeded, return the original search result instead of an error
+                    return Ok(search_result);
+                }
+                Err(e) => return Err(e.into()),
+            };
+
+        debug!("Cohere rerank successful, reordering {} results", search_result.hits.len());
+
+        // Reorder the hits based on Cohere's reranking
+        let mut reranked_hits = Vec::new();
+        for index in reranked_indices.iter() {
+            if let Some(hit) = search_result.hits.get(*index) {
+                reranked_hits.push(hit.clone());
+            }
+        }
+
+        Ok(SearchResult { hits: reranked_hits, ..search_result })
+    }
+
+    async fn call_rerank_with_retry(
+        &self,
+        query: &str,
+        documents: &[String],
+        time_budget: TimeBudget,
+    ) -> Result<Vec<usize>, PersonalizationError> {
+        let request_body = CohereRerankRequest {
+            query: query.to_string(),
+            documents: documents.to_vec(),
+            model: "rerank-english-v3.0".to_string(),
+        };
+
+        // Retry loop similar to vector extraction
+        for attempt in 0..MAX_RETRIES {
+            let response_result = self.send_rerank_request(&request_body).await;
+
+            let retry_duration = match self.handle_response(response_result).await {
+                Ok(indices) => return Ok(indices),
+                Err(retry) => {
+                    warn!("Cohere rerank attempt #{} failed: {}", attempt, retry.error);
+
+                    if time_budget.exceeded() {
+                        warn!("Could not rerank due to deadline");
+                        return Err(PersonalizationError::DeadlineExceeded);
+                    } else {
+                        match retry.into_duration(attempt) {
+                            Ok(d) => d,
+                            Err(error) => return Err(error),
+                        }
+                    }
+                }
+            };
+
+            // randomly up to double the retry duration
+            let retry_duration = retry_duration
+                + rand::thread_rng().gen_range(std::time::Duration::ZERO..retry_duration);
+
+            warn!("Retrying after {}ms", retry_duration.as_millis());
+            tokio::time::sleep(retry_duration).await;
+        }
+
+        // Final attempt without retry
+        let response_result = self.send_rerank_request(&request_body).await;
+
+        match self.handle_response(response_result).await {
+            Ok(indices) => Ok(indices),
+            Err(retry) => Err(retry.into_error()),
+        }
+    }
+
+    async fn send_rerank_request(
+        &self,
+        request_body: &CohereRerankRequest,
+    ) -> Result<reqwest::Response, reqwest::Error> {
+        self.client
+            .post(COHERE_API_URL)
+            .header("Authorization", format!("Bearer {}", self.api_key))
+            .header("Content-Type", "application/json")
+            .json(request_body)
+            .send()
+            .await
+    }
+
+    async fn handle_response(
+        &self,
+        response_result: Result<reqwest::Response, reqwest::Error>,
+    ) -> Result<Vec<usize>, Retry> {
+        let response = match response_result {
+            Ok(r) => r,
+            Err(e) if e.is_timeout() => {
+                return Err(Retry::retry_later(PersonalizationError::Network(format!(
+                    "Request timeout: {}",
+                    e
+                ))));
+            }
+            Err(e) => {
+                return Err(Retry::retry_later(PersonalizationError::Network(format!(
+                    "Network error: {}",
+                    e
+                ))));
+            }
+        };
+
+        let status = response.status();
+        let status_code = status.as_u16();
+
+        if status.is_success() {
+            let rerank_response: CohereRerankResponse = match response.json().await {
+                Ok(r) => r,
+                Err(e) => {
+                    return Err(Retry::retry_later(PersonalizationError::Parse(format!(
+                        "Failed to parse response: {}",
+                        e
+                    ))));
+                }
+            };
+
+            // Extract indices from rerank results
+            let indices: Vec<usize> =
+                rerank_response.results.iter().map(|result| result.index as usize).collect();
+
+            return Ok(indices);
+        }
+
+        // Handle error status codes
+        let error_body = response.text().await.unwrap_or_else(|_| "Unknown error".to_string());
+
+        let retry = match status_code {
+            401 => Retry::give_up(PersonalizationError::Unauthorized),
+            429 => Retry::rate_limited(PersonalizationError::RateLimited),
+            400 => Retry::give_up(PersonalizationError::BadRequest(error_body)),
+            500..=599 => Retry::retry_later(PersonalizationError::InternalServerError(format!(
+                "Status {}: {}",
+                status_code, error_body
+            ))),
+            402..=499 => Retry::give_up(PersonalizationError::Api(format!(
+                "Status {}: {}",
+                status_code, error_body
+            ))),
+            _ => Retry::retry_later(PersonalizationError::Api(format!(
+                "Unexpected status {}: {}",
+                status_code, error_body
+            ))),
+        };
+
+        Err(retry)
+    }
+}
+
+#[derive(Serialize)]
+struct CohereRerankRequest {
+    query: String,
+    documents: Vec<String>,
+    model: String,
+}
+
+#[derive(Deserialize)]
+struct CohereRerankResponse {
+    results: Vec<CohereRerankResult>,
+}
+
+#[derive(Deserialize)]
+struct CohereRerankResult {
+    index: u32,
+}
+
+// Retry strategy similar to vector extraction
+struct Retry {
+    error: PersonalizationError,
+    strategy: RetryStrategy,
+}
+
+enum RetryStrategy {
+    GiveUp,
+    Retry,
+    RetryAfterRateLimit,
+}
+
+impl Retry {
+    fn give_up(error: PersonalizationError) -> Self {
+        Self { error, strategy: RetryStrategy::GiveUp }
+    }
+
+    fn retry_later(error: PersonalizationError) -> Self {
+        Self { error, strategy: RetryStrategy::Retry }
+    }
+
+    fn rate_limited(error: PersonalizationError) -> Self {
+        Self { error, strategy: RetryStrategy::RetryAfterRateLimit }
+    }
+
+    fn into_duration(self, attempt: u32) -> Result<Duration, PersonalizationError> {
+        match self.strategy {
+            RetryStrategy::GiveUp => Err(self.error),
+            RetryStrategy::Retry => {
+                // Exponential backoff: 10^attempt milliseconds
+                Ok(Duration::from_millis((10u64).pow(attempt)))
+            }
+            RetryStrategy::RetryAfterRateLimit => {
+                // Longer backoff for rate limits: 100ms + exponential
+                Ok(Duration::from_millis(100 + (10u64).pow(attempt)))
+            }
+        }
+    }
+
+    fn into_error(self) -> PersonalizationError {
+        self.error
+    }
+}
+
+pub enum PersonalizationService {
+    Cohere(CohereService),
+    Disabled,
+}
+
+impl PersonalizationService {
+    pub fn cohere(api_key: String) -> Self {
+        // If the API key is empty, consider the personalization service as disabled
+        if api_key.trim().is_empty() {
+            Self::disabled()
+        } else {
+            Self::Cohere(CohereService::new(api_key))
+        }
+    }
+
+    pub fn disabled() -> Self {
+        debug!("Personalization service disabled");
+        Self::Disabled
+    }
+
+    pub async fn rerank_search_results(
+        &self,
+        search_result: SearchResult,
+        personalize: &Personalize,
+        query: Option<&str>,
+        time_budget: TimeBudget,
+    ) -> Result<SearchResult, ResponseError> {
+        match self {
+            Self::Cohere(cohere_service) => {
+                cohere_service
+                    .rerank_search_results(search_result, personalize, query, time_budget)
+                    .await
+            }
+            Self::Disabled => Err(PersonalizationError::FeatureNotEnabled(
+                index_scheduler::error::FeatureNotEnabledError {
+                    disabled_action: "reranking search results",
+                    feature: "personalization",
+                    issue_link: "https://github.com/orgs/meilisearch/discussions/866",
+                },
+            )
+            .into()),
+        }
+    }
+}

crates/meilisearch/src/routes/indexes/compact.rs

@@ -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");

crates/meilisearch/src/routes/indexes/documents.rs

@@ -333,10 +333,12 @@ impl Aggregate for DocumentsDeletionAggregator {
 pub async fn delete_document(
     index_scheduler: GuardedData<ActionPolicy<{ actions::DOCUMENTS_DELETE }>, Data<IndexScheduler>>,
     path: web::Path<DocumentParam>,
+    params: AwebQueryParameter<CustomMetadataQuery, DeserrQueryParamError>,
     req: HttpRequest,
     opt: web::Data<Opt>,
     analytics: web::Data<Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
+    let CustomMetadataQuery { custom_metadata } = params.into_inner();
     let DocumentParam { index_uid, document_id } = path.into_inner();
     let index_uid = IndexUid::try_from(index_uid)?;
     let network = index_scheduler.network();
@@ -359,7 +361,10 @@ pub async fn delete_document(
     let dry_run = is_dry_run(&req, &opt)?;
     let task = {
         let index_scheduler = index_scheduler.clone();
-        tokio::task::spawn_blocking(move || index_scheduler.register(task, uid, dry_run)).await??
+        tokio::task::spawn_blocking(move || {
+            index_scheduler.register_with_custom_metadata(task, uid, custom_metadata, dry_run)
+        })
+        .await??
     };
 
     if network.sharding && !dry_run {
@@ -678,6 +683,19 @@ pub struct UpdateDocumentsQuery {
     #[param(value_type = char, default = ",", example = ";")]
     #[deserr(default, try_from(char) = from_char_csv_delimiter -> DeserrQueryParamError<InvalidDocumentCsvDelimiter>, error = DeserrQueryParamError<InvalidDocumentCsvDelimiter>)]
     pub csv_delimiter: Option<u8>,
+
+    #[param(example = "custom")]
+    #[deserr(default, error = DeserrQueryParamError<InvalidIndexCustomMetadata>)]
+    pub custom_metadata: Option<String>,
+}
+
+#[derive(Deserialize, Debug, Deserr, IntoParams)]
+#[deserr(error = DeserrQueryParamError, rename_all = camelCase, deny_unknown_fields)]
+#[into_params(parameter_in = Query, rename_all = "camelCase")]
+pub struct CustomMetadataQuery {
+    #[param(example = "custom")]
+    #[deserr(default, error = DeserrQueryParamError<InvalidIndexCustomMetadata>)]
+    pub custom_metadata: Option<String>,
 }
 
 fn from_char_csv_delimiter(
@@ -819,6 +837,7 @@ pub async fn replace_documents(
         body,
         IndexDocumentsMethod::ReplaceDocuments,
         uid,
+        params.custom_metadata,
         dry_run,
         allow_index_creation,
         &req,
@@ -921,6 +940,7 @@ pub async fn update_documents(
         body,
         IndexDocumentsMethod::UpdateDocuments,
         uid,
+        params.custom_metadata,
         dry_run,
         allow_index_creation,
         &req,
@@ -940,6 +960,7 @@ async fn document_addition(
     body: Payload,
     method: IndexDocumentsMethod,
     task_id: Option<TaskId>,
+    custom_metadata: Option<String>,
     dry_run: bool,
     allow_index_creation: bool,
     req: &HttpRequest,
@@ -1065,8 +1086,10 @@ async fn document_addition(
     };
 
     let scheduler = index_scheduler.clone();
-    let task = match tokio::task::spawn_blocking(move || scheduler.register(task, task_id, dry_run))
-        .await?
+    let task = match tokio::task::spawn_blocking(move || {
+        scheduler.register_with_custom_metadata(task, task_id, custom_metadata, dry_run)
+    })
+    .await?
     {
         Ok(task) => task,
         Err(e) => {
@@ -1130,7 +1153,7 @@ async fn copy_body_to_file(
 /// Delete a set of documents based on an array of document ids.
 #[utoipa::path(
     post,
-    path = "{indexUid}/delete-batch",
+    path = "{indexUid}/documents/delete-batch",
     tag = "Documents",
     security(("Bearer" = ["documents.delete", "documents.*", "*"])),
     params(
@@ -1161,11 +1184,14 @@ pub async fn delete_documents_batch(
     index_scheduler: GuardedData<ActionPolicy<{ actions::DOCUMENTS_DELETE }>, Data<IndexScheduler>>,
     index_uid: web::Path<String>,
     body: web::Json<Vec<Value>>,
+    params: AwebQueryParameter<CustomMetadataQuery, DeserrQueryParamError>,
     req: HttpRequest,
     opt: web::Data<Opt>,
     analytics: web::Data<Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
     debug!(parameters = ?body, "Delete documents by batch");
+    let CustomMetadataQuery { custom_metadata } = params.into_inner();
+
     let index_uid = IndexUid::try_from(index_uid.into_inner())?;
     let network = index_scheduler.network();
 
