Fix compact openapi generation

Add support for XLM Roberta models

.github/ISSUE_TEMPLATE/new_feature_issue.md

@@ -24,6 +24,11 @@ TBD
 - [ ] If not, add the `no db change` label to your PR, and you're good to merge.
 - [ ] If yes, add the `db change` label to your PR. You'll receive a message explaining you what to do.
 
+### Reminders when adding features
+
+- [ ] Write unit tests using insta
+- [ ] Write declarative integration tests in [workloads/tests](https://github.com/meilisearch/meilisearch/tree/main/workloads/test). Specify the routes to call and then call `cargo xtask test workloads/tests/YOUR_TEST.json --update-responses` so that responses are automatically filled.
+
 ### Reminders when modifying the API
 
 - [ ] Update the openAPI file with utoipa:

.github/release-draft-template.yml

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

.github/workflows/bench-pr.yml

@@ -67,8 +67,6 @@ jobs:
           ref: ${{ steps.comment-branch.outputs.head_ref }}
 
       - uses: dtolnay/rust-toolchain@1.89
-        with:
-          profile: minimal
 
       - name: Run benchmarks on PR ${{ github.event.issue.id }}
         run: |

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

@@ -13,8 +13,6 @@ jobs:
     steps:
       - uses: actions/checkout@v5
       - uses: dtolnay/rust-toolchain@1.89
-        with:
-          profile: minimal
 
       # Run benchmarks
       - name: Run benchmarks - Dataset ${BENCH_NAME} - Branch main - Commit ${{ github.sha }}

.github/workflows/db-change-comments.yml

@@ -6,7 +6,7 @@ on:
 
 env:
   MESSAGE: |
-    ### Hello, I'm a bot 🤖 
+    ### Hello, I'm a bot 🤖
 
     You are receiving this message because you declared that this PR make changes to the Meilisearch database.
     Depending on the nature of the change, additional actions might be required on your part. The following sections detail the additional actions depending on the nature of the change, please copy the relevant section in the description of your PR, and make sure to perform the required actions.
@@ -19,6 +19,7 @@ env:
 
     - [ ] Detail the change to the DB format and why they are forward compatible
     - [ ] Forward-compatibility: A database created before this PR and using the features touched by this PR was able to be opened by a Meilisearch produced by the code of this PR.
+    - [ ] Declarative test: add a [declarative test containing a dumpless upgrade](https://github.com/meilisearch/meilisearch/blob/main/TESTING.md#typical-usage)
 
 
     ## This PR makes breaking changes
@@ -35,8 +36,7 @@ env:
       - [ ] Write the code to go from the old database to the new one
         - If the change happened in milli, the upgrade function should be written and called [here](https://github.com/meilisearch/meilisearch/blob/3fd86e8d76d7d468b0095d679adb09211ca3b6c0/crates/milli/src/update/upgrade/mod.rs#L24-L47)
         - If the change happened in the index-scheduler, we've never done it yet, but the right place to do it should be [here](https://github.com/meilisearch/meilisearch/blob/3fd86e8d76d7d468b0095d679adb09211ca3b6c0/crates/index-scheduler/src/scheduler/process_upgrade/mod.rs#L13)
-      - [ ] Write an integration test [here](https://github.com/meilisearch/meilisearch/blob/main/crates/meilisearch/tests/upgrade/mod.rs) ensuring you can read the old database, upgrade to the new database, and read the new database as expected
-    
+      - [ ] Declarative test: add a [declarative test containing a dumpless upgrade](https://github.com/meilisearch/meilisearch/blob/main/TESTING.md#typical-usage)
 
 jobs:
   add-comment:

.github/workflows/flaky-tests.yml

@@ -13,6 +13,12 @@ jobs:
       image: ubuntu:22.04
     steps:
       - uses: actions/checkout@v5
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
+        run: |
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - name: Install needed dependencies
         run: |
           apt-get update && apt-get install -y curl

.github/workflows/fuzzer-indexing.yml

@@ -13,8 +13,6 @@ jobs:
     steps:
       - uses: actions/checkout@v5
       - uses: dtolnay/rust-toolchain@1.89
-        with:
-          profile: minimal
 
       # Run benchmarks
       - name: Run the fuzzer

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

@@ -25,6 +25,12 @@ jobs:
         run: |
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
+        run: |
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - uses: dtolnay/rust-toolchain@1.89
       - name: Install cargo-deb
         run: cargo install cargo-deb

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

@@ -14,10 +14,105 @@ on:
   workflow_dispatch:
 
 jobs:
-  docker:
-    runs-on: docker
+  build:
+    runs-on: ${{ matrix.runner }}
+
+    strategy:
+      matrix:
+        platform: [amd64, arm64]
+        edition: [community, enterprise]
+        include:
+          - platform: amd64
+            runner: ubuntu-24.04
+          - platform: arm64
+            runner: ubuntu-24.04-arm
+          - edition: community
+            registry: getmeili/meilisearch
+            feature-flag: ""
+          - edition: enterprise
+            registry: getmeili/meilisearch-enterprise
+            feature-flag: "--features enterprise"
+
+    permissions: {}
+    steps:
+      - uses: actions/checkout@v5
+
+      - name: Prepare
+        run: |
+          platform=linux/${{ matrix.platform }}
+          echo "PLATFORM_PAIR=${platform//\//-}" >> $GITHUB_ENV
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+        with:
+          platforms: linux/${{ matrix.platform }}
+          install: true
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Docker meta
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ matrix.registry }}
+          # Prevent `latest` to be updated for each new tag pushed.
+          # We need latest and `vX.Y` tags to only be pushed for the stable Meilisearch releases.
+          flavor: latest=false
+          tags: |
+            type=ref,event=tag
+            type=raw,value=nightly,enable=${{ github.event_name != 'push' }}
+            type=semver,pattern=v{{major}}.{{minor}},enable=${{ steps.check-tag-format.outputs.stable == 'true' }}
+            type=semver,pattern=v{{major}},enable=${{ steps.check-tag-format.outputs.stable == 'true' }}
+            type=raw,value=latest,enable=${{ steps.check-tag-format.outputs.stable == 'true' && steps.check-tag-format.outputs.latest == 'true' }}
+
+      - name: Build and push by digest
+        uses: docker/build-push-action@v6
+        id: build-and-push
+        with:
+          platforms: linux/${{ matrix.platform }}
+          labels: ${{ steps.meta.outputs.labels }}
+          tags: ${{ matrix.registry }}
+          outputs: type=image,push-by-digest=true,name-canonical=true,push=true
+          build-args: |
+            COMMIT_SHA=${{ github.sha }}
+            COMMIT_DATE=${{ steps.build-metadata.outputs.date }}
+            GIT_TAG=${{ github.ref_name }}
+            EXTRA_ARGS=${{ matrix.feature-flag }}
+
+      - name: Export digest
+        run: |
+          mkdir -p ${{ runner.temp }}/digests
+          digest="${{ steps.build-and-push.outputs.digest }}"
+          touch "${{ runner.temp }}/digests/${digest#sha256:}"
+
+      - name: Upload digest
+        uses: actions/upload-artifact@v4
+        with:
+          name: digests-${{ matrix.edition }}-${{ env.PLATFORM_PAIR }}
+          path: ${{ runner.temp }}/digests/*
+          if-no-files-found: error
+          retention-days: 1
+
+  merge:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        edition: [community, enterprise]
+        include:
+          - edition: community
+            registry: getmeili/meilisearch
+          - edition: enterprise
+            registry: getmeili/meilisearch-enterprise
+    needs:
+      - build
+
     permissions:
       id-token: write # This is needed to use Cosign in keyless mode
+
     steps:
       - uses: actions/checkout@v5
 
@@ -58,26 +153,30 @@ jobs:
 
           echo "date=$commit_date" >> $GITHUB_OUTPUT
 
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
       - name: Install cosign
         uses: sigstore/cosign-installer@d7543c93d881b35a8faa02e8e3605f69b7a1ce62 # tag=v3.10.0
 
+      - name: Download digests
+        uses: actions/download-artifact@v4
+        with:
+          path: ${{ runner.temp }}/digests
+          pattern: digests-${{ matrix.edition }}-*
+          merge-multiple: true
+
       - name: Login to Docker Hub
         uses: docker/login-action@v3
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
 
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
       - name: Docker meta
         id: meta
         uses: docker/metadata-action@v5
         with:
-          images: getmeili/meilisearch
+          images: ${{ matrix.registry }}
           # Prevent `latest` to be updated for each new tag pushed.
           # We need latest and `vX.Y` tags to only be pushed for the stable Meilisearch releases.
           flavor: latest=false
@@ -88,33 +187,31 @@ jobs:
             type=semver,pattern=v{{major}},enable=${{ steps.check-tag-format.outputs.stable == 'true' }}
             type=raw,value=latest,enable=${{ steps.check-tag-format.outputs.stable == 'true' && steps.check-tag-format.outputs.latest == 'true' }}
 
-      - name: Build and push
-        uses: docker/build-push-action@v6
-        id: build-and-push
-        with:
-          push: true
-          platforms: linux/amd64,linux/arm64
-          tags: ${{ steps.meta.outputs.tags }}
-          build-args: |
-            COMMIT_SHA=${{ github.sha }}
-            COMMIT_DATE=${{ steps.build-metadata.outputs.date }}
-            GIT_TAG=${{ github.ref_name }}
+      - name: Create manifest list and push
+        working-directory: ${{ runner.temp }}/digests
+        run: |
+          docker buildx imagetools create $(jq -cr '.tags | map("-t " + .) | join(" ")' <<< "$DOCKER_METADATA_OUTPUT_JSON") \
+            $(printf '${{ matrix.registry }}@sha256:%s ' *)
+
+      - name: Inspect image to fetch digest to sign
+        run: |
+          digest=$(docker buildx imagetools inspect --format='{{ json .Manifest }}' ${{ matrix.registry }}:${{ steps.meta.outputs.version }} | jq -r '.digest')
+          echo "DIGEST=${digest}" >> $GITHUB_ENV
 
       - name: Sign the images with GitHub OIDC Token
         env:
-          DIGEST: ${{ steps.build-and-push.outputs.digest }}
           TAGS: ${{ steps.meta.outputs.tags }}
         run: |
           images=""
           for tag in ${TAGS}; do
-            images+="${tag}@${DIGEST} "
+            images+="${tag}@${{ env.DIGEST }} "
           done
           cosign sign --yes ${images}
 
-      # /!\ Don't touch this without checking with Cloud team
-      - name: Send CI information to Cloud team
+      # /!\ Don't touch this without checking with engineers working on the Cloud code base on #discussion-engineering Slack channel
+      - name: Notify meilisearch-cloud 
         # Do not send if nightly build (i.e. 'schedule' or 'workflow_dispatch' event)
-        if: github.event_name == 'push'
+        if: ${{ (github.event_name == 'push') && (matrix.edition == 'enterprise') }}
         uses: peter-evans/repository-dispatch@v3
         with:
           token: ${{ secrets.MEILI_BOT_GH_PAT }}
@@ -122,21 +219,13 @@ jobs:
           event-type: cloud-docker-build
           client-payload: '{ "meilisearch_version": "${{ github.ref_name }}", "stable": "${{ steps.check-tag-format.outputs.stable }}" }'
 
-      # Send notification to Swarmia to notify of a deployment: https://app.swarmia.com
-      # - name: 'Setup jq'
-      #   uses: dcarbone/install-jq-action
-      # - name: Send deployment to Swarmia
-      #   if: github.event_name == 'push' && success()
-      #   run: |
-      #     JSON_STRING=$( jq --null-input --compact-output \
-      #     --arg version "${{ github.ref_name }}" \
-      #     --arg appName "meilisearch" \
-      #     --arg environment "production" \
-      #     --arg commitSha "${{ github.sha }}" \
-      #     --arg repositoryFullName "${{ github.repository }}" \
-      #     '{"version": $version, "appName": $appName, "environment": $environment, "commitSha": $commitSha, "repositoryFullName": $repositoryFullName}' )
-
-      #     curl -H "Authorization: ${{ secrets.SWARMIA_DEPLOYMENTS_AUTHORIZATION }}" \
-      #       -H "Content-Type: application/json" \
-      #       -d "$JSON_STRING" \
-      #       https://hook.swarmia.com/deployments
+      # /!\ Don't touch this without checking with integration team members on #discussion-integrations Slack channel 
+      - name: Notify meilisearch-kubernetes
+        # Do not send if nightly build (i.e. 'schedule' or 'workflow_dispatch' event), or if not stable
+        if: ${{ github.event_name == 'push' && matrix.edition == 'community' && steps.check-tag-format.outputs.stable == 'true' }}
+        uses: peter-evans/repository-dispatch@v3
+        with:
+          token: ${{ secrets.MEILI_BOT_GH_PAT }}
+          repository: meilisearch/meilisearch-kubernetes
+          event-type: meilisearch-release
+          client-payload: '{ "version": "${{ github.ref_name }}" }'

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

@@ -32,157 +32,61 @@ jobs:
         if: github.event_name == 'release' && steps.check-tag-format.outputs.stable == 'true'
         run: bash .github/scripts/check-release.sh
 