@@ -1190,7 +1216,10 @@ pub async fn delete_documents_batch(
     let dry_run = is_dry_run(&req, &opt)?;
     let task = {
         let index_scheduler = index_scheduler.clone();
-        tokio::task::spawn_blocking(move || index_scheduler.register(task, uid, dry_run)).await??
+        tokio::task::spawn_blocking(move || {
+            index_scheduler.register_with_custom_metadata(task, uid, custom_metadata, dry_run)
+        })
+        .await??
     };
 
     if network.sharding && !dry_run {
@@ -1244,12 +1273,15 @@ pub struct DocumentDeletionByFilter {
 pub async fn delete_documents_by_filter(
     index_scheduler: GuardedData<ActionPolicy<{ actions::DOCUMENTS_DELETE }>, Data<IndexScheduler>>,
     index_uid: web::Path<String>,
+    params: AwebQueryParameter<CustomMetadataQuery, DeserrQueryParamError>,
     body: AwebJson<DocumentDeletionByFilter, DeserrJsonError>,
     req: HttpRequest,
     opt: web::Data<Opt>,
     analytics: web::Data<Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
     debug!(parameters = ?body, "Delete documents by filter");
+    let CustomMetadataQuery { custom_metadata } = params.into_inner();
+
     let index_uid = IndexUid::try_from(index_uid.into_inner())?;
     let index_uid = index_uid.into_inner();
     let filter = body.into_inner();
@@ -1282,7 +1314,10 @@ pub async fn delete_documents_by_filter(
     let dry_run = is_dry_run(&req, &opt)?;
     let task = {
         let index_scheduler = index_scheduler.clone();
-        tokio::task::spawn_blocking(move || index_scheduler.register(task, uid, dry_run)).await??
+        tokio::task::spawn_blocking(move || {
+            index_scheduler.register_with_custom_metadata(task, uid, custom_metadata, dry_run)
+        })
+        .await??
     };
 
     if network.sharding && !dry_run {
@@ -1372,12 +1407,14 @@ impl Aggregate for EditDocumentsByFunctionAggregator {
 pub async fn edit_documents_by_function(
     index_scheduler: GuardedData<ActionPolicy<{ actions::DOCUMENTS_ALL }>, Data<IndexScheduler>>,
     index_uid: web::Path<String>,
-    params: AwebJson<DocumentEditionByFunction, DeserrJsonError>,
+    params: AwebQueryParameter<CustomMetadataQuery, DeserrQueryParamError>,
+    body: AwebJson<DocumentEditionByFunction, DeserrJsonError>,
     req: HttpRequest,
     opt: web::Data<Opt>,
     analytics: web::Data<Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
-    debug!(parameters = ?params, "Edit documents by function");
+    debug!(parameters = ?body, "Edit documents by function");
+    let CustomMetadataQuery { custom_metadata } = params.into_inner();
 
     index_scheduler
         .features()
@@ -1387,23 +1424,23 @@ pub async fn edit_documents_by_function(
 
     let index_uid = IndexUid::try_from(index_uid.into_inner())?;
     let index_uid = index_uid.into_inner();
-    let params = params.into_inner();
+    let body = body.into_inner();
 
     analytics.publish(
         EditDocumentsByFunctionAggregator {
-            filtered: params.filter.is_some(),
-            with_context: params.context.is_some(),
+            filtered: body.filter.is_some(),
+            with_context: body.context.is_some(),
             index_creation: index_scheduler.index(&index_uid).is_err(),
         },
         &req,
     );
 
     let engine = milli::rhai::Engine::new();
-    if let Err(e) = engine.compile(&params.function) {
+    if let Err(e) = engine.compile(&body.function) {
         return Err(ResponseError::from_msg(e.to_string(), Code::BadRequest));
     }
 
-    if let Some(ref filter) = params.filter {
+    if let Some(ref filter) = body.filter {
         // we ensure the filter is well formed before enqueuing it
         crate::search::parse_filter(
             filter,
@@ -1414,8 +1451,8 @@ pub async fn edit_documents_by_function(
     }
     let task = KindWithContent::DocumentEdition {
         index_uid: index_uid.clone(),
-        filter_expr: params.filter.clone(),
-        context: match params.context.clone() {
+        filter_expr: body.filter.clone(),
+        context: match body.context.clone() {
             Some(Value::Object(m)) => Some(m),
             None => None,
             _ => {
@@ -1425,18 +1462,21 @@ pub async fn edit_documents_by_function(
                 ))
             }
         },
-        function: params.function.clone(),
+        function: body.function.clone(),
     };
 
     let uid = get_task_id(&req, &opt)?;
     let dry_run = is_dry_run(&req, &opt)?;
     let task = {
         let index_scheduler = index_scheduler.clone();
-        tokio::task::spawn_blocking(move || index_scheduler.register(task, uid, dry_run)).await??
+        tokio::task::spawn_blocking(move || {
+            index_scheduler.register_with_custom_metadata(task, uid, custom_metadata, dry_run)
+        })
+        .await??
     };
 
     if network.sharding && !dry_run {
-        proxy(&index_scheduler, &index_uid, &req, network, Body::Inline(params), &task).await?;
+        proxy(&index_scheduler, &index_uid, &req, network, Body::Inline(body), &task).await?;
     }
 
     let task: SummarizedTaskView = task.into();
@@ -1477,12 +1517,14 @@ pub async fn edit_documents_by_function(
 pub async fn clear_all_documents(
     index_scheduler: GuardedData<ActionPolicy<{ actions::DOCUMENTS_DELETE }>, Data<IndexScheduler>>,
     index_uid: web::Path<String>,
+    params: AwebQueryParameter<CustomMetadataQuery, DeserrQueryParamError>,
     req: HttpRequest,
     opt: web::Data<Opt>,
     analytics: web::Data<Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
     let index_uid = IndexUid::try_from(index_uid.into_inner())?;
     let network = index_scheduler.network();
+    let CustomMetadataQuery { custom_metadata } = params.into_inner();
 
     analytics.publish(
         DocumentsDeletionAggregator {
@@ -1501,7 +1543,10 @@ pub async fn clear_all_documents(
     let task = {
         let index_scheduler = index_scheduler.clone();
 
-        tokio::task::spawn_blocking(move || index_scheduler.register(task, uid, dry_run)).await??
+        tokio::task::spawn_blocking(move || {
+            index_scheduler.register_with_custom_metadata(task, uid, custom_metadata, dry_run)
+        })
+        .await??
     };
 
     if network.sharding && !dry_run {

crates/meilisearch/src/routes/indexes/facet_search.rs

@@ -343,6 +343,7 @@ impl From<FacetSearchQuery> for SearchQuery {
             hybrid,
             ranking_score_threshold,
             locales,
+            personalize: None,
         }
     }
 }

crates/meilisearch/src/routes/indexes/mod.rs

@@ -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)),
     );
 }
 

crates/meilisearch/src/routes/indexes/search.rs

@@ -13,6 +13,7 @@ use meilisearch_types::serde_cs::vec::CS;
 use serde_json::Value;
 use tracing::debug;
 use utoipa::{IntoParams, OpenApi};
+use uuid::Uuid;
 
 use crate::analytics::Analytics;
 use crate::error::MeilisearchHttpError;
@@ -21,11 +22,12 @@ use crate::extractors::authentication::GuardedData;
 use crate::extractors::sequential_extractor::SeqHandler;
 use crate::metrics::MEILISEARCH_DEGRADED_SEARCH_REQUESTS;
 use crate::routes::indexes::search_analytics::{SearchAggregator, SearchGET, SearchPOST};
+use crate::routes::parse_include_metadata_header;
 use crate::search::{
-    add_search_rules, perform_search, HybridQuery, MatchingStrategy, RankingScoreThreshold,
-    RetrieveVectors, SearchKind, SearchQuery, SearchResult, SemanticRatio, DEFAULT_CROP_LENGTH,
-    DEFAULT_CROP_MARKER, DEFAULT_HIGHLIGHT_POST_TAG, DEFAULT_HIGHLIGHT_PRE_TAG,
-    DEFAULT_SEARCH_LIMIT, DEFAULT_SEARCH_OFFSET, DEFAULT_SEMANTIC_RATIO,
+    add_search_rules, perform_search, HybridQuery, MatchingStrategy, Personalize,
+    RankingScoreThreshold, RetrieveVectors, SearchKind, SearchParams, SearchQuery, SearchResult,
+    SemanticRatio, DEFAULT_CROP_LENGTH, DEFAULT_CROP_MARKER, DEFAULT_HIGHLIGHT_POST_TAG,
+    DEFAULT_HIGHLIGHT_PRE_TAG, DEFAULT_SEARCH_LIMIT, DEFAULT_SEARCH_OFFSET, DEFAULT_SEMANTIC_RATIO,
 };
 use crate::search_queue::SearchQueue;
 
@@ -132,6 +134,8 @@ pub struct SearchQueryGet {
     #[deserr(default, error = DeserrQueryParamError<InvalidSearchLocales>)]
     #[param(value_type = Vec<Locale>, explode = false)]
     pub locales: Option<CS<Locale>>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidSearchPersonalizeUserContext>)]
+    pub personalize_user_context: Option<String>,
 }
 
 #[derive(Debug, Clone, Copy, PartialEq, deserr::Deserr)]
@@ -203,6 +207,9 @@ impl TryFrom<SearchQueryGet> for SearchQuery {
             ));
         }
 
+        let personalize =
+            other.personalize_user_context.map(|user_context| Personalize { user_context });
+
         Ok(Self {
             q: other.q,
             // `media` not supported for `GET`
@@ -232,6 +239,7 @@ impl TryFrom<SearchQueryGet> for SearchQuery {
             hybrid,
             ranking_score_threshold: other.ranking_score_threshold.map(|o| o.0),
             locales: other.locales.map(|o| o.into_iter().collect()),
+            personalize,
         })
     }
 }
@@ -320,12 +328,14 @@ pub fn fix_sort_query_parameters(sort_query: &str) -> Vec<String> {
 pub async fn search_with_url_query(
     index_scheduler: GuardedData<ActionPolicy<{ actions::SEARCH }>, Data<IndexScheduler>>,
     search_queue: web::Data<SearchQueue>,
+    personalization_service: web::Data<crate::personalization::PersonalizationService>,
     index_uid: web::Path<String>,
     params: AwebQueryParameter<SearchQueryGet, DeserrQueryParamError>,
     req: HttpRequest,
     analytics: web::Data<Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
-    debug!(parameters = ?params, "Search get");
+    let request_uid = Uuid::now_v7();
+    debug!(request_uid = ?request_uid, parameters = ?params, "Search get");
     let index_uid = IndexUid::try_from(index_uid.into_inner())?;
 
     let mut query: SearchQuery = params.into_inner().try_into()?;
@@ -339,31 +349,56 @@ pub async fn search_with_url_query(
 
     let index = index_scheduler.index(&index_uid)?;
 
+    // Extract personalization and query string before moving query
+    let personalize = query.personalize.take();
+
     let search_kind =
         search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index)?;
     let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors);
+
+    // Save the query string for personalization if requested
+    let personalize_query = personalize.is_some().then(|| query.q.clone()).flatten();
+
     let permit = search_queue.try_get_search_permit().await?;
+    let include_metadata = parse_include_metadata_header(&req);
+
     let search_result = tokio::task::spawn_blocking(move || {
         perform_search(
-            index_uid.to_string(),
+            SearchParams {
+                index_uid: index_uid.to_string(),
+                query,
+                search_kind,
+                retrieve_vectors: retrieve_vector,
+                features: index_scheduler.features(),
+                request_uid,
+                include_metadata,
+            },
             &index,
-            query,
-            search_kind,
-            retrieve_vector,
-            index_scheduler.features(),
         )
     })
     .await;
     permit.drop().await;
     let search_result = search_result?;
-    if let Ok(ref search_result) = search_result {
+    if let Ok((search_result, _)) = search_result.as_ref() {
         aggregate.succeed(search_result);
     }
     analytics.publish(aggregate, &req);
 
-    let search_result = search_result?;
+    let (mut search_result, time_budget) = search_result?;
 
-    debug!(returns = ?search_result, "Search get");
+    // Apply personalization if requested
+    if let Some(personalize) = personalize.as_ref() {
+        search_result = personalization_service
+            .rerank_search_results(
+                search_result,
+                personalize,
+                personalize_query.as_deref(),
+                time_budget,
+            )
+            .await?;
+    }
+
+    debug!(request_uid = ?request_uid, returns = ?search_result, "Search get");
     Ok(HttpResponse::Ok().json(search_result))
 }
 
@@ -426,15 +461,17 @@ pub async fn search_with_url_query(
 pub async fn search_with_post(
     index_scheduler: GuardedData<ActionPolicy<{ actions::SEARCH }>, Data<IndexScheduler>>,
     search_queue: web::Data<SearchQueue>,
+    personalization_service: web::Data<crate::personalization::PersonalizationService>,
     index_uid: web::Path<String>,
     params: AwebJson<SearchQuery, DeserrJsonError>,
     req: HttpRequest,
     analytics: web::Data<Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
     let index_uid = IndexUid::try_from(index_uid.into_inner())?;
+    let request_uid = Uuid::now_v7();
 
     let mut query = params.into_inner();
-    debug!(parameters = ?query, "Search post");
+    debug!(request_uid = ?request_uid, parameters = ?query, "Search post");
 
     // Tenant token search_rules.
     if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) {
@@ -445,25 +482,37 @@ pub async fn search_with_post(
 
     let index = index_scheduler.index(&index_uid)?;
 
+    // Extract personalization and query string before moving query
+    let personalize = query.personalize.take();
+
     let search_kind =
         search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index)?;
     let retrieve_vectors = RetrieveVectors::new(query.retrieve_vectors);
 
+    let include_metadata = parse_include_metadata_header(&req);
+
+    // Save the query string for personalization if requested
+    let personalize_query = personalize.is_some().then(|| query.q.clone()).flatten();
+
     let permit = search_queue.try_get_search_permit().await?;
     let search_result = tokio::task::spawn_blocking(move || {
         perform_search(
-            index_uid.to_string(),
+            SearchParams {
+                index_uid: index_uid.to_string(),
+                query,
+                search_kind,
+                retrieve_vectors,
+                features: index_scheduler.features(),
+                request_uid,
+                include_metadata,
+            },
             &index,
-            query,
-            search_kind,
-            retrieve_vectors,
-            index_scheduler.features(),
         )
     })
     .await;
     permit.drop().await;
     let search_result = search_result?;
-    if let Ok(ref search_result) = search_result {
+    if let Ok((ref search_result, _)) = search_result {
         aggregate.succeed(search_result);
         if search_result.degraded {
             MEILISEARCH_DEGRADED_SEARCH_REQUESTS.inc();
@@ -471,9 +520,21 @@ pub async fn search_with_post(
     }
     analytics.publish(aggregate, &req);
 
-    let search_result = search_result?;
+    let (mut search_result, time_budget) = search_result?;
 
-    debug!(returns = ?search_result, "Search post");
+    // Apply personalization if requested
+    if let Some(personalize) = personalize.as_ref() {
+        search_result = personalization_service
+            .rerank_search_results(
+                search_result,
+                personalize,
+                personalize_query.as_deref(),
+                time_budget,
+            )
+            .await?;
+    }
+
+    debug!(request_uid = ?request_uid, returns = ?search_result, "Search post");
     Ok(HttpResponse::Ok().json(search_result))
 }
 

crates/meilisearch/src/routes/indexes/search_analytics.rs

@@ -7,6 +7,7 @@ use serde_json::{json, Value};
 
 use crate::aggregate_methods;
 use crate::analytics::{Aggregate, AggregateMethod};
+use crate::metrics::MEILISEARCH_PERSONALIZED_SEARCH_REQUESTS;
 use crate::search::{
     SearchQuery, SearchResult, DEFAULT_CROP_LENGTH, DEFAULT_CROP_MARKER,
     DEFAULT_HIGHLIGHT_POST_TAG, DEFAULT_HIGHLIGHT_PRE_TAG, DEFAULT_SEARCH_LIMIT,
@@ -95,6 +96,9 @@ pub struct SearchAggregator<Method: AggregateMethod> {
     show_ranking_score_details: bool,
     ranking_score_threshold: bool,
 
+    // personalization
+    total_personalized: usize,
+
     marker: std::marker::PhantomData<Method>,
 }
 
@@ -129,6 +133,7 @@ impl<Method: AggregateMethod> SearchAggregator<Method> {
             hybrid,
             ranking_score_threshold,
             locales,
+            personalize,
         } = query;
 
         let mut ret = Self::default();
@@ -204,6 +209,12 @@ impl<Method: AggregateMethod> SearchAggregator<Method> {
             ret.locales = locales.iter().copied().collect();
         }
 
+        // personalization
+        if personalize.is_some() {
+            ret.total_personalized = 1;
+            MEILISEARCH_PERSONALIZED_SEARCH_REQUESTS.inc();
+        }
+
         ret.highlight_pre_tag = *highlight_pre_tag != DEFAULT_HIGHLIGHT_PRE_TAG();
         ret.highlight_post_tag = *highlight_post_tag != DEFAULT_HIGHLIGHT_POST_TAG();
         ret.crop_marker = *crop_marker != DEFAULT_CROP_MARKER();
@@ -234,6 +245,8 @@ impl<Method: AggregateMethod> SearchAggregator<Method> {
             facet_stats: _,
             degraded,
             used_negative_operator,
+            request_uid: _,
+            metadata: _,
         } = result;
 
         self.total_succeeded = self.total_succeeded.saturating_add(1);
@@ -294,6 +307,7 @@ impl<Method: AggregateMethod> Aggregate for SearchAggregator<Method> {
             total_used_negative_operator,
             ranking_score_threshold,
             mut locales,
+            total_personalized,
             marker: _,
         } = *new;
 
@@ -379,6 +393,9 @@ impl<Method: AggregateMethod> Aggregate for SearchAggregator<Method> {
         // locales
         self.locales.append(&mut locales);
 
+        // personalization
+        self.total_personalized = self.total_personalized.saturating_add(total_personalized);
+
         self
     }
 
@@ -424,6 +441,7 @@ impl<Method: AggregateMethod> Aggregate for SearchAggregator<Method> {
             total_used_negative_operator,
             ranking_score_threshold,
             locales,
+            total_personalized,
             marker: _,
         } = *self;
 
@@ -497,6 +515,9 @@ impl<Method: AggregateMethod> Aggregate for SearchAggregator<Method> {
                 "show_ranking_score_details": show_ranking_score_details,
                 "ranking_score_threshold": ranking_score_threshold,
             },
+            "personalization": {
+                "total_personalized": total_personalized,
+            },
         })
     }
 }

crates/meilisearch/src/routes/metrics.rs

@@ -180,12 +180,6 @@ pub async fn get_metrics(
 
     let response = String::from_utf8(buffer).expect("Failed to convert bytes to string");
 
-    // We cannot specify the version with ContentType(TEXT_PLAIN_UTF_8) so we have to write everything by hand :(
-    // see the following for what should be returned: https://prometheus.io/docs/instrumenting/content_negotiation/#content-type-response
-    let content_type = ("content-type", "text/plain; version=0.0.4; charset=utf-8");
-
-    Ok(HttpResponse::Ok()
-        // .insert_header(header::ContentType(mime::TEXT_PLAIN_UTF_8))
-        .insert_header(content_type)
-        .body(response))
+    let content_type = ("content-type", prometheus::TEXT_FORMAT);
+    Ok(HttpResponse::Ok().insert_header(content_type).body(response))
 }