-  publish-linux:
-    name: Publish binary for Linux
-    runs-on: ubuntu-latest
-    needs: check-version
-    container:
-      # Use ubuntu-22.04 to compile with glibc 2.35
-      image: ubuntu:22.04
-    steps:
-      - uses: actions/checkout@v5
-      - name: Install needed dependencies
-        run: |
-          apt-get update && apt-get install -y curl
-          apt-get install build-essential -y
-      - uses: dtolnay/rust-toolchain@1.89
-      - name: Build
-        run: cargo build --release --locked
-      # 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
-        with:
-          repo_token: ${{ secrets.MEILI_BOT_GH_PAT }}
-          file: target/release/meilisearch
-          asset_name: meilisearch-linux-amd64
-          tag: ${{ github.ref }}
-
-  publish-macos-windows:
-    name: Publish binary for ${{ matrix.os }}
+  publish-binaries:
+    name: Publish binary for ${{ matrix.release }} ${{ matrix.edition }} edition
     runs-on: ${{ matrix.os }}
-    needs: check-version
     strategy:
       fail-fast: false
       matrix:
-        os: [macos-13, windows-2022]
+        edition: [community, enterprise]
+        release:
+          [macos-amd64, macos-aarch64, windows, linux-amd64, linux-aarch64]
         include:
-          - os: macos-13
-            artifact_name: meilisearch
-            asset_name: meilisearch-macos-amd64
-          - os: windows-2022
-            artifact_name: meilisearch.exe
-            asset_name: meilisearch-windows-amd64.exe
+          - edition: "community"
+            feature-flag: ""
+            edition-suffix: ""
+          - edition: "enterprise"
+            feature-flag: "--features enterprise"
+            edition-suffix: "enterprise-"
+          - release: macos-amd64
+            os: macos-15-intel
+            binary_path: release/meilisearch
+            asset_name: macos-amd64
+            extra-args: ""
+          - release: macos-aarch64
+            os: macos-14
+            binary_path: aarch64-apple-darwin/release/meilisearch
+            asset_name: macos-apple-silicon
+            extra-args: "--target aarch64-apple-darwin"
+          - release: windows
+            os: windows-2022
+            binary_path: release/meilisearch.exe
+            asset_name: windows-amd64.exe
+            extra-args: ""
+          - release: linux-amd64
+            os: ubuntu-22.04
+            binary_path: x86_64-unknown-linux-gnu/release/meilisearch
+            asset_name: linux-amd64
+            extra-args: "--target x86_64-unknown-linux-gnu"
+          - release: linux-aarch64
+            os: ubuntu-22.04-arm
+            binary_path: aarch64-unknown-linux-gnu/release/meilisearch
+            asset_name: linux-aarch64
+            extra-args: "--target aarch64-unknown-linux-gnu"
+    needs: check-version
     steps:
       - uses: actions/checkout@v5
       - uses: dtolnay/rust-toolchain@1.89
       - name: Build
-        run: cargo build --release --locked
+        run: cargo build --release --locked ${{ matrix.feature-flag }} ${{ matrix.extra-args }}
       # 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
         with:
           repo_token: ${{ secrets.MEILI_BOT_GH_PAT }}
-          file: target/release/${{ matrix.artifact_name }}
-          asset_name: ${{ matrix.asset_name }}
-          tag: ${{ github.ref }}
-
-  publish-macos-apple-silicon:
-    name: Publish binary for macOS silicon
-    runs-on: macos-13
-    needs: check-version
-    strategy:
-      matrix:
-        include:
-          - target: aarch64-apple-darwin
-            asset_name: meilisearch-macos-apple-silicon
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v5
-      - name: Installing Rust toolchain
-        uses: dtolnay/rust-toolchain@1.89
-        with:
-          profile: minimal
-          target: ${{ matrix.target }}
-      - name: Cargo build
-        uses: actions-rs/cargo@v1
-        with:
-          command: build
-          args: --release --target ${{ matrix.target }}
-      - name: Upload the binary to release
-        # 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:
-          repo_token: ${{ secrets.MEILI_BOT_GH_PAT }}
-          file: target/${{ matrix.target }}/release/meilisearch
-          asset_name: ${{ matrix.asset_name }}
-          tag: ${{ github.ref }}
-
-  publish-aarch64:
-    name: Publish binary for aarch64
-    runs-on: ubuntu-latest
-    needs: check-version
-    env:
-      DEBIAN_FRONTEND: noninteractive
-    container:
-      # Use ubuntu-22.04 to compile with glibc 2.35
-      image: ubuntu:22.04
-    strategy:
-      matrix:
-        include:
-          - target: aarch64-unknown-linux-gnu
-            asset_name: meilisearch-linux-aarch64
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v5
-      - name: Install needed dependencies
-        run: |
-          apt-get update -y && apt upgrade -y
-          apt-get install -y curl build-essential gcc-aarch64-linux-gnu
-      - name: Set up Docker for cross compilation
-        run: |
-          apt-get install -y curl apt-transport-https ca-certificates software-properties-common
-          curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add -
-          add-apt-repository "deb [arch=$(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@1.89
-        with:
-          profile: minimal
-          target: ${{ matrix.target }}
-      - name: Configure target aarch64 GNU
-        ## Environment variable is not passed using env:
-        ## LD gold won't work with MUSL
-        # env:
-        #   JEMALLOC_SYS_WITH_LG_PAGE: 16
-        #   RUSTFLAGS: '-Clink-arg=-fuse-ld=gold'
-        run: |
-          echo '[target.aarch64-unknown-linux-gnu]' >> ~/.cargo/config
-          echo 'linker = "aarch64-linux-gnu-gcc"' >> ~/.cargo/config
-          echo 'JEMALLOC_SYS_WITH_LG_PAGE=16' >> $GITHUB_ENV
-      - name: Install a default toolchain that will be used to build cargo cross
-        run: |
-          rustup default stable
-      - name: Cargo build
-        uses: actions-rs/cargo@v1
-        with:
-          command: build
-          use-cross: true
-          args: --release --target ${{ matrix.target }}
-        env:
-          CROSS_DOCKER_IN_DOCKER: true
-      - name: List target output files
-        run: ls -lR ./target
-      - name: Upload the binary to release
-        # 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:
-          repo_token: ${{ secrets.MEILI_BOT_GH_PAT }}
-          file: target/${{ matrix.target }}/release/meilisearch
-          asset_name: ${{ matrix.asset_name }}
+          file: target/${{ matrix.binary_path }}
+          asset_name: meilisearch-${{ matrix.edition-suffix }}${{ matrix.asset_name }}
           tag: ${{ github.ref }}
 
   publish-openapi-file:
@@ -200,13 +104,13 @@ jobs:
       - name: Generate OpenAPI file
         run: |
           cd crates/openapi-generator
-          cargo run --release -- --pretty --output ../../meilisearch.json
+          cargo run --release -- --pretty --output ../../meilisearch-openapi.json
       - name: Upload OpenAPI to Release
         # 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:
           repo_token: ${{ secrets.MEILI_BOT_GH_PAT }}
-          file: ./meilisearch.json
+          file: ./meilisearch-openapi.json
           asset_name: meilisearch-openapi.json
           tag: ${{ github.ref }}

.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

@@ -68,7 +68,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -92,7 +92,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -122,7 +122,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -149,7 +149,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -184,7 +184,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -213,7 +213,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -238,7 +238,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -263,7 +263,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -284,7 +284,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -307,7 +307,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -338,7 +338,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -370,7 +370,7 @@ jobs:
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
+        image: getmeili/meilisearch-enterprise:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}

.github/workflows/test-suite.yml

@@ -15,31 +15,40 @@ env:
 
 jobs:
   test-linux:
-    name: Tests on ubuntu-22.04
-    runs-on: ubuntu-latest
-    container:
-      # Use ubuntu-22.04 to compile with glibc 2.35
-      image: ubuntu:22.04
+    name: Tests on Ubuntu
+    runs-on: ${{ matrix.runner }}
+    strategy:
+      matrix:
+        runner: [ubuntu-22.04, ubuntu-22.04-arm]
+        features: ["", "--features enterprise"]
     steps:
       - uses: actions/checkout@v5
-      - name: Install needed dependencies
+      - name: check free space before
+        run: df -h
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
         run: |
-          apt-get update && apt-get install -y curl
-          apt-get install build-essential -y
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
+      - name: check free space after
+        run: df -h
       - name: Setup test with Rust stable
         uses: dtolnay/rust-toolchain@1.89
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
-      - name: Run cargo check without any default features
+        with:
+          key: ${{ matrix.features }}
+      - name: Run cargo build without any default features
         uses: actions-rs/cargo@v1
         with:
           command: build
-          args: --locked --release --no-default-features --all
+          args: --locked --no-default-features --all
       - name: Run cargo test
         uses: actions-rs/cargo@v1
         with:
           command: test
-          args: --locked --release --all
+          args: --locked --all ${{ matrix.features }}
 
   test-others:
     name: Tests on ${{ matrix.os }}
@@ -47,51 +56,58 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [macos-13, windows-2022]
+        os: [macos-14, windows-2022]
+        features: ["", "--features enterprise"]
+    if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
     steps:
       - uses: actions/checkout@v5
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
       - uses: dtolnay/rust-toolchain@1.89
-      - name: Run cargo check without any default features
+      - name: Run cargo build without any default features
         uses: actions-rs/cargo@v1
         with:
           command: build
-          args: --locked --release --no-default-features --all
+          args: --locked --no-default-features --all
       - name: Run cargo test
         uses: actions-rs/cargo@v1
         with:
           command: test
-          args: --locked --release --all
+          args: --locked --all ${{ matrix.features }}
 
   test-all-features:
     name: Tests almost all features
-    runs-on: ubuntu-latest
-    container:
-      # Use ubuntu-22.04 to compile with glibc 2.35
-      image: ubuntu:22.04
+    runs-on: ubuntu-22.04
     if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
     steps:
       - uses: actions/checkout@v5
-      - name: Install needed dependencies
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
         run: |
-          apt-get update
-          apt-get install --assume-yes build-essential curl
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - uses: dtolnay/rust-toolchain@1.89
       - name: Run cargo build with almost all features
         run: |
-          cargo build --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda,test-ollama)"
+          cargo build --workspace --locked --features "$(cargo xtask list-features --exclude-feature cuda,test-ollama)"
       - name: Run cargo test with almost all features
         run: |
-          cargo test --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda,test-ollama)"
+          cargo test --workspace --locked --features "$(cargo xtask list-features --exclude-feature cuda,test-ollama)"
 
   ollama-ubuntu:
     name: Test with Ollama
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     env:
       MEILI_TEST_OLLAMA_SERVER: "http://localhost:11434"
     steps:
       - uses: actions/checkout@v5
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
+        run: |
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - name: Install Ollama
         run: |
           curl -fsSL https://ollama.com/install.sh | sudo -E sh
@@ -115,20 +131,20 @@ jobs:
         uses: actions-rs/cargo@v1
         with:
           command: test
-          args: --locked --release --all --features test-ollama ollama
+          args: --locked -p meilisearch --features test-ollama ollama
 
   test-disabled-tokenization:
     name: Test disabled tokenization
-    runs-on: ubuntu-latest
-    container:
-      image: ubuntu:22.04
+    runs-on: ubuntu-22.04
     if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
     steps:
       - uses: actions/checkout@v5
-      - name: Install needed dependencies
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
         run: |
-          apt-get update
-          apt-get install --assume-yes build-essential curl
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - uses: dtolnay/rust-toolchain@1.89
       - name: Run cargo tree without default features and check lindera is not present
         run: |
@@ -140,36 +156,39 @@ jobs:
         run: |
           cargo tree -f '{p} {f}' -e normal | grep lindera -qz
 
-  # We run tests in debug also, to make sure that the debug_assertions are hit
-  test-debug:
-    name: Run tests in debug
-    runs-on: ubuntu-latest
-    container:
-      # Use ubuntu-22.04 to compile with glibc 2.35
-      image: ubuntu:22.04
+  build:
+    name: Build in release
+    runs-on: ubuntu-22.04
     steps:
       - uses: actions/checkout@v5
-      - name: Install needed dependencies
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
         run: |
-          apt-get update && apt-get install -y curl
-          apt-get install build-essential -y
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - uses: dtolnay/rust-toolchain@1.89
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
-      - name: Run tests in debug
-        uses: actions-rs/cargo@v1
-        with:
-          command: test
-          args: --locked --all
+      - name: Build
+        run: cargo build --release --locked --target x86_64-unknown-linux-gnu
 
   clippy:
     name: Run Clippy
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
+    strategy:
+      matrix:
+        features: ["", "--features enterprise"]
     steps:
       - uses: actions/checkout@v5
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
+        run: |
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - uses: dtolnay/rust-toolchain@1.89
         with:
-          profile: minimal
           components: clippy
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
@@ -177,18 +196,21 @@ jobs:
         uses: actions-rs/cargo@v1
         with:
           command: clippy
-          args: --all-targets -- --deny warnings
+          args: --all-targets ${{ matrix.features }} -- --deny warnings
 
   fmt:
     name: Run Rustfmt
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     steps:
       - uses: actions/checkout@v5
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
+        run: |
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - uses: dtolnay/rust-toolchain@1.89
         with:
-          profile: minimal
-          toolchain: nightly-2024-07-09
-          override: true
           components: rustfmt
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.8.0
@@ -199,3 +221,23 @@ jobs:
         run: |
           echo -ne "\n" > crates/benchmarks/benches/datasets_paths.rs
           cargo fmt --all -- --check
+
+  declarative-tests:
+    name: Run declarative tests
+    runs-on: ubuntu-22.04-arm
+    permissions:
+      contents: read
+    steps:
+      - uses: actions/checkout@v5
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
+        run: |
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
+      - uses: dtolnay/rust-toolchain@1.89
+      - name: Cache dependencies
+        uses: Swatinem/rust-cache@v2.8.0
+      - name: Run declarative tests
+        run: |
+          cargo xtask test workloads/tests/*.json

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

@@ -18,9 +18,13 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v5
+      - name: Clean space as per https://github.com/actions/virtual-environments/issues/709
+        run: |
+          sudo rm -rf "/opt/ghc" || true
+          sudo rm -rf "/usr/share/dotnet" || true
+          sudo rm -rf "/usr/local/lib/android" || true
+          sudo rm -rf "/usr/local/share/boost" || true
       - uses: dtolnay/rust-toolchain@1.89
-        with:
-          profile: minimal
       - name: Install sd
         run: cargo install sd
       - name: Update Cargo.toml file

.gitignore

@@ -29,3 +29,6 @@ crates/meilisearch/db.snapshot
 
 # Fuzzcheck data for the facet indexing fuzz test
 crates/milli/fuzz/update::facet::incremental::fuzz::fuzz/
+
+# OpenAPI generator
+**/meilisearch-openapi.json

BENCHMARKS.md

@@ -124,6 +124,7 @@ They are JSON files with the following structure (comments are not actually supp
 {
   // Name of the workload. Must be unique to the workload, as it will be used to group results on the dashboard.
   "name": "hackernews.ndjson_1M,no-threads",
+  "type": "bench",
   // Number of consecutive runs of the commands that should be performed.
   // Each run uses a fresh instance of Meilisearch and a fresh database.
   // Each run produces its own report file.

CONTRIBUTING.md

@@ -117,7 +117,7 @@ With swagger:
 With the internal crate:
 ```bash
 cd crates/openapi-generator
-cargo run --release -- --pretty --output meilisearch.json
+cargo run --release -- --pretty
 ```
 
 ### Logging

Cargo.toml

@@ -23,7 +23,7 @@ members = [
 ]
 
 [workspace.package]
-version = "1.22.1"
+version = "1.28.2"
 authors = [
     "Quentin de Quelen <quentin@dequelen.me>",
     "Clément Renault <clement@meilisearch.com>",
@@ -50,3 +50,5 @@ opt-level = 3
 opt-level = 3
 [profile.dev.package.roaring]
 opt-level = 3
+[profile.dev.package.gemm-f16]
+opt-level = 3

Cross.toml

@@ -1,7 +0,0 @@
-[build.env]
-passthrough = [
-    "RUST_BACKTRACE",
-    "CARGO_TERM_COLOR",
-    "RUSTFLAGS",
-    "JEMALLOC_SYS_WITH_LG_PAGE"
-]

Dockerfile

@@ -1,5 +1,5 @@
 # Compile
-FROM    rust:1.89-alpine3.20 AS compiler
+FROM    rust:1.89-alpine3.22 AS compiler
 
 RUN     apk add -q --no-cache build-base openssl-dev
 
@@ -8,19 +8,17 @@ WORKDIR /
 ARG     COMMIT_SHA
 ARG     COMMIT_DATE
 ARG     GIT_TAG
+ARG     EXTRA_ARGS
 ENV     VERGEN_GIT_SHA=${COMMIT_SHA} VERGEN_GIT_COMMIT_TIMESTAMP=${COMMIT_DATE} VERGEN_GIT_DESCRIBE=${GIT_TAG}
 ENV     RUSTFLAGS="-C target-feature=-crt-static"
 
 COPY    . .
 RUN     set -eux; \
         apkArch="$(apk --print-arch)"; \
-        if [ "$apkArch" = "aarch64" ]; then \
-            export JEMALLOC_SYS_WITH_LG_PAGE=16; \
-        fi && \
-        cargo build --release -p meilisearch -p meilitool
+        cargo build --release -p meilisearch -p meilitool ${EXTRA_ARGS}
 
 # 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.

TESTING.md

@@ -0,0 +1,326 @@
+# Declarative tests
+
+Declarative tests ensure that Meilisearch features remain stable across versions.
+
+While we already have unit tests, those are run against **temporary databases** that are created fresh each time and therefore never risk corruption.
+
+Declarative tests instead **simulate the lifetime of a database**: they chain together commands and requests to change the binary, verifying that database state and API responses remain consistent.
+
+## Basic example
+
+```jsonc
+{
+  "type": "test",
+  "name": "api-keys",
+  "binary": { // the first command will run on the binary following this specification.
+    "source": "release", // get the binary as a release from GitHub
+    "version": "1.19.0", // version to fetch
+    "edition": "community" // edition to fetch
+  },
+  "commands": []
+}
+```
+
+This example defines a no-op test (it does nothing).
+
+If the file is saved at `workloads/tests/example.json`, you can run it with:
+
+```bash
+cargo xtask test workloads/tests/example.json
+```
+
+## Commands
+
+Commands represent API requests sent to Meilisearch endpoints during a test.
+
+They are executed sequentially, and their responses can be validated to ensure consistent behavior across upgrades.
+
+```jsonc
+
+{
+  "route": "keys",
+  "method": "POST",
+  "body": {
+    "inline": {
+      "actions": [
+        "search",
+        "documents.add"
+      ],
+      "description": "Test API Key",
+      "expiresAt": null,
+      "indexes": [ "movies" ]
+    }
+  }
+}
+```
+
+This command issues a `POST /keys` request, creating an API key with permissions to search and add documents in the `movies` index.
+
+### Using assets in commands
+
+To keep tests concise and reusable, you can define **assets** at the root of the workload file.
+
+Assets are external data sources (such as datasets) that are cached between runs, making tests faster and easier to read.
+
+```jsonc
+{
+  "type": "test",
+  "name": "movies",
+  "binary": {
+    "source": "release",
+    "version": "1.19.0",
+    "edition": "community"
+  },
+  "assets": {
+    "movies.json": {
+      "local_location": null,
+      "remote_location": "https://milli-benchmarks.fra1.digitaloceanspaces.com/bench/datasets/movies.json",
+      "sha256": "5b6e4cb660bc20327776e8a33ea197b43d9ec84856710ead1cc87ab24df77de1"
+    }
+  },
+  "commands": [
+    {
+      "route": "indexes/movies/documents",
+      "method": "POST",
+      "body": {
+        "asset": "movies.json"
+      }
+    }
+  ]
+}
+```
+
+In this example:
+- The `movies.json` dataset is defined as an asset, pointing to a remote URL.
+- The SHA-256 checksum ensures integrity.
+- The `POST /indexes/movies/documents` command uses this asset as the request body.
+
+This makes the test much cleaner than inlining a large dataset directly into the command.
+
+For asset handling, please refer to the [declarative benchmarks documentation](/BENCHMARKS.md#adding-new-assets).
+
+### Asserting responses
+
+Commands can specify both the **expected status code** and the **expected response body**.
+
+```jsonc
+{
+  "route": "indexes/movies/documents",
+  "method": "POST",
+  "body": {
+    "asset": "movies.json"
+  },
+  "expectedStatus": 202,
+  "expectedResponse": {
+    "enqueuedAt": "[timestamp]", // Set to a bracketed string to ignore the value
+    "indexUid": "movies",
+    "status": "enqueued",
+    "taskUid": 1,
+    "type": "documentAdditionOrUpdate"
+  },
+  "synchronous": "WaitForTask"
+}
+```
+
+Manually writing `expectedResponse` fields can be tedious.
+
+Instead, you can let the test runner populate them automatically:
+
+```bash
+# Run the workload to populate expected fields. Only adds the missing ones, doesn't change existing data
+cargo xtask test workloads/tests/example.json --add-missing-responses
+
+# OR
+
+# Run the workload to populate expected fields. Updates all fields including existing ones
+cargo xtask test workloads/tests/example.json --update-responses
+```
+
+This workflow is recommended:
+
+1. Write the test without expected fields.
+2. Run it with `--add-missing-responses` to capture the actual responses.
+3. Review and commit the generated expectations.
+
+## Changing binary
+
+It is possible to insert an instruction to change the current Meilisearch instance from one binary specification to another during a test.
+
+When executed, such an instruction will:
+1. Stop the current Meilisearch instance.
+2. Fetch the binary specified by the instruction.
+3. Restart the server with the specified binary on the same database.
+
+```jsonc
+{
+  "type": "test",
+  "name": "movies",
+  "binary": {
+    "source": "release",
+    "version": "1.19.0", // start with version v1.19.0
+    "edition": "community"
+  },
+  "assets": {
+    "movies.json": {
+      "local_location": null,
+      "remote_location": "https://milli-benchmarks.fra1.digitaloceanspaces.com/bench/datasets/movies.json",
+      "sha256": "5b6e4cb660bc20327776e8a33ea197b43d9ec84856710ead1cc87ab24df77de1"
+    }
+  },
+  "commands": [
+    // setup some data
+    {
+      "route": "indexes/movies/documents",
+      "method": "POST",
+      "body": {
+        "asset": "movies.json"
+      }
+    },
+    // switch binary to v1.24.0
+    {
+      "binary": {
+        "source": "release",
+        "version": "1.24.0",
+        "edition": "community"
+      }
+    }
+  ]
+}
+```
+
+### Typical Usage
+
+In most cases, the change binary instruction will be used to update a database.
+
+- **Set up** some data using commands on an older version.
+- **Upgrade** to the latest version.
+- **Assert** that the data and API behavior remain correct after the upgrade.
+
+To properly test the dumpless upgrade, one should typically:
+
+1. Open the database without processing the update task: Use a `binary` instruction to switch to the desired version, passing `--experimental-dumpless-upgrade` and `--experimental-max-number-of-batched-tasks=0` as extra CLI arguments
+2. Check that the search, stats and task queue still work.
+3. Open the database and process the update task: Use a `binary` instruction to switch to the desired version, passing `--experimental-dumpless-upgrade` as the extra CLI argument. Use a `health` command to wait for the upgrade task to finish.
+4. Check that the indexing, search, stats, and task queue still work.
+
+```jsonc
+{
+  "type": "test",
+  "name": "movies",
+  "binary": {
+    "source": "release",
+    "version": "1.12.0",
+    "edition": "community"
+  },
+  "commands": [
+    // 0. Run commands to populate the database
+    {
+      // ..
+    },
+    // 1. Open the database with new MS without processing the update task
+    {
+      "binary": {
+        "source": "build", // build the binary from the sources in the current git repository
+        "edition": "community",
+        "extraCliArgs": [
+          "--experimental-dumpless-upgrade", // allows to open with a newer MS
+          "--experimental-max-number-of-batched-tasks=0" // prevent processing of the update task
+        ]
+      }
+    },
+    // 2. Check the search etc.
+    {
+      // ..
+    },
+    // 3. Open the database with new MS and processing the update task
+    {
+      "binary": {
+        "source": "build", // build the binary from the sources in the current git repository
+        "edition": "community",
+        "extraCliArgs": [
+          "--experimental-dumpless-upgrade" // allows to open with a newer MS
+          // no `--experimental-max-number-of-batched-tasks=0`
+        ]
+      }
+    },
+    // 4. Check the indexing, search, etc.
+    {
+      // ..
+    }
+  ]
+}
+```
+
+This ensures backward compatibility: databases created with older Meilisearch versions should remain functional and consistent after an upgrade.
+
+## Variables
+
+Sometimes a command needs to use a value returned by a **previous response**.
+These values can be captured and reused using the register field.
+
+```jsonc
+{
+  "route": "keys",
+  "method": "POST",
+  "body": {
+    "inline": {
+        "actions": [
+        "search",
+        "documents.add"
+        ],
+        "description": "Test API Key",
+        "expiresAt": null,
+        "indexes": [ "movies" ]
+    }
+  },
+  "expectedResponse": {
+      "key": "c6f64630bad2996b1f675007c8800168e14adf5d6a7bb1a400a6d2b158050eaf",
+      // ...
+  },
+  "register": {
+    "key": "/key"
+  },
+  "synchronous": "WaitForResponse"
+}
+```
+
+The `register` field captures the value at the JSON path `/key` from the response.
+Paths follow the **JavaScript Object Notation Pointer (RFC 6901)** format.
+Registered variables are available for all subsequent commands.
+
+Registered variables can be referenced by wrapping their name in double curly braces:
+
+In the route/path:
+
+```jsonc
+{
+  "route": "tasks/{{ task_id }}",
+  "method": "GET"
+}
+```
+
+In the request body:
+
+```jsonc
+{
+  "route": "indexes/movies/documents",
+  "method": "PATCH",
+  "body": {
+    "inline": {
+      "id": "{{ document_id }}",
+      "overview": "Shazam turns evil and the world is in danger.",
+    }
+  }
+}
+```
+
+Or they can be referenced by their name (**without curly braces**) as an API key:
+
+```jsonc
+{
+  "route": "indexes/movies/documents",
+  "method": "POST",
+  "body": { /* ... */ },
+  "apiKeyVariable": "key" // The **content** of the key variable will be used as an API key
+}
+```

crates/benchmarks/Cargo.toml

@@ -11,27 +11,27 @@ edition.workspace = true
 license.workspace = true
 
 [dependencies]
-anyhow = "1.0.98"
-bumpalo = "3.18.1"
-csv = "1.3.1"
-memmap2 = "0.9.7"
+anyhow = "1.0.100"
+bumpalo = "3.19.0"
+csv = "1.4.0"
+memmap2 = "0.9.9"
 milli = { path = "../milli" }
-mimalloc = { version = "0.1.47", default-features = false }
-serde_json = { version = "1.0.140", features = ["preserve_order"] }
-tempfile = "3.20.0"
+mimalloc = { version = "0.1.48", default-features = false }
+serde_json = { version = "1.0.145", features = ["preserve_order"] }
+tempfile = "3.23.0"
 
 [dev-dependencies]
-criterion = { version = "0.6.0", features = ["html_reports"] }
+criterion = { version = "0.7.0", features = ["html_reports"] }
 rand = "0.8.5"
 rand_chacha = "0.3.1"
 roaring = "0.10.12"
 
 [build-dependencies]
-anyhow = "1.0.98"
-bytes = "1.10.1"
-convert_case = "0.8.0"
-flate2 = "1.1.2"
-reqwest = { version = "0.12.20", features = ["blocking", "rustls-tls"], default-features = false }
+anyhow = "1.0.100"
+bytes = "1.11.0"
+convert_case = "0.9.0"
+flate2 = "1.1.5"
+reqwest = { version = "0.12.24", features = ["blocking", "rustls-tls"], default-features = false }
 
 [features]
 default = ["milli/all-tokenizations"]

crates/benchmarks/benches/indexing.rs

@@ -21,6 +21,10 @@ use roaring::RoaringBitmap;
 #[global_allocator]
 static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
 
+fn no_cancel() -> bool {
+    false
+}
+
 const BENCHMARK_ITERATION: usize = 10;
 
 fn setup_dir(path: impl AsRef<Path>) {
@@ -65,7 +69,7 @@ fn setup_settings<'t>(
     let sortable_fields = sortable_fields.iter().map(|s| s.to_string()).collect();
     builder.set_sortable_fields(sortable_fields);
 