crates/meilisearch/src/routes/mod.rs

@@ -41,10 +41,13 @@ use crate::routes::indexes::IndexView;
 use crate::routes::multi_search::SearchResults;
 use crate::routes::network::{Network, Remote};
 use crate::routes::swap_indexes::SwapIndexesPayload;
-use crate::routes::webhooks::{WebhookResults, WebhookSettings, WebhookWithMetadata};
+use crate::routes::webhooks::{
+    WebhookResults, WebhookSettings, WebhookWithMetadataRedactedAuthorization,
+};
 use crate::search::{
     FederatedSearch, FederatedSearchResult, Federation, FederationOptions, MergeFacets,
     SearchQueryWithIndex, SearchResultWithIndex, SimilarQuery, SimilarResult,
+    INCLUDE_METADATA_HEADER,
 };
 use crate::search_queue::SearchQueue;
 use crate::Opt;
@@ -102,7 +105,7 @@ mod webhooks;
         url = "/",
         description = "Local server",
     )),
-    components(schemas(PaginationView<KeyView>, PaginationView<IndexView>, IndexView, DocumentDeletionByFilter, AllBatches, BatchStats, ProgressStepView, ProgressView, BatchView, RuntimeTogglableFeatures, SwapIndexesPayload, DocumentEditionByFunction, MergeFacets, FederationOptions, SearchQueryWithIndex, Federation, FederatedSearch, FederatedSearchResult, SearchResults, SearchResultWithIndex, SimilarQuery, SimilarResult, PaginationView<serde_json::Value>, BrowseQuery, UpdateIndexRequest, IndexUid, IndexCreateRequest, KeyView, Action, CreateApiKey, UpdateStderrLogs, LogMode, GetLogs, IndexStats, Stats, HealthStatus, HealthResponse, VersionResponse, Code, ErrorType, AllTasks, TaskView, Status, DetailsView, ResponseError, Settings<Unchecked>, Settings<Checked>, TypoSettings, MinWordSizeTyposSetting, FacetingSettings, PaginationSettings, SummarizedTaskView, Kind, Network, Remote, FilterableAttributesRule, FilterableAttributesPatterns, AttributePatterns, FilterableAttributesFeatures, FilterFeatures, Export, WebhookSettings, WebhookResults, WebhookWithMetadata, meilisearch_types::milli::vector::VectorStoreBackend))
+    components(schemas(PaginationView<KeyView>, PaginationView<IndexView>, IndexView, DocumentDeletionByFilter, AllBatches, BatchStats, ProgressStepView, ProgressView, BatchView, RuntimeTogglableFeatures, SwapIndexesPayload, DocumentEditionByFunction, MergeFacets, FederationOptions, SearchQueryWithIndex, Federation, FederatedSearch, FederatedSearchResult, SearchResults, SearchResultWithIndex, SimilarQuery, SimilarResult, PaginationView<serde_json::Value>, BrowseQuery, UpdateIndexRequest, IndexUid, IndexCreateRequest, KeyView, Action, CreateApiKey, UpdateStderrLogs, LogMode, GetLogs, IndexStats, Stats, HealthStatus, HealthResponse, VersionResponse, Code, ErrorType, AllTasks, TaskView, Status, DetailsView, ResponseError, Settings<Unchecked>, Settings<Checked>, TypoSettings, MinWordSizeTyposSetting, FacetingSettings, PaginationSettings, SummarizedTaskView, Kind, Network, Remote, FilterableAttributesRule, FilterableAttributesPatterns, AttributePatterns, FilterableAttributesFeatures, FilterFeatures, Export, WebhookSettings, WebhookResults, WebhookWithMetadataRedactedAuthorization, meilisearch_types::milli::vector::VectorStoreBackend))
 )]
 pub struct MeilisearchApi;
 
@@ -184,6 +187,18 @@ pub fn is_dry_run(req: &HttpRequest, opt: &Opt) -> Result<bool, ResponseError> {
         .is_some_and(|s| s.to_lowercase() == "true"))
 }
 
+/// Parse the `Meili-Include-Metadata` header from an HTTP request.
+///
+/// Returns `true` if the header is present and set to "true" or "1" (case-insensitive).
+/// Returns `false` if the header is not present or has any other value.
+pub fn parse_include_metadata_header(req: &HttpRequest) -> bool {
+    req.headers()
+        .get(INCLUDE_METADATA_HEADER)
+        .and_then(|h| h.to_str().ok())
+        .map(|v| matches!(v.to_lowercase().as_str(), "true" | "1"))
+        .unwrap_or(false)
+}
+
 #[derive(Debug, Serialize, Deserialize, ToSchema)]
 #[serde(rename_all = "camelCase")]
 pub struct SummarizedTaskView {
@@ -203,6 +218,8 @@ pub struct SummarizedTaskView {
         deserialize_with = "time::serde::rfc3339::deserialize"
     )]
     enqueued_at: OffsetDateTime,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    custom_metadata: Option<String>,
 }
 
 impl From<Task> for SummarizedTaskView {
@@ -213,6 +230,7 @@ impl From<Task> for SummarizedTaskView {
             status: task.status,
             kind: task.kind.as_kind(),
             enqueued_at: task.enqueued_at,
+            custom_metadata: task.custom_metadata,
         }
     }
 }

crates/meilisearch/src/routes/multi_search.rs

@@ -9,6 +9,7 @@ use meilisearch_types::keys::actions;
 use serde::Serialize;
 use tracing::debug;
 use utoipa::{OpenApi, ToSchema};
+use uuid::Uuid;
 
 use super::multi_search_analytics::MultiSearchAggregator;
 use crate::analytics::Analytics;
@@ -17,10 +18,11 @@ use crate::extractors::authentication::policies::ActionPolicy;
 use crate::extractors::authentication::{AuthenticationError, GuardedData};
 use crate::extractors::sequential_extractor::SeqHandler;
 use crate::routes::indexes::search::search_kind;
+use crate::routes::parse_include_metadata_header;
 use crate::search::{
     add_search_rules, perform_federated_search, perform_search, FederatedSearch,
-    FederatedSearchResult, RetrieveVectors, SearchQueryWithIndex, SearchResultWithIndex,
-    PROXY_SEARCH_HEADER, PROXY_SEARCH_HEADER_VALUE,
+    FederatedSearchResult, RetrieveVectors, SearchParams, SearchQueryWithIndex,
+    SearchResultWithIndex, PROXY_SEARCH_HEADER, PROXY_SEARCH_HEADER_VALUE,
 };
 use crate::search_queue::SearchQueue;
 
@@ -144,6 +146,7 @@ pub struct SearchResults {
 pub async fn multi_search_with_post(
     index_scheduler: GuardedData<ActionPolicy<{ actions::SEARCH }>, Data<IndexScheduler>>,
     search_queue: Data<SearchQueue>,
+    personalization_service: web::Data<crate::personalization::PersonalizationService>,
     params: AwebJson<FederatedSearch, DeserrJsonError>,
     req: HttpRequest,
     analytics: web::Data<Analytics>,
@@ -151,6 +154,7 @@ pub async fn multi_search_with_post(
     // Since we don't want to process half of the search requests and then get a permit refused
     // we're going to get one permit for the whole duration of the multi-search request.
     let permit = search_queue.try_get_search_permit().await?;
+    let request_uid = Uuid::now_v7();
 
     let federated_search = params.into_inner();
 
@@ -186,16 +190,31 @@ pub async fn multi_search_with_post(
         err
     })?;
 
+    let include_metadata = parse_include_metadata_header(&req);
     let response = match federation {
         Some(federation) => {
+            debug!(
+                request_uid = ?request_uid,
+                federation = ?federation,
+                parameters = ?queries,
+                "Federated-search"
+            );
+
             // check remote header
             let is_proxy = req
                 .headers()
                 .get(PROXY_SEARCH_HEADER)
                 .is_some_and(|value| value.as_bytes() == PROXY_SEARCH_HEADER_VALUE.as_bytes());
-            let search_result =
-                perform_federated_search(&index_scheduler, queries, federation, features, is_proxy)
-                    .await;
+            let search_result = perform_federated_search(
+                &index_scheduler,
+                queries,
+                federation,
+                features,
+                is_proxy,
+                request_uid,
+                include_metadata,
+            )
+            .await;
             permit.drop().await;
 
             if search_result.is_ok() {
@@ -203,6 +222,13 @@ pub async fn multi_search_with_post(
             }
 
             analytics.publish(multi_aggregate, &req);
+
+            debug!(
+                request_uid = ?request_uid,
+                returns = ?search_result,
+                "Federated-search"
+            );
+
             HttpResponse::Ok().json(search_result?)
         }
         None => {
@@ -211,12 +237,17 @@ pub async fn multi_search_with_post(
             // changes.
             let search_results: Result<_, (ResponseError, usize)> = async {
                 let mut search_results = Vec::with_capacity(queries.len());
-                for (query_index, (index_uid, query, federation_options)) in queries
+                for (query_index, (index_uid, mut query, federation_options)) in queries
                     .into_iter()
                     .map(SearchQueryWithIndex::into_index_query_federation)
                     .enumerate()
                 {
-                    debug!(on_index = query_index, parameters = ?query, "Multi-search");
+                    debug!(
+                        request_uid = ?request_uid,
+                        on_index = query_index,
+                        parameters = ?query,
+                        "Multi-search"
+                    );
 
                     if federation_options.is_some() {
                         return Err((
@@ -239,6 +270,13 @@ pub async fn multi_search_with_post(
                         })
                         .with_index(query_index)?;
 
+                    // Extract personalization and query string before moving query
+                    let personalize = query.personalize.take();
+
+                    // Save the query string for personalization if requested
+                    let personalize_query =
+                        personalize.is_some().then(|| query.q.clone()).flatten();
+
                     let index_uid_str = index_uid.to_string();
 
                     let search_kind = search_kind(
@@ -250,22 +288,40 @@ pub async fn multi_search_with_post(
                     .with_index(query_index)?;
                     let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors);
 
-                    let search_result = tokio::task::spawn_blocking(move || {
+                    let (mut search_result, time_budget) = tokio::task::spawn_blocking(move || {
                         perform_search(
-                            index_uid_str.clone(),
+                            SearchParams {
+                                index_uid: index_uid_str.clone(),
+                                query,
+                                search_kind,
+                                retrieve_vectors: retrieve_vector,
+                                features,
+                                request_uid,
+                                include_metadata,
+                            },
                             &index,
-                            query,
-                            search_kind,
-                            retrieve_vector,
-                            features,
                         )
                     })
                     .await
+                    .with_index(query_index)?
                     .with_index(query_index)?;
 
+                    // Apply personalization if requested
+                    if let Some(personalize) = personalize.as_ref() {
+                        search_result = personalization_service
+                            .rerank_search_results(
+                                search_result,
+                                personalize,
+                                personalize_query.as_deref(),
+                                time_budget,
+                            )
+                            .await
+                            .with_index(query_index)?;
+                    }
+
                     search_results.push(SearchResultWithIndex {
                         index_uid: index_uid.into_inner(),
-                        result: search_result.with_index(query_index)?,
+                        result: search_result,
                     });
                 }
                 Ok(search_results)
@@ -286,7 +342,11 @@ pub async fn multi_search_with_post(
                 err
             })?;
 
-            debug!(returns = ?search_results, "Multi-search");
+            debug!(
+                request_uid = ?request_uid,
+                returns = ?search_results,
+                "Multi-search"
+            );
 
             HttpResponse::Ok().json(SearchResults { results: search_results })
         }

crates/meilisearch/src/routes/multi_search_analytics.rs

@@ -67,6 +67,7 @@ impl MultiSearchAggregator {
             hybrid: _,
             ranking_score_threshold: _,
             locales: _,
+            personalize: _,
         } in &federated_search.queries
         {
             if let Some(federation_options) = federation_options {

crates/meilisearch/src/routes/tasks_test.rs

@@ -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]

crates/meilisearch/src/routes/webhooks.rs

@@ -1,3 +1,4 @@
+use core::convert::Infallible;
 use std::collections::BTreeMap;
 use std::str::FromStr;
 
@@ -7,7 +8,6 @@ use actix_http::header::{
 };
 use actix_web::web::{self, Data, Path};
 use actix_web::{HttpRequest, HttpResponse};
-use core::convert::Infallible;
 use deserr::actix_web::AwebJson;
 use deserr::{DeserializeError, Deserr, ValuePointerRef};
 use index_scheduler::IndexScheduler;
@@ -24,12 +24,12 @@ use tracing::debug;
 use url::Url;
 use utoipa::{OpenApi, ToSchema};
 use uuid::Uuid;
+use WebhooksError::*;
 
 use crate::analytics::{Aggregate, Analytics};
 use crate::extractors::authentication::policies::ActionPolicy;
 use crate::extractors::authentication::GuardedData;
 use crate::extractors::sequential_extractor::SeqHandler;
-use WebhooksError::*;
 
 #[derive(OpenApi)]
 #[openapi(
@@ -90,7 +90,7 @@ fn deny_immutable_fields_webhook(
 #[derive(Debug, Serialize, ToSchema)]
 #[serde(rename_all = "camelCase")]
 #[schema(rename_all = "camelCase")]
-pub(super) struct WebhookWithMetadata {
+pub(super) struct WebhookWithMetadataRedactedAuthorization {
     uuid: Uuid,
     is_editable: bool,
     #[schema(value_type = WebhookSettings)]
@@ -98,8 +98,9 @@ pub(super) struct WebhookWithMetadata {
     webhook: Webhook,
 }
 
-impl WebhookWithMetadata {
-    pub fn from(uuid: Uuid, webhook: Webhook) -> Self {
+impl WebhookWithMetadataRedactedAuthorization {
+    pub fn from(uuid: Uuid, mut webhook: Webhook) -> Self {
+        webhook.redact_authorization_header();
         Self { uuid, is_editable: uuid != Uuid::nil(), webhook }
     }
 }
@@ -107,7 +108,7 @@ impl WebhookWithMetadata {
 #[derive(Debug, Serialize, ToSchema)]
 #[serde(rename_all = "camelCase")]
 pub(super) struct WebhookResults {
-    results: Vec<WebhookWithMetadata>,
+    results: Vec<WebhookWithMetadataRedactedAuthorization>,
 }
 
 #[utoipa::path(
@@ -150,7 +151,7 @@ async fn get_webhooks(
     let results = webhooks
         .webhooks
         .into_iter()
-        .map(|(uuid, webhook)| WebhookWithMetadata::from(uuid, webhook))
+        .map(|(uuid, webhook)| WebhookWithMetadataRedactedAuthorization::from(uuid, webhook))
         .collect::<Vec<_>>();
     let results = WebhookResults { results };
 
@@ -301,7 +302,7 @@ fn check_changed(uuid: Uuid, webhook: &Webhook) -> Result<(), WebhooksError> {
     tag = "Webhooks",
     security(("Bearer" = ["webhooks.get", "webhooks.*", "*.get", "*"])),
     responses(
-        (status = 200, description = "Webhook found", body = WebhookWithMetadata, content_type = "application/json", example = json!({
+        (status = 200, description = "Webhook found", body = WebhookWithMetadataRedactedAuthorization, content_type = "application/json", example = json!({
             "uuid": "550e8400-e29b-41d4-a716-446655440000",
             "url": "https://your.site/on-tasks-completed",
             "headers": {
@@ -324,7 +325,7 @@ async fn get_webhook(
     let mut webhooks = index_scheduler.webhooks_view();
 
     let webhook = webhooks.webhooks.remove(&uuid).ok_or(WebhookNotFound(uuid))?;
-    let webhook = WebhookWithMetadata::from(uuid, webhook);
+    let webhook = WebhookWithMetadataRedactedAuthorization::from(uuid, webhook);
 
     debug!(returns = ?webhook, "Get webhook");
     Ok(HttpResponse::Ok().json(webhook))
@@ -337,7 +338,7 @@ async fn get_webhook(
     request_body = WebhookSettings,
     security(("Bearer" = ["webhooks.create", "webhooks.*", "*"])),
     responses(
-        (status = 201, description = "Webhook created successfully", body = WebhookWithMetadata, content_type = "application/json", example = json!({
+        (status = 201, description = "Webhook created successfully", body = WebhookWithMetadataRedactedAuthorization, content_type = "application/json", example = json!({
             "uuid": "550e8400-e29b-41d4-a716-446655440000",
             "url": "https://your.site/on-tasks-completed",
             "headers": {
@@ -383,7 +384,7 @@ async fn post_webhook(
 
     analytics.publish(PostWebhooksAnalytics, &req);
 
-    let response = WebhookWithMetadata::from(uuid, webhook);
+    let response = WebhookWithMetadataRedactedAuthorization::from(uuid, webhook);
     debug!(returns = ?response, "Post webhook");
     Ok(HttpResponse::Created().json(response))
 }
@@ -395,7 +396,7 @@ async fn post_webhook(
     request_body = WebhookSettings,
     security(("Bearer" = ["webhooks.update", "webhooks.*", "*"])),
     responses(
-        (status = 200, description = "Webhook updated successfully", body = WebhookWithMetadata, content_type = "application/json", example = json!({
+        (status = 200, description = "Webhook updated successfully", body = WebhookWithMetadataRedactedAuthorization, content_type = "application/json", example = json!({
             "uuid": "550e8400-e29b-41d4-a716-446655440000",
             "url": "https://your.site/on-tasks-completed",
             "headers": {
@@ -435,7 +436,7 @@ async fn patch_webhook(
 
     analytics.publish(PatchWebhooksAnalytics, &req);
 
-    let response = WebhookWithMetadata::from(uuid, webhook);
+    let response = WebhookWithMetadataRedactedAuthorization::from(uuid, webhook);
     debug!(returns = ?response, "Patch webhook");
     Ok(HttpResponse::Ok().json(response))
 }

crates/meilisearch/src/search/federated/perform.rs

@@ -17,11 +17,13 @@ use meilisearch_types::milli::vector::Embedding;
 use meilisearch_types::milli::{self, DocumentId, OrderBy, TimeBudget, DEFAULT_VALUES_PER_FACET};
 use roaring::RoaringBitmap;
 use tokio::task::JoinHandle;
+use uuid::Uuid;
 
 use super::super::ranking_rules::{self, RankingRules};
 use super::super::{
     compute_facet_distribution_stats, prepare_search, AttributesFormat, ComputedFacets, HitMaker,
-    HitsInfo, RetrieveVectors, SearchHit, SearchKind, SearchQuery, SearchQueryWithIndex,
+    HitsInfo, RetrieveVectors, SearchHit, SearchKind, SearchMetadata, SearchQuery,
+    SearchQueryWithIndex,
 };
 use super::proxy::{proxy_search, ProxySearchError, ProxySearchParams};
 use super::types::{
@@ -39,32 +41,58 @@ pub async fn perform_federated_search(
     federation: Federation,
     features: RoFeatures,
     is_proxy: bool,
+    request_uid: Uuid,
+    include_metadata: bool,
 ) -> Result<FederatedSearchResult, ResponseError> {
     if is_proxy {
         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);
 
     let network = index_scheduler.network();
 
+    // Preconstruct metadata keeping the original queries order for later metadata building
+    let precomputed_query_metadata: Option<Vec<_>> = include_metadata.then(|| {
+        queries
+            .iter()
+            .map(|q| {
+                (
+                    q.index_uid.to_string(),
+                    q.federation_options.as_ref().and_then(|o| o.remote.clone()),
+                )
+            })
+            .collect()
+    });
+
     // this implementation partition the queries by index to guarantee an important property:
     // - all the queries to a particular index use the same read transaction.
     // This is an important property, otherwise we cannot guarantee the self-consistency of the results.
 
     // 1. partition queries by host and index
     let mut partitioned_queries = PartitionedQueries::new();
+
     for (query_index, federated_query) in queries.into_iter().enumerate() {
         partitioned_queries.partition(federated_query, query_index, &network, features)?
     }
 
     // 2. perform queries, merge and make hits index by index
     // 2.1. start remote queries
-    let remote_search =
-        RemoteSearch::start(partitioned_queries.remote_queries_by_host, &federation, deadline);
+    let remote_search = RemoteSearch::start(
+        partitioned_queries.remote_queries_by_host,
+        &federation,
+        deadline,
+        include_metadata,
+    );
 
     // 2.2. concurrently execute local queries
     let params = SearchByIndexParams {
@@ -106,11 +134,25 @@ pub async fn perform_federated_search(
     let after_waiting_remote_results = std::time::Instant::now();
 
     // 3. merge hits and metadata across indexes and hosts
-    // 3.1. merge metadata
+
+    // 3.1. Build metadata in the same order as the original queries
+    let query_metadata = precomputed_query_metadata.map(|precomputed_query_metadata| {
+        // If a remote is present, set the local remote name
+        let local_remote_name = network.local.clone().filter(|_| partitioned_queries.has_remote);
+
+        build_query_metadata(
+            precomputed_query_metadata,
+            local_remote_name,
+            &remote_results,
+            &results_by_index,
+        )
+    });
+
+    // 3.2. merge federation metadata
     let (estimated_total_hits, degraded, used_negative_operator, facets, max_remote_duration) =
         merge_metadata(&mut results_by_index, &remote_results);
 
-    // 3.2. merge hits
+    // 3.3. merge hits
     let merged_hits: Vec<_> = merge_index_global_results(results_by_index, &mut remote_results)
         .skip(federation.offset)
         .take(federation.limit)
@@ -125,7 +167,7 @@ pub async fn perform_federated_search(
         .map(|hit| hit.hit())
         .collect();
 
-    // 3.3. merge query vectors
+    // 3.4. merge query vectors
     let query_vectors = if retrieve_vectors {
         for remote_results in remote_results.iter_mut() {
             if let Some(remote_vectors) = remote_results.query_vectors.take() {
@@ -144,7 +186,7 @@ pub async fn perform_federated_search(
         None
     };
 
-    // 3.4. merge facets
+    // 3.5. merge facets
     let (facet_distribution, facet_stats, facets_by_index) =
         facet_order.merge(federation.merge_facets, remote_results, facets);
 
@@ -170,6 +212,8 @@ pub async fn perform_federated_search(
         facet_stats,
         facets_by_index,
         remote_errors: partitioned_queries.has_remote.then_some(remote_errors),
+        request_uid: Some(request_uid),
+        metadata: query_metadata,
     })
 }
 
@@ -219,7 +263,7 @@ struct SearchResultByQueryIterItem<'a> {
 
 fn merge_index_local_results(
     results_by_query: Vec<SearchResultByQuery<'_>>,
-) -> impl Iterator<Item = SearchResultByQueryIterItem> + '_ {
+) -> impl Iterator<Item = SearchResultByQueryIterItem<'_>> + '_ {
     itertools::kmerge_by(
         results_by_query.into_iter().map(SearchResultByQueryIter::new),
         |left: &SearchResultByQueryIterItem, right: &SearchResultByQueryIterItem| {
@@ -393,6 +437,7 @@ struct SearchHitByIndex {
 
 struct SearchResultByIndex {
     index: String,
+    primary_key: Option<String>,
     hits: Vec<SearchHitByIndex>,
     estimated_total_hits: usize,
     degraded: bool,
@@ -400,6 +445,61 @@ struct SearchResultByIndex {
     facets: Option<ComputedFacets>,
 }
 