-    builder.execute(&|| false, &Progress::default(), Default::default()).unwrap();
+    builder.execute(&no_cancel, &Progress::default(), Default::default()).unwrap();
 }
 
 fn setup_index_with_settings(
@@ -152,7 +156,7 @@ fn indexing_songs_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -168,7 +172,7 @@ fn indexing_songs_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -220,7 +224,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -236,7 +240,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -266,7 +270,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -282,7 +286,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -336,7 +340,7 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -352,7 +356,7 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -414,7 +418,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -430,7 +434,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -460,7 +464,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -476,7 +480,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -502,7 +506,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -518,7 +522,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -571,7 +575,7 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -587,7 +591,7 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -639,7 +643,7 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -655,7 +659,7 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -707,7 +711,7 @@ fn indexing_wiki(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -723,7 +727,7 @@ fn indexing_wiki(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -774,7 +778,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -790,7 +794,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -820,7 +824,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -836,7 +840,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -889,7 +893,7 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -905,7 +909,7 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -967,7 +971,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -983,7 +987,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1014,7 +1018,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1030,7 +1034,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1057,7 +1061,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1073,7 +1077,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1125,7 +1129,7 @@ fn indexing_movies_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1141,7 +1145,7 @@ fn indexing_movies_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1192,7 +1196,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1208,7 +1212,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1238,7 +1242,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1254,7 +1258,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1307,7 +1311,7 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1323,7 +1327,7 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1372,7 +1376,7 @@ fn delete_documents_from_ids(index: Index, document_ids_to_delete: Vec<RoaringBi
             Some(primary_key),
             &document_changes,
             RuntimeEmbedders::default(),
-            &|| false,
+            &no_cancel,
             &Progress::default(),
             &Default::default(),
         )
@@ -1422,7 +1426,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1438,7 +1442,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1468,7 +1472,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1484,7 +1488,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1510,7 +1514,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1526,7 +1530,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1601,7 +1605,7 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1617,7 +1621,7 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1693,7 +1697,7 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1709,7 +1713,7 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1777,7 +1781,7 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1793,7 +1797,7 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1845,7 +1849,7 @@ fn indexing_geo(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1861,7 +1865,7 @@ fn indexing_geo(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1912,7 +1916,7 @@ fn reindexing_geo(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1928,7 +1932,7 @@ fn reindexing_geo(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -1958,7 +1962,7 @@ fn reindexing_geo(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -1974,7 +1978,7 @@ fn reindexing_geo(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )
@@ -2027,7 +2031,7 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
                         &rtxn,
                         None,
                         &mut new_fields_ids_map,
-                        &|| false,
+                        &no_cancel,
                         Progress::default(),
                         None,
                     )
@@ -2043,7 +2047,7 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
                     primary_key,
                     &document_changes,
                     RuntimeEmbedders::default(),
-                    &|| false,
+                    &no_cancel,
                     &Progress::default(),
                     &Default::default(),
                 )

crates/build-info/Cargo.toml

@@ -11,8 +11,8 @@ license.workspace = true
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-time = { version = "0.3.41", features = ["parsing"] }
+time = { version = "0.3.44", features = ["parsing"] }
 
 [build-dependencies]
-anyhow = "1.0.98"
-vergen-git2 = "1.0.7"
+anyhow = "1.0.100"
+vergen-gitcl = "1.0.8"

crates/build-info/build.rs

@@ -15,7 +15,7 @@ fn emit_git_variables() -> anyhow::Result<()> {
     // Note: any code that needs VERGEN_ environment variables should take care to define them manually in the Dockerfile and pass them
     // in the corresponding GitHub workflow (publish_docker.yml).
     // This is due to the Dockerfile building the binary outside of the git directory.
-    let mut builder = vergen_git2::Git2Builder::default();
+    let mut builder = vergen_gitcl::GitclBuilder::default();
 
     builder.branch(true);
     builder.commit_timestamp(true);
@@ -25,5 +25,5 @@ fn emit_git_variables() -> anyhow::Result<()> {
 
     let git2 = builder.build()?;
 
-    vergen_git2::Emitter::default().fail_on_error().add_instructions(&git2)?.emit()
+    vergen_gitcl::Emitter::default().fail_on_error().add_instructions(&git2)?.emit()
 }

crates/build-info/src/main.rs

@@ -0,0 +1,6 @@
+use build_info::BuildInfo;
+
+fn main() {
+    let info = BuildInfo::from_build();
+    dbg!(info);
+}

crates/dump/Cargo.toml

@@ -11,24 +11,27 @@ readme.workspace = true
 license.workspace = true
 
 [dependencies]
-anyhow = "1.0.98"
-flate2 = "1.1.2"
+anyhow = "1.0.100"
+flate2 = "1.1.5"
 http = "1.3.1"
 meilisearch-types = { path = "../meilisearch-types" }
 once_cell = "1.21.3"
-regex = "1.11.1"
+regex = "1.12.2"
 roaring = { version = "0.10.12", features = ["serde"] }
-serde = { version = "1.0.219", features = ["derive"] }
-serde_json = { version = "1.0.140", features = ["preserve_order"] }
+serde = { version = "1.0.228", features = ["derive"] }
+serde_json = { version = "1.0.145", features = ["preserve_order"] }
 tar = "0.4.44"
-tempfile = "3.20.0"
-thiserror = "2.0.12"
-time = { version = "0.3.41", features = ["serde-well-known", "formatting", "parsing", "macros"] }
+tempfile = "3.23.0"
+thiserror = "2.0.17"
+time = { version = "0.3.44", features = ["serde-well-known", "formatting", "parsing", "macros"] }
 tracing = "0.1.41"
-uuid = { version = "1.17.0", features = ["serde", "v4"] }
+uuid = { version = "1.18.1", features = ["serde", "v4"] }
 
 [dev-dependencies]
 big_s = "1.0.2"
 maplit = "1.0.2"
 meili-snap = { path = "../meili-snap" }
 meilisearch-types = { path = "../meilisearch-types" }
+
+[features]
+enterprise = ["meilisearch-types/enterprise"]

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.
@@ -178,6 +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,
         }
     }
 }
@@ -259,13 +262,13 @@ pub(crate) mod test {
     use big_s::S;
     use maplit::{btreemap, btreeset};
     use meilisearch_types::batches::{Batch, BatchEnqueuedAt, BatchStats};
-    use meilisearch_types::enterprise_edition::network::{Network, Remote};
     use meilisearch_types::facet_values_sort::FacetValuesSort;
     use meilisearch_types::features::RuntimeTogglableFeatures;
     use meilisearch_types::index_uid_pattern::IndexUidPattern;
     use meilisearch_types::keys::{Action, Key};
     use meilisearch_types::milli::update::Setting;
     use meilisearch_types::milli::{self, FilterableAttributesRule};
+    use meilisearch_types::network::{Network, Remote};
     use meilisearch_types::settings::{Checked, FacetingSettings, Settings};
     use meilisearch_types::task_view::DetailsView;
     use meilisearch_types::tasks::{BatchStopReason, Details, Kind, Status};
@@ -396,6 +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,
             ),
@@ -421,6 +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(),
@@ -441,6 +446,7 @@ pub(crate) mod test {
                     started_at: None,
                     finished_at: None,
                     network: None,
+                    custom_metadata: None,
                 },
                 None,
             ),

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/dump/src/reader/v6/mod.rs

@@ -24,7 +24,7 @@ pub type Batch = meilisearch_types::batches::Batch;
 pub type Key = meilisearch_types::keys::Key;
 pub type ChatCompletionSettings = meilisearch_types::features::ChatCompletionSettings;
 pub type RuntimeTogglableFeatures = meilisearch_types::features::RuntimeTogglableFeatures;
-pub type Network = meilisearch_types::enterprise_edition::network::Network;
+pub type Network = meilisearch_types::network::Network;
 pub type Webhooks = meilisearch_types::webhooks::WebhooksDumpView;
 
 // ===== Other types to clarify the code of the compat module

crates/dump/src/writer.rs

@@ -5,9 +5,9 @@ use std::path::PathBuf;
 use flate2::write::GzEncoder;
 use flate2::Compression;
 use meilisearch_types::batches::Batch;
-use meilisearch_types::enterprise_edition::network::Network;
 use meilisearch_types::features::{ChatCompletionSettings, RuntimeTogglableFeatures};
 use meilisearch_types::keys::Key;
+use meilisearch_types::network::Network;
 use meilisearch_types::settings::{Checked, Settings};
 use meilisearch_types::webhooks::WebhooksDumpView;
 use serde_json::{Map, Value};

crates/file-store/Cargo.toml

@@ -11,7 +11,7 @@ edition.workspace = true
 license.workspace = true
 
 [dependencies]
-tempfile = "3.20.0"
-thiserror = "2.0.12"
+tempfile = "3.23.0"
+thiserror = "2.0.17"
 tracing = "0.1.41"
-uuid = { version = "1.17.0", features = ["serde", "v4"] }
+uuid = { version = "1.18.1", features = ["serde", "v4"] }

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/flatten-serde-json/Cargo.toml

@@ -16,7 +16,7 @@ license.workspace = true
 serde_json = "1.0"
 
 [dev-dependencies]
-criterion = { version = "0.6.0", features = ["html_reports"] }
+criterion = { version = "0.7.0", features = ["html_reports"] }
 
 [[bench]]
 name = "benchmarks"

crates/fuzzers/Cargo.toml

@@ -11,12 +11,12 @@ edition.workspace = true
 license.workspace = true
 
 [dependencies]
-arbitrary = { version = "1.4.1", features = ["derive"] }
-bumpalo = "3.18.1"
-clap = { version = "4.5.40", features = ["derive"] }
+arbitrary = { version = "1.4.2", features = ["derive"] }
+bumpalo = "3.19.0"
+clap = { version = "4.5.52", features = ["derive"] }
 either = "1.15.0"
 fastrand = "2.3.0"
 milli = { path = "../milli" }
-serde = { version = "1.0.219", features = ["derive"] }
-serde_json = { version = "1.0.140", features = ["preserve_order"] }
-tempfile = "3.20.0"
+serde = { version = "1.0.228", features = ["derive"] }
+serde_json = { version = "1.0.145", features = ["preserve_order"] }
+tempfile = "3.23.0"

crates/index-scheduler/Cargo.toml

@@ -11,31 +11,33 @@ edition.workspace = true
 license.workspace = true
 
 [dependencies]
-anyhow = "1.0.98"
+anyhow = "1.0.100"
 bincode = "1.3.3"
 byte-unit = "5.1.6"
-bumpalo = "3.18.1"
+bytes = "1.11.0"
+bumpalo = "3.19.0"
 bumparaw-collections = "0.1.4"
-convert_case = "0.8.0"
-csv = "1.3.1"
+convert_case = "0.9.0"
+csv = "1.4.0"
 derive_builder = "0.20.2"
 dump = { path = "../dump" }
-enum-iterator = "2.1.0"
+enum-iterator = "2.3.0"
 file-store = { path = "../file-store" }
-flate2 = "1.1.2"
-indexmap = "2.9.0"
+flate2 = "1.1.5"
+indexmap = "2.12.0"
 meilisearch-auth = { path = "../meilisearch-auth" }
 meilisearch-types = { path = "../meilisearch-types" }
-memmap2 = "0.9.7"
+memmap2 = "0.9.9"
 page_size = "0.6.0"
-rayon = "1.10.0"
+rayon = "1.11.0"
 roaring = { version = "0.10.12", features = ["serde"] }
-serde = { version = "1.0.219", features = ["derive"] }
-serde_json = { version = "1.0.140", features = ["preserve_order"] }
+serde = { version = "1.0.228", features = ["derive"] }
+serde_json = { version = "1.0.145", features = ["preserve_order"] }
+tar = "0.4.44"
 synchronoise = "1.0.1"
-tempfile = "3.20.0"
-thiserror = "2.0.12"
-time = { version = "0.3.41", features = [
+tempfile = "3.23.0"
+thiserror = "2.0.17"
+time = { version = "0.3.44", features = [
     "serde-well-known",
     "formatting",
     "parsing",
@@ -43,8 +45,11 @@ time = { version = "0.3.41", features = [
 ] }
 tracing = "0.1.41"
 ureq = "2.12.1"
-uuid = { version = "1.17.0", features = ["serde", "v4"] }
+uuid = { version = "1.18.1", features = ["serde", "v4"] }
 backoff = "0.4.0"
+reqwest = { version = "0.12.24", features = ["rustls-tls", "http2"], default-features = false }
+rusty-s3 = "0.8.1"
+tokio = { version = "1.48.0", features = ["full"] }
 
 [dev-dependencies]
 big_s = "1.0.2"

crates/index-scheduler/src/dump.rs

@@ -150,6 +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,

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/features.rs

@@ -1,9 +1,9 @@
 use std::sync::{Arc, RwLock};
 
-use meilisearch_types::enterprise_edition::network::Network;
 use meilisearch_types::features::{InstanceTogglableFeatures, RuntimeTogglableFeatures};
 use meilisearch_types::heed::types::{SerdeJson, Str};
 use meilisearch_types::heed::{Database, Env, RwTxn, WithoutTls};
+use meilisearch_types::network::Network;
 
 use crate::error::FeatureNotEnabledError;
 use crate::Result;

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

@@ -220,9 +220,11 @@ impl IndexMap {
         uuid: &Uuid,
         enable_mdb_writemap: bool,
         map_size_growth: usize,
-    ) -> Option<EnvClosingEvent> {
-        let index = self.available.remove(uuid)?;
-        Some(self.close(*uuid, index, enable_mdb_writemap, map_size_growth))
+    ) {
+        let Some(index) = self.available.remove(uuid) else {
+            return;
+        };
+        self.close(*uuid, index, enable_mdb_writemap, map_size_growth);
     }
 
     fn close(
@@ -231,21 +233,14 @@ impl IndexMap {
         index: Index,
         enable_mdb_writemap: bool,
         map_size_growth: usize,
-    ) -> EnvClosingEvent {
+    ) {
         let map_size = index.map_size() + map_size_growth;
         let closing_event = index.prepare_for_closing();
         let generation = self.next_generation();
         self.unavailable.insert(
             uuid,
-            Some(ClosingIndex {
-                uuid,
-                closing_event: closing_event.clone(),
-                enable_mdb_writemap,
-                map_size,
-                generation,
-            }),
+            Some(ClosingIndex { uuid, closing_event, enable_mdb_writemap, map_size, generation }),
         );
-        closing_event
     }
 
     /// Attempts to delete and index.

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

@@ -4,7 +4,7 @@ use std::time::Duration;
 use std::{fs, thread};
 
 use meilisearch_types::heed::types::{SerdeJson, Str};
-use meilisearch_types::heed::{Database, Env, EnvClosingEvent, RoTxn, RwTxn, WithoutTls};
+use meilisearch_types::heed::{Database, Env, RoTxn, RwTxn, WithoutTls};
 use meilisearch_types::milli;
 use meilisearch_types::milli::database_stats::DatabaseStats;
 use meilisearch_types::milli::index::RollbackOutcome;
@@ -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"))
@@ -349,14 +349,16 @@ impl IndexMapper {
     ///
     /// - 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<Option<EnvClosingEvent>> {
+    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.
-        Ok(self.index_map.write().unwrap().close_for_resize(&uuid, self.enable_mdb_writemap, 0))
+        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.
@@ -406,7 +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)
@@ -423,7 +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(
@@ -450,6 +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,
@@ -490,7 +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

@@ -6,7 +6,7 @@ use meilisearch_types::heed::types::{SerdeBincode, SerdeJson, Str};
 use meilisearch_types::heed::{Database, RoTxn};
 use meilisearch_types::milli::{CboRoaringBitmapCodec, RoaringBitmapCodec, BEU32};
 use meilisearch_types::tasks::{Details, Kind, Status, Task};
-use meilisearch_types::versioning;
+use meilisearch_types::versioning::{self, VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
 use roaring::RoaringBitmap;
 
 use crate::index_mapper::IndexMapper;
@@ -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
@@ -315,7 +320,11 @@ fn snapshot_details(d: &Details) -> String {
             format!("{{ url: {url:?}, api_key: {api_key:?}, payload_size: {payload_size:?}, indexes: {indexes:?} }}")
         }
         Details::UpgradeDatabase { from, to } => {
-            format!("{{ from: {from:?}, to: {to:?} }}")
+            if to == &(VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH) {
+                format!("{{ from: {from:?}, to: [current version] }}")
+            } else {
+                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:?} }}")
@@ -395,7 +404,21 @@ pub fn snapshot_batch(batch: &Batch) -> String {
 
     snap.push('{');
     snap.push_str(&format!("uid: {uid}, "));
-    snap.push_str(&format!("details: {}, ", serde_json::to_string(details).unwrap()));
+    let details = if let Some(upgrade_to) = &details.upgrade_to {
+        if upgrade_to.as_str()
+            == format!("v{VERSION_MAJOR}.{VERSION_MINOR}.{VERSION_PATCH}").as_str()
+        {
+            let mut details = details.clone();
+
+            details.upgrade_to = Some("[current version]".into());
+            serde_json::to_string(&details).unwrap()
+        } else {
+            serde_json::to_string(details).unwrap()
+        }
+    } else {
+        serde_json::to_string(details).unwrap()
+    };
+    snap.push_str(&format!("details: {details}, "));
     snap.push_str(&format!("stats: {}, ", serde_json::to_string(&stats).unwrap()));
     if !embedder_stats.skip_serializing() {
         snap.push_str(&format!(

crates/index-scheduler/src/lib.rs

@@ -54,7 +54,6 @@ pub use features::RoFeatures;
 use flate2::bufread::GzEncoder;
 use flate2::Compression;
 use meilisearch_types::batches::Batch;
-use meilisearch_types::enterprise_edition::network::Network;
 use meilisearch_types::features::{
     ChatCompletionSettings, InstanceTogglableFeatures, RuntimeTogglableFeatures,
 };
@@ -67,6 +66,7 @@ use meilisearch_types::milli::vector::{
     Embedder, EmbedderOptions, RuntimeEmbedder, RuntimeEmbedders, RuntimeFragment,
 };
 use meilisearch_types::milli::{self, Index};
+use meilisearch_types::network::Network;
 use meilisearch_types::task_view::TaskView;
 use meilisearch_types::tasks::{KindWithContent, Task, TaskNetwork};
 use meilisearch_types::webhooks::{Webhook, WebhooksDumpView, WebhooksView};
@@ -216,6 +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 {
@@ -242,6 +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(),
         }
     }
 
@@ -255,13 +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)?;
@@ -316,8 +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,
@@ -333,15 +346,32 @@ 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)
@@ -726,6 +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())
@@ -736,7 +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).

crates/index-scheduler/src/processing.rs

@@ -75,6 +75,7 @@ make_enum_progress! {
     pub enum TaskCancelationProgress {
         RetrievingTasks,
         CancelingUpgrade,
+        CleaningCompactionLeftover,
         UpdatingTasks,
     }
 }
@@ -145,6 +146,7 @@ make_enum_progress! {
         CopyAndCompactTheIndex,
         PersistTheCompactedIndex,
         CloseTheIndex,
+        ReopenTheIndex,
     }
 }
 

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.
@@ -344,6 +346,7 @@ impl Queue {
                 tasks: to_delete,
             },
             None,
+            None,
             false,
         )?;
 

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

@@ -25,7 +25,6 @@ enum AutobatchKind {
     IndexDeletion,
     IndexUpdate,
     IndexSwap,
-    IndexCompaction,
 }
 
 impl AutobatchKind {
@@ -69,14 +68,14 @@ impl From<KindWithContent> for AutobatchKind {
             KindWithContent::IndexCreation { .. } => AutobatchKind::IndexCreation,
             KindWithContent::IndexUpdate { .. } => AutobatchKind::IndexUpdate,
             KindWithContent::IndexSwap { .. } => AutobatchKind::IndexSwap,
-            KindWithContent::IndexCompaction { .. } => AutobatchKind::IndexCompaction,
-            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.")
             }
         }
     }
@@ -120,9 +119,6 @@ pub enum BatchKind {
     IndexSwap {
         id: TaskId,
     },
-    IndexCompaction {
-        id: TaskId,
-    },
 }
 
 impl BatchKind {
@@ -188,13 +184,6 @@ impl BatchKind {
                 )),
                 false,
             ),
-            K::IndexCompaction => (
-                Break((
-                    BatchKind::IndexCompaction { id: task_id },
-                    BatchStopReason::TaskCannotBeBatched { kind, id: task_id },
-                )),
-                false,
-            ),
             K::DocumentClear => (Continue(BatchKind::DocumentClear { ids: vec![task_id] }), false),
             K::DocumentImport { allow_index_creation, primary_key: pk }
                 if primary_key.is_none() || pk.is_none() || primary_key == pk.as_deref() =>
@@ -300,7 +289,7 @@ impl BatchKind {
 
         match (self, autobatch_kind) {
             // We don't batch any of these operations
-            (this, K::IndexCreation | K::IndexUpdate | K::IndexSwap | K::DocumentEdition | K::IndexCompaction) => {
+            (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.
@@ -497,7 +486,6 @@ impl BatchKind {
                 | BatchKind::IndexDeletion { .. }
                 | BatchKind::IndexUpdate { .. }
                 | BatchKind::IndexSwap { .. }
-                | BatchKind::IndexCompaction { .. }
                 | BatchKind::DocumentEdition { .. },
                 _,
             ) => {

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

@@ -437,12 +437,6 @@ impl IndexScheduler {
                 current_batch.processing(Some(&mut task));
                 Ok(Some(Batch::IndexSwap { task }))
             }
-            BatchKind::IndexCompaction { id } => {
-                let mut task =
-                    self.queue.tasks.get_task(rtxn, id)?.ok_or(Error::CorruptedTaskQueue)?;
-                current_batch.processing(Some(&mut task));
-                Ok(Some(Batch::IndexCompaction { index_uid, task }))
-            }
         }
     }
 
@@ -525,17 +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)?;
@@ -545,7 +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 =
@@ -558,7 +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

@@ -1,5 +1,6 @@
 use std::collections::{BTreeSet, HashMap, HashSet};
-use std::io::{Seek, SeekFrom};
+use std::fs::{remove_file, File};
+use std::io::{ErrorKind, Seek, SeekFrom};
 use std::panic::{catch_unwind, AssertUnwindSafe};
 use std::sync::atomic::Ordering;
 
@@ -13,7 +14,7 @@ use meilisearch_types::tasks::{Details, IndexSwap, Kind, KindWithContent, Status
 use meilisearch_types::versioning::{VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
 use milli::update::Settings as MilliSettings;
 use roaring::RoaringBitmap;
-use tempfile::PersistError;
+use tempfile::{PersistError, TempPath};
 use time::OffsetDateTime;
 
 use super::create_batch::Batch;
@@ -28,6 +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.
@@ -460,6 +464,7 @@ impl IndexScheduler {
                     *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 } => {
@@ -551,12 +556,18 @@ impl IndexScheduler {
         // 2. We retrieve the index and create a temporary file in the index directory
         progress.update_progress(IndexCompaction::RetrieveTheIndex);
         let index = self.index_mapper.index(rtxn, index_uid)?;
+
+        // the index operation can take a long time, so save this handle to make it available to the search for the duration of the tick
+        self.index_mapper
+            .set_currently_updating_index(Some((index_uid.to_string(), index.clone())));
+
         progress.update_progress(IndexCompaction::CreateTemporaryFile);
-        let pre_size = std::fs::metadata(index.path().join("data.mdb"))?.len();
-        let mut file = tempfile::Builder::new()
-            .suffix("data.")
-            .prefix(".mdb.cpy")
-            .tempfile_in(index.path())?;
+        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);
@@ -568,23 +579,51 @@ impl IndexScheduler {
 
         // 4. We replace the index data file with the temporary file
         progress.update_progress(IndexCompaction::PersistTheCompactedIndex);
-        let post_size = file.as_file().metadata()?.len();
-        match file.persist(index.path().join("data.mdb")) {
+        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 and wait for it
-        // The next time the index is opened, it will use the new compacted data
-        let closing_event = self.index_mapper.close_index(rtxn, index_uid)?;
+        // 5. Prepare to close the index
+        progress.update_progress(IndexCompaction::CloseTheIndex);
 
-        if let Some(closing_event) = closing_event {
-            progress.update_progress(IndexCompaction::CloseTheIndex);
-            drop(index);
-            closing_event.wait();
-        }
+        // 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))
     }
@@ -876,9 +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)?;
@@ -943,7 +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_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 +230,407 @@ 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;
+    use std::io;
+    use std::os::fd::OwnedFd;
+    use std::path::PathBuf;
+
+    use bytes::{Bytes, BytesMut};
+    use reqwest::{Client, Response};
+    use rusty_s3::actions::CreateMultipartUpload;
+    use rusty_s3::{Bucket, BucketError, Credentials, S3Action as _, 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
+
+            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() => {
+                    Err(backoff::Error::Permanent(Error::S3Error {
+                        status: resp.status(),
+                        body: resp.text().await.unwrap_or_default(),
+                    }))
+                }
+                Ok(resp) => Ok(resp),
+                Err(e) => Err(backoff::Error::transient(Error::S3HttpError(e))),
+            }
+        }
+    })
+    .await?;
+
+    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, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: [current version] }, kind: UpgradeDatabase { from: (1, 12, 0) }}
 1 {uid: 1, batch_uid: 1, status: succeeded, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
 2 {uid: 2, batch_uid: 2, status: succeeded, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
 3 {uid: 3, batch_uid: 3, status: failed, error: ResponseError { code: 200, message: "Index `doggo` already exists.", error_code: "index_already_exists", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#index_already_exists" }, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
@@ -57,7 +57,7 @@ girafo: { number_of_documents: 0, field_distribution: {} }
 [timestamp] [4,]
 ----------------------------------------------------------------------
 ### All Batches:
-0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"[current version]"}, 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, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: [current version] }, 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, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: [current version] }, 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, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: [current version] }, kind: UpgradeDatabase { from: (1, 12, 0) }}
 1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
 ----------------------------------------------------------------------
 ### Status:
@@ -37,7 +37,7 @@ catto [1,]
 [timestamp] [0,]
 ----------------------------------------------------------------------
 ### All Batches:
-0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"[current version]"}, 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, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: [current version] }, kind: UpgradeDatabase { from: (1, 12, 0) }}
 1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
 2 {uid: 2, status: enqueued, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
 ----------------------------------------------------------------------
@@ -40,7 +40,7 @@ doggo [2,]
 [timestamp] [0,]
 ----------------------------------------------------------------------
 ### All Batches:
-0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"[current version]"}, 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, 22, 1) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
+0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: [current version] }, kind: UpgradeDatabase { from: (1, 12, 0) }}
 1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
 2 {uid: 2, status: enqueued, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
 3 {uid: 3, status: enqueued, details: { primary_key: Some("bone"), old_new_uid: None, new_index_uid: None }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
@@ -43,7 +43,7 @@ doggo [2,3,]
 [timestamp] [0,]
 ----------------------------------------------------------------------
 ### All Batches:
-0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.22.1"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "stopped after the last task of type `upgradeDatabase` because they cannot be batched with tasks of any other type.", }
+0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"[current version]"}, 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/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

@@ -1,7 +1,7 @@
 use anyhow::bail;
 use meilisearch_types::heed::{Env, RwTxn, WithoutTls};
 use meilisearch_types::tasks::{Details, KindWithContent, Status, Task};
-use meilisearch_types::versioning::{VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
+use meilisearch_types::versioning;
 use time::OffsetDateTime;
 use tracing::info;
 
@@ -9,77 +9,82 @@ use crate::queue::TaskQueue;
 use crate::versioning::Versioning;
 
 trait UpgradeIndexScheduler {
-    fn upgrade(
-        &self,
-        env: &Env<WithoutTls>,
-        wtxn: &mut RwTxn,
-        original: (u32, u32, u32),
-    ) -> anyhow::Result<()>;
-    fn target_version(&self) -> (u32, u32, u32);
+    fn upgrade(&self, env: &Env<WithoutTls>, wtxn: &mut RwTxn) -> anyhow::Result<()>;
+    /// Whether the migration should be applied, depending on the initial version of the index scheduler before
+    /// any migration was applied
+    fn must_upgrade(&self, initial_version: (u32, u32, u32)) -> bool;
+    /// A progress-centric description of the migration
+    fn description(&self) -> &'static str;
 }
 
+/// Upgrade the index scheduler to the binary version.
+///
+/// # Warning
+///
+/// The current implementation uses a single wtxn to the index scheduler for the whole duration of the upgrade.
+/// If migrations start taking take a long time, it might prevent tasks from being registered.
+/// If this issue manifests, then it can be mitigated by adding a `fn target_version` to `UpgradeIndexScheduler`,
+/// to be able to write intermediate versions and drop the wtxn between applying migrations.
 pub fn upgrade_index_scheduler(
     env: &Env<WithoutTls>,
     versioning: &Versioning,
-    from: (u32, u32, u32),
-    to: (u32, u32, u32),
+    initial_version: (u32, u32, u32),
 ) -> anyhow::Result<()> {
-    let current_major = to.0;
-    let current_minor = to.1;
-    let current_patch = to.2;
+    let target_major: u32 = versioning::VERSION_MAJOR;
+    let target_minor: u32 = versioning::VERSION_MINOR;
+    let target_patch: u32 = versioning::VERSION_PATCH;
+    let target_version = (target_major, target_minor, target_patch);
 