+/// Builds query metadata for federated search results.
+///
+/// This function creates metadata for each query in the same order as the original queries,
+/// combining information from both local and remote search results. It handles the mapping
+/// of primary keys to their respective indexes and remotes to prevent collisions when
+/// multiple remotes have the same index_uid but different primary keys.
+fn build_query_metadata(
+    precomputed_query_metadata: Vec<(String, Option<String>)>,
+    local_remote_name: Option<String>,
+    remote_results: &[FederatedSearchResult],
+    results_by_index: &[SearchResultByIndex],
+) -> Vec<SearchMetadata> {
+    // Create a map of (remote, index_uid) -> primary_key for quick lookup
+    // This prevents collisions when multiple remotes have the same index_uid but different primary keys
+    let mut primary_key_per_index = std::collections::HashMap::new();
+
+    // Build metadata for remote results
+    for remote_result in remote_results {
+        if let Some(remote_metadata) = &remote_result.metadata {
+            for remote_meta in remote_metadata {
+                if let SearchMetadata {
+                    remote: Some(remote_name),
+                    index_uid,
+                    primary_key: Some(primary_key),
+                    ..
+                } = remote_meta
+                {
+                    let key = (Some(remote_name), index_uid);
+                    primary_key_per_index.insert(key, primary_key);
+                }
+            }
+        }
+    }
+
+    // Build metadata for local results
+    for local_meta in results_by_index {
+        if let SearchResultByIndex { index, primary_key: Some(primary_key), .. } = local_meta {
+            let key = (None, index);
+            primary_key_per_index.insert(key, primary_key);
+        }
+    }
+
+    // Build metadata in the same order as the original queries
+    let mut query_metadata = Vec::new();
+    for (index_uid, remote) in precomputed_query_metadata {
+        let primary_key =
+            primary_key_per_index.get(&(remote.as_ref(), &index_uid)).map(|pk| pk.to_string());
+        let query_uid = Uuid::now_v7();
+        // if the remote is not set, use the local remote name
+        let remote = remote.or_else(|| local_remote_name.clone());
+        query_metadata.push(SearchMetadata { query_uid, primary_key, index_uid, remote });
+    }
+    query_metadata
+}
+
 fn merge_metadata(
     results_by_index: &mut Vec<SearchResultByIndex>,
     remote_results: &Vec<FederatedSearchResult>,
@@ -411,6 +511,7 @@ fn merge_metadata(
     let mut max_remote_duration = Duration::ZERO;
     for SearchResultByIndex {
         index,
+        primary_key: _,
         hits: _,
         estimated_total_hits: estimated_total_hits_by_index,
         facets: facets_by_index,
@@ -439,6 +540,8 @@ fn merge_metadata(
         degraded: degraded_for_host,
         used_negative_operator: host_used_negative_operator,
         remote_errors: _,
+        metadata: _,
+        request_uid: _,
     } in remote_results
     {
         let this_remote_duration = Duration::from_millis(*processing_time_ms as u64);
@@ -498,6 +601,10 @@ impl PartitionedQueries {
             .into());
         }
 
+        if federated_query.has_personalize() {
+            return Err(MeilisearchHttpError::PersonalizationInFederatedQuery(query_index).into());
+        }
+
         let (index_uid, query, federation_options) = federated_query.into_index_query_federation();
 
         let federation_options = federation_options.unwrap_or_default();
@@ -566,7 +673,12 @@ struct RemoteSearch {
 }
 
 impl RemoteSearch {
-    fn start(queries: RemoteQueriesByHost, federation: &Federation, deadline: Instant) -> Self {
+    fn start(
+        queries: RemoteQueriesByHost,
+        federation: &Federation,
+        deadline: Instant,
+        include_metadata: bool,
+    ) -> Self {
         let mut in_flight_remote_queries = BTreeMap::new();
         let client = reqwest::ClientBuilder::new()
             .connect_timeout(std::time::Duration::from_millis(200))
@@ -586,7 +698,10 @@ impl RemoteSearch {
                     // never merge distant facets
                     proxy_federation.merge_facets = None;
                     let params = params.clone();
-                    async move { proxy_search(&node, queries, proxy_federation, &params).await }
+                    async move {
+                        proxy_search(&node, queries, proxy_federation, &params, include_metadata)
+                            .await
+                    }
                 }),
             );
         }
@@ -630,6 +745,13 @@ impl RemoteSearch {
                             continue 'remote_queries;
                         }
 
+                        // Add remote name to metadata
+                        if let Some(metadata) = res.metadata.as_mut() {
+                            for meta in metadata {
+                                meta.remote = Some(node_name.clone());
+                            }
+                        }
+
                         federation.insert(
                             FEDERATION_REMOTE.to_string(),
                             serde_json::Value::String(node_name.clone()),
@@ -725,6 +847,7 @@ impl SearchByIndex {
             }
         };
         let rtxn = index.read_txn()?;
+        let primary_key = index.primary_key(&rtxn)?.map(|pk| pk.to_string());
         let criteria = index.criteria(&rtxn)?;
         let dictionary = index.dictionary(&rtxn)?;
         let dictionary: Option<Vec<_>> =
@@ -751,6 +874,12 @@ impl SearchByIndex {
             return Err(error);
         }
         let mut results_by_query = Vec::with_capacity(queries.len());
+
+        // all queries for an index share the same budget
+        let time_budget = match cutoff {
+            Some(cutoff) => TimeBudget::new(Duration::from_millis(cutoff)),
+            None => TimeBudget::default(),
+        };
         for QueryByIndex { query, weight, query_index } in queries {
             // use an immediately invoked lambda to capture the result without returning from the function
 
@@ -820,17 +949,13 @@ impl SearchByIndex {
 
                 let retrieve_vectors = RetrieveVectors::new(query.retrieve_vectors);
 
-                let time_budget = match cutoff {
-                    Some(cutoff) => TimeBudget::new(Duration::from_millis(cutoff)),
-                    None => TimeBudget::default(),
-                };
-
                 let (mut search, _is_finite_pagination, _max_total_hits, _offset) = prepare_search(
                     &index,
                     &rtxn,
                     &query,
                     &search_kind,
-                    time_budget,
+                    // clones of `TimeBudget` share the budget rather than restart it
+                    time_budget.clone(),
                     params.features,
                 )?;
 
@@ -977,6 +1102,7 @@ impl SearchByIndex {
             })?;
         self.results_by_index.push(SearchResultByIndex {
             index: index_uid,
+            primary_key,
             hits: merged_result,
             estimated_total_hits,
             degraded,

crates/meilisearch/src/search/federated/proxy.rs

@@ -7,7 +7,7 @@ use serde::de::DeserializeOwned;
 use serde_json::Value;
 
 use super::types::{FederatedSearch, FederatedSearchResult, Federation};
-use crate::search::SearchQueryWithIndex;
+use crate::search::{SearchQueryWithIndex, INCLUDE_METADATA_HEADER};
 
 pub const PROXY_SEARCH_HEADER: &str = "Meili-Proxy-Search";
 pub const PROXY_SEARCH_HEADER_VALUE: &str = "true";
@@ -98,6 +98,7 @@ pub async fn proxy_search(
     queries: Vec<SearchQueryWithIndex>,
     federation: Federation,
     params: &ProxySearchParams,
+    include_metadata: bool,
 ) -> Result<FederatedSearchResult, ProxySearchError> {
     let url = format!("{}/multi-search", node.url);
 
@@ -105,7 +106,12 @@ 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)
@@ -114,7 +120,16 @@ pub async fn proxy_search(
     };
 
     for i in 0..params.try_count {
-        match try_proxy_search(&url, search_api_key, &federated, &params.client, deadline).await {
+        match try_proxy_search(
+            &url,
+            search_api_key,
+            &federated,
+            &params.client,
+            deadline,
+            include_metadata,
+        )
+        .await
+        {
             Ok(response) => return Ok(response),
             Err(retry) => {
                 let duration = retry.into_duration(i)?;
@@ -122,7 +137,7 @@ pub async fn proxy_search(
             }
         }
     }
-    try_proxy_search(&url, search_api_key, &federated, &params.client, deadline)
+    try_proxy_search(&url, search_api_key, &federated, &params.client, deadline, include_metadata)
         .await
         .map_err(Retry::into_error)
 }
@@ -133,6 +148,7 @@ async fn try_proxy_search(
     federated: &FederatedSearch,
     client: &Client,
     deadline: std::time::Instant,
+    include_metadata: bool,
 ) -> Result<FederatedSearchResult, Retry> {
     let timeout = deadline.saturating_duration_since(std::time::Instant::now());
 
@@ -143,6 +159,8 @@ async fn try_proxy_search(
         request
     };
     let request = request.header(PROXY_SEARCH_HEADER, PROXY_SEARCH_HEADER_VALUE);
+    let request =
+        if include_metadata { request.header(INCLUDE_METADATA_HEADER, "true") } else { request };
 
     let response = request.send().await;
     let response = match response {

crates/meilisearch/src/search/federated/types.rs

@@ -16,6 +16,9 @@ use meilisearch_types::milli::order_by_map::OrderByMap;
 use meilisearch_types::milli::OrderBy;
 use serde::{Deserialize, Serialize};
 use utoipa::ToSchema;
+use uuid::Uuid;
+
+use crate::search::SearchMetadata;
 
 use super::super::{ComputedFacets, FacetStats, HitsInfo, SearchHit, SearchQueryWithIndex};
 use crate::milli::vector::Embedding;
@@ -131,6 +134,10 @@ pub struct FederatedSearchResult {
     pub facet_stats: Option<BTreeMap<String, FacetStats>>,
     #[serde(default, skip_serializing_if = "FederatedFacets::is_empty")]
     pub facets_by_index: FederatedFacets,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub request_uid: Option<Uuid>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub metadata: Option<Vec<SearchMetadata>>,
 
     #[serde(default, skip_serializing_if = "Option::is_none")]
     pub remote_errors: Option<BTreeMap<String, ResponseError>>,
@@ -156,6 +163,8 @@ impl fmt::Debug for FederatedSearchResult {
             facet_stats,
             facets_by_index,
             remote_errors,
+            request_uid,
+            metadata,
         } = self;
 
         let mut debug = f.debug_struct("SearchResult");
@@ -188,6 +197,12 @@ impl fmt::Debug for FederatedSearchResult {
         if let Some(remote_errors) = remote_errors {
             debug.field("remote_errors", &remote_errors);
         }
+        if let Some(request_uid) = request_uid {
+            debug.field("request_uid", &request_uid);
+        }
+        if let Some(metadata) = metadata {
+            debug.field("metadata", &metadata);
+        }
 
         debug.finish()
     }

crates/meilisearch/src/search/mod.rs

@@ -36,6 +36,7 @@ use serde_json::{json, Value};
 #[cfg(test)]
 mod mod_test;
 use utoipa::ToSchema;
+use uuid::Uuid;
 
 use crate::error::MeilisearchHttpError;
 
@@ -56,6 +57,14 @@ pub const DEFAULT_CROP_MARKER: fn() -> String = || "…".to_string();
 pub const DEFAULT_HIGHLIGHT_PRE_TAG: fn() -> String = || "<em>".to_string();
 pub const DEFAULT_HIGHLIGHT_POST_TAG: fn() -> String = || "</em>".to_string();
 pub const DEFAULT_SEMANTIC_RATIO: fn() -> SemanticRatio = || SemanticRatio(0.5);
+pub const INCLUDE_METADATA_HEADER: &str = "Meili-Include-Metadata";
+
+#[derive(Clone, Default, PartialEq, Deserr, ToSchema, Debug)]
+#[deserr(error = DeserrJsonError<InvalidSearchPersonalize>, rename_all = camelCase, deny_unknown_fields)]
+pub struct Personalize {
+    #[deserr(error = DeserrJsonError<InvalidSearchPersonalizeUserContext>)]
+    pub user_context: String,
+}
 
 #[derive(Clone, Default, PartialEq, Deserr, ToSchema)]
 #[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
@@ -120,6 +129,8 @@ pub struct SearchQuery {
     pub ranking_score_threshold: Option<RankingScoreThreshold>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchLocales>)]
     pub locales: Option<Vec<Locale>>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchPersonalize>, default)]
+    pub personalize: Option<Personalize>,
 }
 
 impl From<SearchParameters> for SearchQuery {
@@ -167,6 +178,7 @@ impl From<SearchParameters> for SearchQuery {
             highlight_post_tag: DEFAULT_HIGHLIGHT_POST_TAG(),
             crop_marker: DEFAULT_CROP_MARKER(),
             locales: None,
+            personalize: None,
         }
     }
 }
@@ -248,6 +260,7 @@ impl fmt::Debug for SearchQuery {
             attributes_to_search_on,
             ranking_score_threshold,
             locales,
+            personalize,
         } = self;
 
         let mut debug = f.debug_struct("SearchQuery");
@@ -336,6 +349,10 @@ impl fmt::Debug for SearchQuery {
             debug.field("locales", &locales);
         }
 
+        if let Some(personalize) = personalize {
+            debug.field("personalize", &personalize);
+        }
+
         debug.finish()
     }
 }
@@ -541,6 +558,9 @@ pub struct SearchQueryWithIndex {
     pub ranking_score_threshold: Option<RankingScoreThreshold>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchLocales>, default)]
     pub locales: Option<Vec<Locale>>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchPersonalize>, default)]
+    #[serde(skip)]
+    pub personalize: Option<Personalize>,
 
     #[deserr(default)]
     pub federation_options: Option<FederationOptions>,
@@ -565,6 +585,10 @@ impl SearchQueryWithIndex {
         self.facets.as_deref().filter(|v| !v.is_empty())
     }
 
+    pub fn has_personalize(&self) -> bool {
+        self.personalize.is_some()
+    }
+
     pub fn from_index_query_federation(
         index_uid: IndexUid,
         query: SearchQuery,
@@ -598,6 +622,7 @@ impl SearchQueryWithIndex {
             attributes_to_search_on,
             ranking_score_threshold,
             locales,
+            personalize,
         } = query;
 
         SearchQueryWithIndex {
@@ -629,6 +654,7 @@ impl SearchQueryWithIndex {
             attributes_to_search_on,
             ranking_score_threshold,
             locales,
+            personalize,
             federation_options,
         }
     }
@@ -664,6 +690,7 @@ impl SearchQueryWithIndex {
             hybrid,
             ranking_score_threshold,
             locales,
+            personalize,
         } = self;
         (
             index_uid,
@@ -695,6 +722,7 @@ impl SearchQueryWithIndex {
                 hybrid,
                 ranking_score_threshold,
                 locales,
+                personalize,
                 // do not use ..Default::default() here,
                 // rather add any missing field from `SearchQuery` to `SearchQueryWithIndex`
             },
@@ -835,6 +863,18 @@ pub struct SearchHit {
     pub ranking_score_details: Option<serde_json::Map<String, serde_json::Value>>,
 }
 
+#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, ToSchema)]
+#[serde(rename_all = "camelCase")]
+#[schema(rename_all = "camelCase")]
+pub struct SearchMetadata {
+    pub query_uid: Uuid,
+    pub index_uid: String,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub primary_key: Option<String>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub remote: Option<String>,
+}
+
 #[derive(Serialize, Clone, PartialEq, ToSchema)]
 #[serde(rename_all = "camelCase")]
 #[schema(rename_all = "camelCase")]
@@ -851,6 +891,10 @@ pub struct SearchResult {
     pub facet_distribution: Option<BTreeMap<String, IndexMap<String, u64>>>,
     #[serde(skip_serializing_if = "Option::is_none")]
     pub facet_stats: Option<BTreeMap<String, FacetStats>>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub request_uid: Option<Uuid>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub metadata: Option<SearchMetadata>,
 
     #[serde(skip_serializing_if = "Option::is_none")]
     pub semantic_hit_count: Option<u32>,
@@ -872,6 +916,8 @@ impl fmt::Debug for SearchResult {
             hits_info,
             facet_distribution,
             facet_stats,
+            request_uid,
+            metadata,
             semantic_hit_count,
             degraded,
             used_negative_operator,
@@ -901,6 +947,12 @@ impl fmt::Debug for SearchResult {
         if let Some(semantic_hit_count) = semantic_hit_count {
             debug.field("semantic_hit_count", &semantic_hit_count);
         }
+        if let Some(request_uid) = request_uid {
+            debug.field("request_uid", &request_uid);
+        }
+        if let Some(metadata) = metadata {
+            debug.field("metadata", &metadata);
+        }
 
         debug.finish()
     }
@@ -1113,15 +1165,31 @@ pub fn prepare_search<'t>(
     Ok((search, is_finite_pagination, max_total_hits, offset))
 }
 
+pub struct SearchParams {
+    pub index_uid: String,
+    pub query: SearchQuery,
+    pub search_kind: SearchKind,
+    pub retrieve_vectors: RetrieveVectors,
+    pub features: RoFeatures,
+    pub request_uid: Uuid,
+    pub include_metadata: bool,
+}
+
 pub fn perform_search(
-    index_uid: String,
+    params: SearchParams,
     index: &Index,
-    query: SearchQuery,
-    search_kind: SearchKind,
-    retrieve_vectors: RetrieveVectors,
-    features: RoFeatures,
-) -> Result<SearchResult, ResponseError> {
+) -> Result<(SearchResult, TimeBudget), ResponseError> {
+    let SearchParams {
+        index_uid,
+        query,
+        search_kind,
+        retrieve_vectors,
+        features,
+        request_uid,
+        include_metadata,
+    } = params;
     let before_search = Instant::now();
+    let index_uid_for_metadata = index_uid.clone();
     let rtxn = index.read_txn()?;
     let time_budget = match index.search_cutoff(&rtxn)? {
         Some(cutoff) => TimeBudget::new(Duration::from_millis(cutoff)),
@@ -1129,7 +1197,7 @@ pub fn perform_search(
     };
 
     let (search, is_finite_pagination, max_total_hits, offset) =
-        prepare_search(index, &rtxn, &query, &search_kind, time_budget, features)?;
+        prepare_search(index, &rtxn, &query, &search_kind, time_budget.clone(), features)?;
 
     let (
         milli::SearchResult {
@@ -1142,7 +1210,20 @@ pub fn perform_search(
             query_vector,
         },
         semantic_hit_count,
-    ) = search_from_kind(index_uid, search_kind, search)?;
+    ) = search_from_kind(index_uid.clone(), search_kind, search)?;
+
+    let metadata = if include_metadata {
+        let query_uid = Uuid::now_v7();
+        let primary_key = index.primary_key(&rtxn)?.map(|pk| pk.to_string());
+        Some(SearchMetadata {
+            query_uid,
+            index_uid: index_uid_for_metadata,
+            primary_key,
+            remote: None, // Local searches don't have a remote
+        })
+    } else {
+        None
+    };
 
     let SearchQuery {
         q,
@@ -1174,6 +1255,7 @@ pub fn perform_search(
         attributes_to_search_on: _,
         filter: _,
         distinct: _,
+        personalize: _,
     } = query;
 
     let format = AttributesFormat {
@@ -1225,7 +1307,6 @@ pub fn perform_search(
         .transpose()?
         .map(|ComputedFacets { distribution, stats }| (distribution, stats))
         .unzip();
-
     let result = SearchResult {
         hits: documents,
         hits_info,
@@ -1237,8 +1318,10 @@ pub fn perform_search(
         degraded,
         used_negative_operator,
         semantic_hit_count,
+        request_uid: Some(request_uid),
+        metadata,
     };
-    Ok(result)
+    Ok((result, time_budget))
 }
 
 #[derive(Debug, Clone, Default, Serialize, Deserialize, ToSchema)]
@@ -2080,7 +2163,7 @@ pub(crate) fn parse_filter(
     facets: &Value,
     filter_parsing_error_code: Code,
     features: RoFeatures,
-) -> Result<Option<Filter>, ResponseError> {
+) -> Result<Option<Filter<'_>>, ResponseError> {
     let filter = match facets {
         Value::String(expr) => Filter::from_str(expr).map_err(|e| e.into()),
         Value::Array(arr) => parse_filter_array(arr).map_err(|e| e.into()),
@@ -2117,7 +2200,7 @@ pub(crate) fn parse_filter(
     Ok(filter)
 }
 
-fn parse_filter_array(arr: &[Value]) -> Result<Option<Filter>, MeilisearchHttpError> {
+fn parse_filter_array(arr: &'_ [Value]) -> Result<Option<Filter<'_>>, MeilisearchHttpError> {
     let mut ands = Vec::new();
     for value in arr {
         match value {

crates/meilisearch/src/search_queue.rs

@@ -13,9 +13,9 @@
 //! What is going to happen at this point is that you're going to send a oneshot::Sender over an async mpsc channel.
 //! Then, the queue/scheduler is going to either:
 //! - Drop your oneshot channel => that means there are too many searches going on, and yours won't be executed.
-//!                                You should exit and free all the RAM you use ASAP.
+//!   You should exit and free all the RAM you use ASAP.
 //! - Sends you a Permit => that will unlock the method, and you will be able to process your search.
-//!                         And should drop the Permit only once you have freed all the RAM consumed by the method.
+//!   And should drop the Permit only once you have freed all the RAM consumed by the method.
 
 use std::num::NonZeroUsize;
 use std::sync::atomic::{AtomicUsize, Ordering};

crates/meilisearch/tests/auth/api_keys.rs

@@ -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]

crates/meilisearch/tests/auth/errors.rs

@@ -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]

crates/meilisearch/tests/auth/tenant_token_multi_search.rs

@@ -1040,7 +1040,7 @@ async fn error_single_search_forbidden_token() {
     ];
 
     let failed_query_indexes: Vec<_> =
-        std::iter::repeat(Some(0)).take(5).chain(std::iter::repeat(None).take(6)).collect();
+        std::iter::repeat_n(Some(0), 5).chain(std::iter::repeat_n(None, 6)).collect();
 
     let failed_query_indexes = vec![failed_query_indexes; ACCEPTED_KEYS_SINGLE.len()];
 
@@ -1118,10 +1118,9 @@ async fn error_multi_search_forbidden_token() {
         },
     ];
 