-    let upgrade_functions: &[&dyn UpgradeIndexScheduler] = &[
-        // This is the last upgrade function, it will be called when the index is up to date.
-        // any other upgrade function should be added before this one.
-        &ToCurrentNoOp {},
-    ];
-
-    let start = match from {
-        (1, 12, _) => 0,
-        (1, 13, _) => 0,
-        (1, 14, _) => 0,
-        (1, 15, _) => 0,
-        (1, 16, _) => 0,
-        (1, 17, _) => 0,
-        (1, 18, _) => 0,
-        (1, 19, _) => 0,
-        (1, 20, _) => 0,
-        (1, 21, _) => 0,
-        (1, 22, _) => 0,
-        (major, minor, patch) => {
-            if major > current_major
-                || (major == current_major && minor > current_minor)
-                || (major == current_major && minor == current_minor && patch > current_patch)
-            {
-                bail!(
-                "Database version {major}.{minor}.{patch} is higher than the Meilisearch version {current_major}.{current_minor}.{current_patch}. Downgrade is not supported",
-                );
-            } else if major < 1 || (major == current_major && minor < 12) {
-                bail!(
-                "Database version {major}.{minor}.{patch} is too old for the experimental dumpless upgrade feature. Please generate a dump using the v{major}.{minor}.{patch} and import it in the v{current_major}.{current_minor}.{current_patch}",
-            );
-            } else {
-                bail!("Unknown database version: v{major}.{minor}.{patch}");
-            }
-        }
-    };
-
-    info!("Upgrading the task queue");
-    let mut local_from = from;
-    for upgrade in upgrade_functions[start..].iter() {
-        let target = upgrade.target_version();
-        info!(
-            "Upgrading from v{}.{}.{} to v{}.{}.{}",
-            local_from.0, local_from.1, local_from.2, target.0, target.1, target.2
-        );
-        let mut wtxn = env.write_txn()?;
-        upgrade.upgrade(env, &mut wtxn, local_from)?;
-        versioning.set_version(&mut wtxn, target)?;
-        wtxn.commit()?;
-        local_from = target;
+    if initial_version == target_version {
+        return Ok(());
     }
 
+    let upgrade_functions: &[&dyn UpgradeIndexScheduler] = &[
+        // List all upgrade functions to apply in order here.
+    ];
+
+    let (initial_major, initial_minor, initial_patch) = initial_version;
+
+    if initial_version > target_version {
+        bail!(
+                "Database version {initial_major}.{initial_minor}.{initial_patch} is higher than the Meilisearch version {target_major}.{target_minor}.{target_patch}. Downgrade is not supported",
+            );
+    }
+
+    if initial_version < (1, 12, 0) {
+        bail!(
+                "Database version {initial_major}.{initial_minor}.{initial_patch} is too old for the experimental dumpless upgrade feature. Please generate a dump using the v{initial_major}.{initial_minor}.{initial_patch} and import it in the v{target_major}.{target_minor}.{target_patch}",
+            );
+    }
+
+    info!("Upgrading the task queue");
     let mut wtxn = env.write_txn()?;
+    let migration_count = upgrade_functions.len();
+    for (migration_index, upgrade) in upgrade_functions.iter().enumerate() {
+        if upgrade.must_upgrade(initial_version) {
+            info!(
+                "[{migration_index}/{migration_count}]Applying migration: {}",
+                upgrade.description()
+            );
+
+            upgrade.upgrade(env, &mut wtxn)?;
+
+            info!(
+                "[{}/{migration_count}]Migration applied: {}",
+                migration_index + 1,
+                upgrade.description()
+            )
+        } else {
+            info!(
+                "[{migration_index}/{migration_count}]Skipping unnecessary migration: {}",
+                upgrade.description()
+            )
+        }
+    }
+
+    versioning.set_version(&mut wtxn, target_version)?;
+    info!("Task queue upgraded, spawning the upgrade database task");
+
     let queue = TaskQueue::new(env, &mut wtxn)?;
     let uid = queue.next_task_id(&wtxn)?;
     queue.register(
@@ -92,31 +97,14 @@ pub fn upgrade_index_scheduler(
             finished_at: None,
             error: None,
             canceled_by: None,
-            details: Some(Details::UpgradeDatabase { from, to }),
+            details: Some(Details::UpgradeDatabase { from: initial_version, to: target_version }),
             status: Status::Enqueued,
-            kind: KindWithContent::UpgradeDatabase { from },
+            kind: KindWithContent::UpgradeDatabase { from: initial_version },
             network: None,
+            custom_metadata: None,
         },
     )?;
     wtxn.commit()?;
 
     Ok(())
 }
-
-#[allow(non_camel_case_types)]
-struct ToCurrentNoOp {}
-
-impl UpgradeIndexScheduler for ToCurrentNoOp {
-    fn upgrade(
-        &self,
-        _env: &Env<WithoutTls>,
-        _wtxn: &mut RwTxn,
-        _original: (u32, u32, u32),
-    ) -> anyhow::Result<()> {
-        Ok(())
-    }
-
-    fn target_version(&self) -> (u32, u32, u32) {
-        (VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH)
-    }
-}

crates/index-scheduler/src/utils.rs

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

crates/index-scheduler/src/versioning.rs

@@ -64,14 +64,7 @@ impl Versioning {
         };
         wtxn.commit()?;
 
-        let bin_major: u32 = versioning::VERSION_MAJOR;
-        let bin_minor: u32 = versioning::VERSION_MINOR;
-        let bin_patch: u32 = versioning::VERSION_PATCH;
-        let to = (bin_major, bin_minor, bin_patch);
-
-        if from != to {
-            upgrade_index_scheduler(env, &this, from, to)?;
-        }
+        upgrade_index_scheduler(env, &this, from)?;
 
         // Once we reach this point it means the upgrade process, if there was one is entirely finished
         // we can safely say we reached the latest version of the index scheduler

crates/json-depth-checker/Cargo.toml

@@ -15,7 +15,7 @@ license.workspace = true
 serde_json = "1.0"
 
 [dev-dependencies]
-criterion = "0.6.0"
+criterion = "0.7.0"
 
 [[bench]]
 name = "depth"

crates/meili-snap/Cargo.toml

@@ -13,7 +13,7 @@ license.workspace = true
 [dependencies]
 # fixed version due to format breakages in v1.40
 insta = { version = "=1.39.0", features = ["json", "redactions"] }
-md5 = "0.7.0"
+md5 = "0.8.0"
 once_cell = "1.21"
-regex-lite = "0.1.6"
-uuid = { version = "1.17.0", features = ["v4"] }
+regex-lite = "0.1.8"
+uuid = { version = "1.18.1", features = ["v4"] }

crates/meilisearch-auth/Cargo.toml

@@ -12,15 +12,15 @@ license.workspace = true
 
 [dependencies]
 base64 = "0.22.1"
-enum-iterator = "2.1.0"
+enum-iterator = "2.3.0"
 hmac = "0.12.1"
 maplit = "1.0.2"
 meilisearch-types = { path = "../meilisearch-types" }
 rand = "0.8.5"
 roaring = { version = "0.10.12", features = ["serde"] }
-serde = { version = "1.0.219", features = ["derive"] }
-serde_json = { version = "1.0.140", features = ["preserve_order"] }
+serde = { version = "1.0.228", features = ["derive"] }
+serde_json = { version = "1.0.145", features = ["preserve_order"] }
 sha2 = "0.10.9"
-thiserror = "2.0.12"
-time = { version = "0.3.41", features = ["serde-well-known", "formatting", "parsing", "macros"] }
-uuid = { version = "1.17.0", features = ["serde", "v4"] }
+thiserror = "2.0.17"
+time = { version = "0.3.44", features = ["serde-well-known", "formatting", "parsing", "macros"] }
+uuid = { version = "1.18.1", features = ["serde", "v4"] }

crates/meilisearch-types/Cargo.toml

@@ -11,38 +11,38 @@ edition.workspace = true
 license.workspace = true
 
 [dependencies]
-actix-web = { version = "4.11.0", default-features = false }
-anyhow = "1.0.98"
-bumpalo = "3.18.1"
+actix-web = { version = "4.12.0", default-features = false }
+anyhow = "1.0.100"
+bumpalo = "3.19.0"
 bumparaw-collections = "0.1.4"
 byte-unit = { version = "5.1.6", features = ["serde"] }
-convert_case = "0.8.0"
-csv = "1.3.1"
-deserr = { version = "0.6.3", features = ["actix-web"] }
+convert_case = "0.9.0"
+csv = "1.4.0"
+deserr = { version = "0.6.4", features = ["actix-web"] }
 either = { version = "1.15.0", features = ["serde"] }
-enum-iterator = "2.1.0"
+enum-iterator = "2.3.0"
 file-store = { path = "../file-store" }
-flate2 = "1.1.2"
+flate2 = "1.1.5"
 fst = "0.4.7"
-memmap2 = "0.9.7"
+memmap2 = "0.9.9"
 milli = { path = "../milli" }
 roaring = { version = "0.10.12", features = ["serde"] }
 rustc-hash = "2.1.1"
-serde = { version = "1.0.219", features = ["derive"] }
+serde = { version = "1.0.228", features = ["derive"] }
 serde-cs = "0.2.4"
-serde_json = { version = "1.0.140", features = ["preserve_order"] }
+serde_json = { version = "1.0.145", features = ["preserve_order"] }
 tar = "0.4.44"
-tempfile = "3.20.0"
-thiserror = "2.0.12"
-time = { version = "0.3.41", features = [
+tempfile = "3.23.0"
+thiserror = "2.0.17"
+time = { version = "0.3.44", features = [
     "serde-well-known",
     "formatting",
     "parsing",
     "macros",
 ] }
-tokio = "1.45"
+tokio = "1.48"
 utoipa = { version = "5.4.0", features = ["macros"] }
-uuid = { version = "1.17.0", features = ["serde", "v4"] }
+uuid = { version = "1.18.1", features = ["serde", "v4"] }
 
 [dev-dependencies]
 # fixed version due to format breakages in v1.40
@@ -56,6 +56,9 @@ all-tokenizations = ["milli/all-tokenizations"]
 # chinese specialized tokenization
 chinese = ["milli/chinese"]
 chinese-pinyin = ["milli/chinese-pinyin"]
+
+enterprise = ["milli/enterprise"]
+
 # hebrew specialized tokenization
 hebrew = ["milli/hebrew"]
 # japanese specialized tokenization

crates/meilisearch-types/src/community_edition.rs

@@ -0,0 +1,16 @@
+pub mod network {
+    use milli::update::new::indexer::current_edition::sharding::Shards;
+
+    use crate::network::Network;
+
+    impl Network {
+        pub fn shards(&self) -> Option<Shards> {
+            None
+        }
+
+        pub fn sharding(&self) -> bool {
+            // always false in CE
+            false
+        }
+    }
+}

crates/meilisearch-types/src/enterprise_edition/network.rs

@@ -3,21 +3,9 @@
 // Use of this source code is governed by the Business Source License 1.1,
 // as found in the LICENSE-EE file or at <https://mariadb.com/bsl11>
 
-use std::collections::BTreeMap;
-
 use milli::update::new::indexer::enterprise_edition::sharding::Shards;
-use serde::{Deserialize, Serialize};
 
-#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, Eq, Default)]
-#[serde(rename_all = "camelCase")]
-pub struct Network {
-    #[serde(default, rename = "self")]
-    pub local: Option<String>,
-    #[serde(default)]
-    pub remotes: BTreeMap<String, Remote>,
-    #[serde(default)]
-    pub sharding: bool,
-}
+use crate::network::Network;
 
 impl Network {
     pub fn shards(&self) -> Option<Shards> {
@@ -34,14 +22,8 @@ impl Network {
             None
         }
     }
-}
 
-#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, Eq)]
-#[serde(rename_all = "camelCase")]
-pub struct Remote {
-    pub url: String,
-    #[serde(default)]
-    pub search_api_key: Option<String>,
-    #[serde(default)]
-    pub write_api_key: Option<String>,
+    pub fn sharding(&self) -> bool {
+        self.sharding
+    }
 }

crates/meilisearch-types/src/error.rs

@@ -254,10 +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 ;
@@ -315,6 +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 ;
@@ -390,6 +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 ;
@@ -426,6 +433,7 @@ InvalidChatCompletionSearchQueryParamPrompt    , InvalidRequest       , BAD_REQU
 InvalidChatCompletionSearchFilterParamPrompt   , InvalidRequest       , BAD_REQUEST ;
 InvalidChatCompletionSearchIndexUidParamPrompt , InvalidRequest       , BAD_REQUEST ;
 InvalidChatCompletionPreQueryPrompt            , InvalidRequest       , BAD_REQUEST ;
+RequiresEnterpriseEdition                      , InvalidRequest       , UNAVAILABLE_FOR_LEGAL_REASONS ;
 // Webhooks
 InvalidWebhooks                                , InvalidRequest       , BAD_REQUEST ;
 InvalidWebhookUrl                              , InvalidRequest       , BAD_REQUEST ;
@@ -679,6 +687,18 @@ 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/lib.rs

@@ -2,10 +2,17 @@
 
 pub mod batch_view;
 pub mod batches;
+#[cfg(not(feature = "enterprise"))]
+pub mod community_edition;
 pub mod compression;
 pub mod deserr;
 pub mod document_formats;
+#[cfg(feature = "enterprise")]
 pub mod enterprise_edition;
+#[cfg(not(feature = "enterprise"))]
+pub use community_edition as current_edition;
+#[cfg(feature = "enterprise")]
+pub use enterprise_edition as current_edition;
 pub mod error;
 pub mod facet_values_sort;
 pub mod features;
@@ -13,6 +20,7 @@ pub mod index_uid;
 pub mod index_uid_pattern;
 pub mod keys;
 pub mod locales;
+pub mod network;
 pub mod settings;
 pub mod star_or;
 pub mod task_view;

crates/meilisearch-types/src/network.rs

@@ -0,0 +1,24 @@
+use std::collections::BTreeMap;
+
+use serde::{Deserialize, Serialize};
+
+#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, Eq, Default)]
+#[serde(rename_all = "camelCase")]
+pub struct Network {
+    #[serde(default, rename = "self")]
+    pub local: Option<String>,
+    #[serde(default)]
+    pub remotes: BTreeMap<String, Remote>,
+    #[serde(default)]
+    pub sharding: bool,
+}
+
+#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, Eq)]
+#[serde(rename_all = "camelCase")]
+pub struct Remote {
+    pub url: String,
+    #[serde(default)]
+    pub search_api_key: Option<String>,
+    #[serde(default)]
+    pub write_api_key: Option<String>,
+}

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(),
         }
     }
 }
@@ -325,7 +329,7 @@ impl DetailsView {
             ) {
                 (None, None) => None,
                 (None, Some(size)) | (Some(size), None) => Some(size),
-                // We should never be able to batch multiple renames at the same time.
+                // We should never be able to batch multiple compactions at the same time.
                 (Some(left), Some(_right)) => Some(left),
             },
             post_compaction_size: match (
@@ -334,7 +338,7 @@ impl DetailsView {
             ) {
                 (None, None) => None,
                 (None, Some(size)) | (Some(size), None) => Some(size),
-                // We should never be able to batch multiple renames at the same time.
+                // We should never be able to batch multiple compactions at the same time.
                 (Some(left), Some(_right)) => Some(left),
             },
         }

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 {

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

@@ -14,91 +14,91 @@ default-run = "meilisearch"
 
 [dependencies]
 actix-cors = "0.7.1"
-actix-http = { version = "3.11.0", default-features = false, features = [
+actix-http = { version = "3.11.2", default-features = false, features = [
     "compress-brotli",
     "compress-gzip",
     "rustls-0_23",
 ] }
 actix-utils = "3.0.1"
-actix-web = { version = "4.11.0", default-features = false, features = [
+actix-web = { version = "4.12.0", default-features = false, features = [
     "macros",
     "compress-brotli",
     "compress-gzip",
     "cookies",
     "rustls-0_23",
 ] }
-anyhow = { version = "1.0.98", features = ["backtrace"] }
-bstr = "1.12.0"
+anyhow = { version = "1.0.100", features = ["backtrace"] }
+bstr = "1.12.1"
 byte-unit = { version = "5.1.6", features = ["serde"] }
-bytes = "1.10.1"
-bumpalo = "3.18.1"
-clap = { version = "4.5.40", features = ["derive", "env"] }
+bytes = "1.11.0"
+bumpalo = "3.19.0"
+clap = { version = "4.5.52", features = ["derive", "env"] }
 crossbeam-channel = "0.5.15"
-deserr = { version = "0.6.3", features = ["actix-web"] }
+deserr = { version = "0.6.4", features = ["actix-web"] }
 dump = { path = "../dump" }
 either = "1.15.0"
 file-store = { path = "../file-store" }
-flate2 = "1.1.2"
+flate2 = "1.1.5"
 fst = "0.4.7"
 futures = "0.3.31"
 futures-util = "0.3.31"
 index-scheduler = { path = "../index-scheduler" }
-indexmap = { version = "2.9.0", features = ["serde"] }
-is-terminal = "0.4.16"
+indexmap = { version = "2.12.0", features = ["serde"] }
+is-terminal = "0.4.17"
 itertools = "0.14.0"
 jsonwebtoken = "9.3.1"
 lazy_static = "1.5.0"
 meilisearch-auth = { path = "../meilisearch-auth" }
 meilisearch-types = { path = "../meilisearch-types" }
-memmap2 = "0.9.7"
-mimalloc = { version = "0.1.47", default-features = false }
+memmap2 = "0.9.9"
+mimalloc = { version = "0.1.48", default-features = false }
 mime = "0.3.17"
 num_cpus = "1.17.0"
 obkv = "0.3.0"
 once_cell = "1.21.3"
-ordered-float = "5.0.0"
-parking_lot = "0.12.4"
+ordered-float = "5.1.0"
+parking_lot = "0.12.5"
 permissive-json-pointer = { path = "../permissive-json-pointer" }
 pin-project-lite = "0.2.16"
 platform-dirs = "0.3.0"
 prometheus = { version = "0.14.0", features = ["process"] }
 rand = "0.8.5"
-rayon = "1.10.0"
-regex = "1.11.1"
-reqwest = { version = "0.12.20", features = [
+rayon = "1.11.0"
+regex = "1.12.2"
+reqwest = { version = "0.12.24", features = [
     "rustls-tls",
     "json",
 ], default-features = false }
-rustls = { version = "0.23.28", features = ["ring"], default-features = false }
-rustls-pki-types = { version = "1.12.0", features = ["alloc"] }
+rustls = { version = "0.23.35", features = ["ring"], default-features = false }
+rustls-pki-types = { version = "1.13.0", features = ["alloc"] }
 rustls-pemfile = "2.2.0"
 segment = { version = "0.2.6" }
-serde = { version = "1.0.219", features = ["derive"] }
-serde_json = { version = "1.0.140", features = ["preserve_order"] }
+serde = { version = "1.0.228", features = ["derive"] }
+serde_json = { version = "1.0.145", features = ["preserve_order"] }
 sha2 = "0.10.9"
 siphasher = "1.0.1"
 slice-group-by = "0.3.1"
-static-files = { version = "0.2.5", optional = true }
-sysinfo = "0.35.2"
+static-files = { version = "0.3.1", optional = true }
+sysinfo = "0.37.2"
 tar = "0.4.44"
-tempfile = "3.20.0"
-thiserror = "2.0.12"
-time = { version = "0.3.41", features = [
+tempfile = "3.23.0"
+thiserror = "2.0.17"
+time = { version = "0.3.44", features = [
     "serde-well-known",
     "formatting",
     "parsing",
     "macros",
 ] }
-tokio = { version = "1.45.1", features = ["full"] }
-toml = "0.8.23"
-uuid = { version = "1.18.0", features = ["serde", "v4", "v7"] }
+tokio = { version = "1.48.0", features = ["full"] }
+toml = "0.9.8"
+uuid = { version = "1.18.1", features = ["serde", "v4", "v7"] }
 serde_urlencoded = "0.7.1"
 termcolor = "1.4.1"
-url = { version = "2.5.4", features = ["serde"] }
+url = { version = "2.5.7", features = ["serde"] }
 tracing = "0.1.41"
 tracing-subscriber = { version = "0.3.20", features = ["json"] }
 tracing-trace = { version = "0.1.0", path = "../tracing-trace" }
-tracing-actix-web = "0.7.18"
+tracing-actix-web = "0.7.19"
 build-info = { version = "1.7.0", path = "../build-info" }
 roaring = "0.10.12"
 mopa-maintained = "0.2.3"
@@ -114,35 +114,35 @@ utoipa = { version = "5.4.0", features = [
 utoipa-scalar = { version = "0.3.0", optional = true, features = ["actix-web"] }
 async-openai = { git = "https://github.com/meilisearch/async-openai", branch = "better-error-handling" }
 secrecy = "0.10.3"
-actix-web-lab = { version = "0.24.1", default-features = false }
+actix-web-lab = { version = "0.24.3", default-features = false }
 urlencoding = "2.1.3"
 backoff = { version = "0.4.0", features = ["tokio"] }
-
+humantime = { version = "2.3.0", default-features = false }
 
 [dev-dependencies]
-actix-rt = "2.10.0"
-brotli = "8.0.1"
+actix-rt = "2.11.0"
+brotli = "8.0.2"
 # fixed version due to format breakages in v1.40
 insta = { version = "=1.39.0", features = ["redactions"] }
 manifest-dir-macros = "0.1.18"
 maplit = "1.0.2"
 meili-snap = { path = "../meili-snap" }
 temp-env = "0.3.6"
-wiremock = "0.6.3"
+wiremock = "0.6.5"
 yaup = "0.3.1"
 
 [build-dependencies]
-anyhow = { version = "1.0.98", optional = true }
-cargo_toml = { version = "0.22.1", optional = true }
+anyhow = { version = "1.0.100", optional = true }
+cargo_toml = { version = "0.22.3", optional = true }
 hex = { version = "0.4.3", optional = true }
-reqwest = { version = "0.12.20", features = [
+reqwest = { version = "0.12.24", features = [
     "blocking",
     "rustls-tls",
 ], default-features = false, optional = true }
 sha-1 = { version = "0.10.1", optional = true }
-static-files = { version = "0.2.5", optional = true }
-tempfile = { version = "3.20.0", optional = true }
-zip = { version = "4.1.0", optional = true }
+static-files = { version = "0.3.1", optional = true }
+tempfile = { version = "3.23.0", optional = true }
+zip = { version = "6.0.0", optional = true }
 
 [features]
 default = ["meilisearch-types/all-tokenizations", "mini-dashboard"]
@@ -160,6 +160,7 @@ mini-dashboard = [
 ]
 chinese = ["meilisearch-types/chinese"]
 chinese-pinyin = ["meilisearch-types/chinese-pinyin"]
+enterprise = ["meilisearch-types/enterprise"]
 hebrew = ["meilisearch-types/hebrew"]
 japanese = ["meilisearch-types/japanese"]
 korean = ["meilisearch-types/korean"]

crates/meilisearch/src/analytics/segment_analytics.rs

@@ -195,7 +195,7 @@ struct Infos {
     experimental_enable_logs_route: bool,
     experimental_reduce_indexing_memory_usage: bool,
     experimental_max_number_of_batched_tasks: usize,
-    experimental_limit_batched_tasks_total_size: u64,
+    experimental_limit_batched_tasks_total_size: Option<u64>,
     experimental_network: bool,
     experimental_multimodal: bool,
     experimental_chat_completions: bool,
@@ -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 {
@@ -344,13 +352,14 @@ 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(),
             http_payload_size_limit,
             experimental_max_number_of_batched_tasks,
             experimental_limit_batched_tasks_total_size:
-                experimental_limit_batched_tasks_total_size.into(),
+                experimental_limit_batched_tasks_total_size.map(|size| size.as_u64()),
             task_queue_webhook: task_webhook_url.is_some(),
             task_webhook_authorization_header: task_webhook_authorization_header.is_some(),
             log_level: log_level.to_string(),
@@ -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

@@ -38,6 +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,
@@ -137,6 +139,9 @@ impl ErrorCode for MeilisearchHttpError {
             MeilisearchHttpError::InconsistentFacetOrder { .. } => {
                 Code::InvalidMultiSearchFacetOrder
             }
+            MeilisearchHttpError::PersonalizationInFederatedQuery(_) => {
+                Code::InvalidMultiSearchQueryPersonalization
+            }
             MeilisearchHttpError::InconsistentOriginHeaders { .. } => {
                 Code::InconsistentDocumentChangeHeaders
             }

crates/meilisearch/src/lib.rs

@@ -11,6 +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;
@@ -58,6 +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.
 ///
@@ -128,12 +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<
@@ -145,17 +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));
 
@@ -216,7 +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"),
@@ -230,12 +221,26 @@ 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,
         max_number_of_batched_tasks: opt.experimental_max_number_of_batched_tasks,
-        batched_tasks_size_limit: opt.experimental_limit_batched_tasks_total_size.into(),
+        batched_tasks_size_limit: opt.experimental_limit_batched_tasks_total_size.map_or_else(
+            || {
+                opt.indexer_options
+                    .max_indexing_memory
+                    // By default, we use half of the available memory to determine the size of batched tasks
+                    .map_or(u64::MAX, |mem| mem.as_u64() / 2)
+                    // And never exceed 10 GiB when we infer the limit
+                    .min(10 * 1024 * 1024 * 1024)
+            },
+            |size| size.as_u64(),
+        ),
         index_growth_amount: byte_unit::Byte::from_str("10GiB").unwrap().as_u64() as usize,
         index_count: DEFAULT_INDEX_COUNT,
         instance_features: opt.to_instance_features(),
@@ -256,6 +261,7 @@ 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)?;
@@ -273,7 +279,7 @@ 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();
@@ -284,6 +290,7 @@ 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),
@@ -304,10 +311,10 @@ 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
@@ -338,6 +345,7 @@ 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.
@@ -345,7 +353,7 @@ 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 (
@@ -452,6 +460,7 @@ 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)?
@@ -459,7 +468,7 @@ 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(
@@ -527,7 +536,11 @@ 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 {
@@ -675,23 +688,26 @@ 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()
@@ -752,3 +768,14 @@ 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

@@ -1,7 +1,8 @@
 use lazy_static::lazy_static;
 use prometheus::{
-    opts, register_gauge, register_histogram_vec, register_int_counter_vec, register_int_gauge,
-    register_int_gauge_vec, Gauge, HistogramVec, IntCounterVec, IntGauge, IntGaugeVec,
+    opts, register_gauge, register_gauge_vec, register_histogram_vec, register_int_counter_vec,
+    register_int_gauge, register_int_gauge_vec, Gauge, GaugeVec, HistogramVec, IntCounterVec,
+    IntGauge, IntGaugeVec,
 };
 
 lazy_static! {
@@ -73,6 +74,20 @@ lazy_static! {
         &["kind", "value"]
     )
     .expect("Can't create a metric");
+    pub static ref MEILISEARCH_BATCH_RUNNING_PROGRESS_TRACE: GaugeVec = register_gauge_vec!(
+        opts!("meilisearch_batch_running_progress_trace", "The currently running progress trace"),
+        &["batch_uid", "step_name"]
+    )
+    .expect("Can't create a metric");
+    pub static ref MEILISEARCH_LAST_FINISHED_BATCHES_PROGRESS_TRACE_MS: IntGaugeVec =
+        register_int_gauge_vec!(
+            opts!(
+                "meilisearch_last_finished_batches_progress_trace_ms",
+                "The last few batches progress trace in milliseconds"
+            ),
+            &["batch_uid", "step_name"]
+        )
+        .expect("Can't create a metric");
     pub static ref MEILISEARCH_LAST_UPDATE: IntGauge =
         register_int_gauge!(opts!("meilisearch_last_update", "Meilisearch Last Update"))
             .expect("Can't create a metric");
@@ -114,4 +129,9 @@ 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";
@@ -448,11 +473,14 @@ pub struct Opt {
     #[serde(default = "default_limit_batched_tasks")]
     pub experimental_max_number_of_batched_tasks: usize,
 