-    let failed_query_indexes: Vec<_> = std::iter::repeat(Some(0))
-        .take(5)
-        .chain(std::iter::repeat(Some(1)).take(5))
-        .chain(std::iter::repeat(None).take(6))
+    let failed_query_indexes: Vec<_> = std::iter::repeat_n(Some(0), 5)
+        .chain(std::iter::repeat_n(Some(1), 5))
+        .chain(std::iter::repeat_n(None, 6))
         .collect();
 
     let failed_query_indexes = vec![failed_query_indexes; ACCEPTED_KEYS_BOTH.len()];

crates/meilisearch/tests/batches/errors.rs

@@ -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]

crates/meilisearch/tests/common/index.rs

@@ -91,7 +91,16 @@ impl<'a> Index<'a, Owned> {
         documents: Value,
         primary_key: Option<&str>,
     ) -> (Value, StatusCode) {
-        self._add_documents(documents, primary_key).await
+        self._add_documents(documents, primary_key, None).await
+    }
+
+    pub async fn add_documents_with_custom_metadata(
+        &self,
+        documents: Value,
+        primary_key: Option<&str>,
+        custom_metadata: Option<&str>,
+    ) -> (Value, StatusCode) {
+        self._add_documents(documents, primary_key, custom_metadata).await
     }
 
     pub async fn raw_add_documents(
@@ -352,12 +361,25 @@ impl<State> Index<'_, State> {
         &self,
         documents: Value,
         primary_key: Option<&str>,
+        custom_metadata: Option<&str>,
     ) -> (Value, StatusCode) {
-        let url = match primary_key {
-            Some(key) => {
-                format!("/indexes/{}/documents?primaryKey={}", urlencode(self.uid.as_ref()), key)
+        let url = match (primary_key, custom_metadata) {
+            (Some(key), Some(meta)) => {
+                format!(
+                    "/indexes/{}/documents?primaryKey={key}&customMetadata={meta}",
+                    urlencode(self.uid.as_ref()),
+                )
             }
-            None => format!("/indexes/{}/documents", urlencode(self.uid.as_ref())),
+            (None, Some(meta)) => {
+                format!(
+                    "/indexes/{}/documents?&customMetadata={meta}",
+                    urlencode(self.uid.as_ref()),
+                )
+            }
+            (Some(key), None) => {
+                format!("/indexes/{}/documents?&primaryKey={key}", urlencode(self.uid.as_ref()),)
+            }
+            (None, None) => format!("/indexes/{}/documents", urlencode(self.uid.as_ref())),
         };
         self.service.post_encoded(url, documents, self.encoder).await
     }
@@ -516,6 +538,18 @@ impl<State> Index<'_, State> {
         self.service.post_encoded(url, query, self.encoder).await
     }
 
+    pub async fn search_with_headers(
+        &self,
+        query: Value,
+        headers: Vec<(&str, &str)>,
+    ) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/search", urlencode(self.uid.as_ref()));
+        let body = serde_json::to_string(&query).unwrap();
+        let mut all_headers = vec![("content-type", "application/json")];
+        all_headers.extend(headers);
+        self.service.post_str(url, body, all_headers).await
+    }
+
     pub async fn search_get(&self, query: &str) -> (Value, StatusCode) {
         let url = format!("/indexes/{}/search{}", urlencode(self.uid.as_ref()), query);
         self.service.get(url).await

crates/meilisearch/tests/common/mod.rs

@@ -241,7 +241,7 @@ pub async fn shared_index_with_documents() -> &'static Index<'static, Shared> {
         let server = Server::new_shared();
         let index = server._index("SHARED_DOCUMENTS").to_shared();
         let documents = DOCUMENTS.clone();
-        let (response, _code) = index._add_documents(documents, None).await;
+        let (response, _code) = index._add_documents(documents, None, None).await;
         server.wait_task(response.uid()).await.succeeded();
         let (response, _code) = index
             ._update_settings(
@@ -284,7 +284,7 @@ pub async fn shared_index_with_score_documents() -> &'static Index<'static, Shar
         let server = Server::new_shared();
         let index = server._index("SHARED_SCORE_DOCUMENTS").to_shared();
         let documents = SCORE_DOCUMENTS.clone();
-        let (response, _code) = index._add_documents(documents, None).await;
+        let (response, _code) = index._add_documents(documents, None, None).await;
         server.wait_task(response.uid()).await.succeeded();
         let (response, _code) = index
             ._update_settings(
@@ -361,7 +361,7 @@ pub async fn shared_index_with_nested_documents() -> &'static Index<'static, Sha
         let server = Server::new_shared();
         let index = server._index("SHARED_NESTED_DOCUMENTS").to_shared();
         let documents = NESTED_DOCUMENTS.clone();
-        let (response, _code) = index._add_documents(documents, None).await;
+        let (response, _code) = index._add_documents(documents, None, None).await;
         server.wait_task(response.uid()).await.succeeded();
         let (response, _code) = index
             ._update_settings(
@@ -508,7 +508,7 @@ pub async fn shared_index_with_geo_documents() -> &'static Index<'static, Shared
         .get_or_init(|| async {
             let server = Server::new_shared();
             let index = server._index("SHARED_GEO_DOCUMENTS").to_shared();
-            let (response, _code) = index._add_documents(GEO_DOCUMENTS.clone(), None).await;
+            let (response, _code) = index._add_documents(GEO_DOCUMENTS.clone(), None, None).await;
             server.wait_task(response.uid()).await.succeeded();
 
             let (response, _code) = index
@@ -522,6 +522,26 @@ pub async fn shared_index_with_geo_documents() -> &'static Index<'static, Shared
         .await
 }
 
+pub async fn shared_index_geojson_documents() -> &'static Index<'static, Shared> {
+    static INDEX: OnceCell<Index<'static, Shared>> = OnceCell::const_new();
+    INDEX
+        .get_or_init(|| async {
+            // Retrieved from https://gitlab-forge.din.developpement-durable.gouv.fr/pub/geomatique/descartes/d-map/-/blob/main/demo/examples/commons/countries.geojson?ref_type=heads
+            let server = Server::new_shared();
+            let index = server._index("SHARED_GEOJSON_DOCUMENTS").to_shared();
+            let countries = include_str!("../documents/geojson/assets/countries.json");
+            let lille = serde_json::from_str::<serde_json::Value>(countries).unwrap();
+            let (response, _code) = index._add_documents(Value(lille), Some("name"), None).await;
+            server.wait_task(response.uid()).await.succeeded();
+
+            let (response, _code) =
+                index._update_settings(json!({"filterableAttributes": ["_geojson"]})).await;
+            server.wait_task(response.uid()).await.succeeded();
+            index
+        })
+        .await
+}
+
 pub async fn shared_index_for_fragments() -> Index<'static, Shared> {
     static INDEX: OnceCell<(Server<Shared>, String)> = OnceCell::const_new();
     let (server, uid) = INDEX

crates/meilisearch/tests/common/server.rs

@@ -49,8 +49,8 @@ impl Server<Owned> {
         }
 
         let options = default_settings(dir.path());
-
-        let (index_scheduler, auth) = setup_meilisearch(&options).unwrap();
+        let handle = tokio::runtime::Handle::current();
+        let (index_scheduler, auth) = setup_meilisearch(&options, handle).unwrap();
         let service = Service { index_scheduler, auth, options, api_key: None };
 
         Server { service, _dir: Some(dir), _marker: PhantomData }
@@ -65,7 +65,9 @@ impl Server<Owned> {
 
         options.master_key = Some("MASTER_KEY".to_string());
 
-        let (index_scheduler, auth) = setup_meilisearch(&options).unwrap();
+        let handle = tokio::runtime::Handle::current();
+
+        let (index_scheduler, auth) = setup_meilisearch(&options, handle).unwrap();
         let service = Service { index_scheduler, auth, options, api_key: None };
 
         Server { service, _dir: Some(dir), _marker: PhantomData }
@@ -78,7 +80,9 @@ impl Server<Owned> {
     }
 
     pub async fn new_with_options(options: Opt) -> Result<Self, anyhow::Error> {
-        let (index_scheduler, auth) = setup_meilisearch(&options)?;
+        let handle = tokio::runtime::Handle::current();
+
+        let (index_scheduler, auth) = setup_meilisearch(&options, handle)?;
         let service = Service { index_scheduler, auth, options, api_key: None };
 
         Ok(Server { service, _dir: None, _marker: PhantomData })
@@ -217,8 +221,9 @@ impl Server<Shared> {
         }
 
         let options = default_settings(dir.path());
+        let handle = tokio::runtime::Handle::current();
 
-        let (index_scheduler, auth) = setup_meilisearch(&options).unwrap();
+        let (index_scheduler, auth) = setup_meilisearch(&options, handle).unwrap();
         let service = Service { index_scheduler, auth, api_key: None, options };
 
         Server { service, _dir: Some(dir), _marker: PhantomData }
@@ -390,6 +395,17 @@ impl<State> Server<State> {
         self.service.post("/multi-search", queries).await
     }
 
+    pub async fn multi_search_with_headers(
+        &self,
+        queries: Value,
+        headers: Vec<(&str, &str)>,
+    ) -> (Value, StatusCode) {
+        let body = serde_json::to_string(&queries).unwrap();
+        let mut all_headers = vec![("content-type", "application/json")];
+        all_headers.extend(headers);
+        self.service.post_str("/multi-search", body, all_headers).await
+    }
+
     pub async fn list_indexes_raw(&self, parameters: &str) -> (Value, StatusCode) {
         self.service.get(format!("/indexes{parameters}")).await
     }
@@ -490,6 +506,8 @@ pub fn default_settings(dir: impl AsRef<Path>) -> Opt {
             max_indexing_threads: MaxThreads::from_str("2").unwrap(),
             experimental_no_edition_2024_for_settings: false,
             experimental_no_edition_2024_for_dumps: false,
+            experimental_no_edition_2024_for_prefix_post_processing: false,
+            experimental_no_edition_2024_for_facet_post_processing: false,
         },
         experimental_enable_metrics: false,
         ..Parser::parse_from(None as Option<&str>)

crates/meilisearch/tests/common/service.rs

@@ -10,8 +10,9 @@ use actix_web::test::TestRequest;
 use actix_web::web::Data;
 use index_scheduler::IndexScheduler;
 use meilisearch::analytics::Analytics;
+use meilisearch::personalization::PersonalizationService;
 use meilisearch::search_queue::SearchQueue;
-use meilisearch::{create_app, Opt, SubscriberForSecondLayer};
+use meilisearch::{create_app, Opt, ServicesData, SubscriberForSecondLayer};
 use meilisearch_auth::AuthController;
 use tracing::level_filters::LevelFilter;
 use tracing_subscriber::Layer;
@@ -135,14 +136,24 @@ impl Service {
             self.options.experimental_search_queue_size,
             NonZeroUsize::new(1).unwrap(),
         );
+        let personalization_service = self
+            .options
+            .experimental_personalization_api_key
+            .clone()
+            .map(PersonalizationService::cohere)
+            .unwrap_or_else(PersonalizationService::disabled);
 
         actix_web::test::init_service(create_app(
-            self.index_scheduler.clone().into(),
-            self.auth.clone().into(),
-            Data::new(search_queue),
+            ServicesData {
+                index_scheduler: self.index_scheduler.clone().into(),
+                auth: self.auth.clone().into(),
+                search_queue: Data::new(search_queue),
+                personalization_service: Data::new(personalization_service),
+                logs_route_handle: Data::new(route_layer_handle),
+                logs_stderr_handle: Data::new(stderr_layer_handle),
+                analytics: Data::new(Analytics::no_analytics()),
+            },
             self.options.clone(),
-            (route_layer_handle, stderr_layer_handle),
-            Data::new(Analytics::no_analytics()),
             true,
         ))
         .await