-    /// Experimentally reduces the maximum total size, in bytes, of tasks that will be processed at once,
-    /// see: <https://github.com/orgs/meilisearch/discussions/801>
-    #[clap(long, env = MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZE, default_value_t = default_limit_batched_tasks_total_size())]
-    #[serde(default = "default_limit_batched_tasks_total_size")]
-    pub experimental_limit_batched_tasks_total_size: Byte,
+    /// Experimentally controls the maximum total size, in bytes, of tasks that will be processed
+    /// simultaneously. When unspecified, defaults to half of the maximum indexing memory and
+    /// clamped to 10 GiB.
+    ///
+    /// See: <https://github.com/orgs/meilisearch/discussions/801>
+    #[clap(long, env = MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZE)]
+    #[serde(default)]
+    pub experimental_limit_batched_tasks_total_size: Option<Byte>,
 
     /// Enables experimental caching of search query embeddings. The value represents the maximal number of entries in the cache of each
     /// distinct embedder.
@@ -471,10 +499,20 @@ 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 +614,8 @@ 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);
@@ -664,10 +704,12 @@ impl Opt {
             MEILI_EXPERIMENTAL_MAX_NUMBER_OF_BATCHED_TASKS,
             experimental_max_number_of_batched_tasks.to_string(),
         );
-        export_to_env_if_not_present(
-            MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZE,
-            experimental_limit_batched_tasks_total_size.to_string(),
-        );
+        if let Some(limit) = experimental_limit_batched_tasks_total_size {
+            export_to_env_if_not_present(
+                MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZE,
+                limit.to_string(),
+            );
+        }
         export_to_env_if_not_present(
             MEILI_EXPERIMENTAL_EMBEDDING_CACHE_ENTRIES,
             experimental_embedding_cache_entries.to_string(),
@@ -676,7 +718,22 @@ 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 +829,22 @@ 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 +856,8 @@ 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 +883,18 @@ 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 +902,16 @@ 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 +919,163 @@ 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(),
         })
     }
 }
@@ -1035,10 +1278,6 @@ fn default_limit_batched_tasks() -> usize {
     usize::MAX
 }
 
-fn default_limit_batched_tasks_total_size() -> Byte {
-    Byte::from_u64(u64::MAX)
-}
-
 fn default_embedding_cache_entries() -> usize {
     0
 }
@@ -1051,6 +1290,22 @@ 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 std::time::Duration;
+
+use meilisearch_types::error::{Code, ErrorCode, ResponseError};
+use meilisearch_types::milli::TimeBudget;
+use rand::Rng;
+use reqwest::Client;
+use serde::{Deserialize, Serialize};
+use tracing::{debug, info, warn};
+
+use crate::search::{Personalize, SearchResult};
+
+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/community_edition.rs

@@ -0,0 +1,39 @@
+pub mod proxy {
+
+    use std::fs::File;
+
+    use actix_web::HttpRequest;
+    use index_scheduler::IndexScheduler;
+
+    use crate::error::MeilisearchHttpError;
+
+    pub enum Body<T: serde::Serialize> {
+        NdJsonPayload,
+        Inline(T),
+        None,
+    }
+
+    impl Body<()> {
+        pub fn with_ndjson_payload(_file: File) -> Self {
+            Self::NdJsonPayload
+        }
+
+        pub fn none() -> Self {
+            Self::None
+        }
+    }
+
+    pub const PROXY_ORIGIN_REMOTE_HEADER: &str = "Meili-Proxy-Origin-Remote";
+    pub const PROXY_ORIGIN_TASK_UID_HEADER: &str = "Meili-Proxy-Origin-TaskUid";
+
+    pub async fn proxy<T: serde::Serialize>(
+        _index_scheduler: &IndexScheduler,
+        _index_uid: &str,
+        _req: &HttpRequest,
+        _network: meilisearch_types::network::Network,
+        _body: Body<T>,
+        _task: &meilisearch_types::tasks::Task,
+    ) -> Result<(), MeilisearchHttpError> {
+        Ok(())
+    }
+}

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

@@ -20,7 +20,7 @@ use crate::routes::SummarizedTaskView;
     tags(
         (
             name = "Compact an index",
-            description = "The /compact route uses compacts the database to reoganize and make it smaller and more efficient.",
+            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"),
         ),
     ),
@@ -34,7 +34,7 @@ pub fn configure(cfg: &mut web::ServiceConfig) {
 /// Compact an index
 #[utoipa::path(
     post,
-    path = "{indexUid}/compact",
+    path = "/{indexUid}/compact",
     tag = "Compact an index",
     security(("Bearer" = ["search", "*"])),
     params(("indexUid" = String, Path, example = "movies", description = "Index Unique Identifier", nullable = false)),

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

@@ -45,7 +45,7 @@ use crate::extractors::authentication::policies::*;
 use crate::extractors::authentication::GuardedData;
 use crate::extractors::payload::Payload;
 use crate::extractors::sequential_extractor::SeqHandler;
-use crate::routes::indexes::enterprise_edition::proxy::{proxy, Body};
+use crate::routes::indexes::current_edition::proxy::{proxy, Body};
 use crate::routes::indexes::search::fix_sort_query_parameters;
 use crate::routes::{
     get_task_id, is_dry_run, PaginationView, SummarizedTaskView, PAGINATION_DEFAULT_LIMIT,
@@ -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();
@@ -359,10 +361,13 @@ 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 {
+    if network.sharding() && !dry_run {
         proxy(&index_scheduler, &index_uid, &req, network, Body::none(), &task).await?;
     }
 
@@ -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) => {
@@ -1075,7 +1098,7 @@ async fn document_addition(
         }
     };
 
-    if network.sharding {
+    if network.sharding() {
         if let Some(file) = file {
             proxy(
                 &index_scheduler,
@@ -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();
 
@@ -1190,10 +1216,13 @@ 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 {
+    if network.sharding() && !dry_run {
         proxy(&index_scheduler, &index_uid, &req, network, Body::Inline(body), &task).await?;
     }
 
@@ -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();
@@ -1282,10 +1314,13 @@ 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 {
+    if network.sharding() && !dry_run {
         proxy(&index_scheduler, &index_uid, &req, network, Body::Inline(filter), &task).await?;
     }
 
@@ -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?;
+    if network.sharding() && !dry_run {
+        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 {
@@ -1501,10 +1543,13 @@ 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 {
+    if network.sharding() && !dry_run {
         proxy(&index_scheduler, &index_uid, &req, network, Body::none(), &task).await?;
     }
 

crates/meilisearch/src/routes/indexes/enterprise_edition/proxy.rs

@@ -52,7 +52,7 @@ pub async fn proxy<T: serde::Serialize>(
     index_scheduler: &IndexScheduler,
     index_uid: &str,
     req: &HttpRequest,
-    network: meilisearch_types::enterprise_edition::network::Network,
+    network: meilisearch_types::network::Network,
     body: Body<T>,
     task: &meilisearch_types::tasks::Task,
 ) -> Result<(), MeilisearchHttpError> {

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

@@ -30,7 +30,16 @@ use crate::Opt;
 
 pub mod compact;
 pub mod documents;
+
+#[cfg(not(feature = "enterprise"))]
+mod community_edition;
+#[cfg(feature = "enterprise")]
 mod enterprise_edition;
+#[cfg(not(feature = "enterprise"))]
+use community_edition as current_edition;
+#[cfg(feature = "enterprise")]
+use enterprise_edition as current_edition;
+
 pub mod facet_search;
 pub mod search;
 mod search_analytics;
@@ -41,7 +50,7 @@ mod settings_analytics;
 pub mod similar;
 mod similar_analytics;
 
-pub use enterprise_edition::proxy::{PROXY_ORIGIN_REMOTE_HEADER, PROXY_ORIGIN_TASK_UID_HEADER};
+pub use current_edition::proxy::{PROXY_ORIGIN_REMOTE_HEADER, PROXY_ORIGIN_TASK_UID_HEADER};
 
 #[derive(OpenApi)]
 #[openapi(

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

@@ -22,11 +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;
 
@@ -133,6 +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)]
@@ -204,6 +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`
@@ -233,6 +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,
         })
     }
 }
@@ -321,6 +328,7 @@ 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,
@@ -341,30 +349,54 @@ 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(),
-            request_uid,
         )
     })
     .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?;
+
+    // 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))
@@ -429,6 +461,7 @@ 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,
@@ -449,26 +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(),
-            request_uid,
         )
     })
     .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();
@@ -476,7 +520,19 @@ pub async fn search_with_post(
     }
     analytics.publish(aggregate, &req);
 
-    let search_result = search_result?;
+    let (mut search_result, time_budget) = search_result?;
+
+    // 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();
@@ -235,6 +246,7 @@ impl<Method: AggregateMethod> SearchAggregator<Method> {
             degraded,
             used_negative_operator,
             request_uid: _,
+            metadata: _,
         } = result;
 
         self.total_succeeded = self.total_succeeded.saturating_add(1);
@@ -295,6 +307,7 @@ impl<Method: AggregateMethod> Aggregate for SearchAggregator<Method> {
             total_used_negative_operator,
             ranking_score_threshold,
             mut locales,
+            total_personalized,
             marker: _,
         } = *new;
 
@@ -380,6 +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
     }
 
@@ -425,6 +441,7 @@ impl<Method: AggregateMethod> Aggregate for SearchAggregator<Method> {
             total_used_negative_operator,
             ranking_score_threshold,
             locales,
+            total_personalized,
             marker: _,
         } = *self;
 
@@ -498,6 +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

@@ -4,6 +4,7 @@ use index_scheduler::{IndexScheduler, Query};
 use meilisearch_auth::AuthController;
 use meilisearch_types::error::ResponseError;
 use meilisearch_types::keys::actions;
+use meilisearch_types::milli::progress::ProgressStepView;
 use meilisearch_types::tasks::Status;
 use prometheus::{Encoder, TextEncoder};
 use time::OffsetDateTime;
@@ -38,6 +39,12 @@ pub fn configure(config: &mut web::ServiceConfig) {
 # HELP meilisearch_db_size_bytes Meilisearch DB Size In Bytes
 # TYPE meilisearch_db_size_bytes gauge
 meilisearch_db_size_bytes 1130496
+# HELP meilisearch_batch_running_progress_trace The currently running progress trace
+# TYPE meilisearch_batch_running_progress_trace gauge
+meilisearch_batch_running_progress_trace{batch_uid="0",step_name="document"} 0.710618582519409
+meilisearch_batch_running_progress_trace{batch_uid="0",step_name="extracting word proximity"} 0.2222222222222222
+meilisearch_batch_running_progress_trace{batch_uid="0",step_name="indexing"} 0.6666666666666666
+meilisearch_batch_running_progress_trace{batch_uid="0",step_name="processing tasks"} 0
 # HELP meilisearch_http_requests_total Meilisearch HTTP requests total
 # TYPE meilisearch_http_requests_total counter
 meilisearch_http_requests_total{method="GET",path="/metrics",status="400"} 1
@@ -61,6 +68,13 @@ meilisearch_http_response_time_seconds_bucket{method="GET",path="/metrics",le="1
 meilisearch_http_response_time_seconds_bucket{method="GET",path="/metrics",le="+Inf"} 0
 meilisearch_http_response_time_seconds_sum{method="GET",path="/metrics"} 0
 meilisearch_http_response_time_seconds_count{method="GET",path="/metrics"} 0
+# HELP meilisearch_last_finished_batches_progress_trace_ms The last few batches progress trace in milliseconds
+# TYPE meilisearch_last_finished_batches_progress_trace_ms gauge
+meilisearch_last_finished_batches_progress_trace_ms{batch_uid="0",step_name="processing tasks"} 19360
+meilisearch_last_finished_batches_progress_trace_ms{batch_uid="0",step_name="processing tasks > computing document changes"} 368
+meilisearch_last_finished_batches_progress_trace_ms{batch_uid="0",step_name="processing tasks > computing document changes > preparing payloads"} 367
+meilisearch_last_finished_batches_progress_trace_ms{batch_uid="0",step_name="processing tasks > computing document changes > preparing payloads > payload"} 367
+meilisearch_last_finished_batches_progress_trace_ms{batch_uid="0",step_name="processing tasks > indexing"} 18970
 # HELP meilisearch_index_count Meilisearch Index Count
 # TYPE meilisearch_index_count gauge
 meilisearch_index_count 1
@@ -148,6 +162,50 @@ pub async fn get_metrics(
         }
     }
 
+    // Fetch and expose the current progressing step
+    crate::metrics::MEILISEARCH_BATCH_RUNNING_PROGRESS_TRACE.reset();
+    let (batches, _total) = index_scheduler.get_batches_from_authorized_indexes(
+        &Query { statuses: Some(vec![Status::Processing]), ..Query::default() },
+        auth_filters,
+    )?;
+    if let Some(batch) = batches.into_iter().next() {
+        let batch_uid = batch.uid.to_string();
+        if let Some(progress) = batch.progress {
+            for ProgressStepView { current_step, finished, total } in progress.steps {
+                crate::metrics::MEILISEARCH_BATCH_RUNNING_PROGRESS_TRACE
+                    .with_label_values(&[batch_uid.as_str(), current_step.as_ref()])
+                    // We return the completion ratio of the current step
+                    .set(finished as f64 / total as f64);
+            }
+        }
+    }
+
+    crate::metrics::MEILISEARCH_LAST_FINISHED_BATCHES_PROGRESS_TRACE_MS.reset();
+    let (batches, _total) = index_scheduler.get_batches_from_authorized_indexes(
+        // Fetch the finished batches...
+        &Query {
+            statuses: Some(vec![Status::Succeeded, Status::Failed]),
+            limit: Some(1),
+            ..Query::default()
+        },
+        auth_filters,
+    )?;
+    // ...and get the last batch only.
+    if let Some(batch) = batches.into_iter().next() {
+        let batch_uid = batch.uid.to_string();
+        for (step_name, duration_str) in batch.stats.progress_trace {
+            let Some(duration_str) = duration_str.as_str() else { continue };
+            match humantime::parse_duration(duration_str) {
+                Ok(duration) => {
+                    crate::metrics::MEILISEARCH_LAST_FINISHED_BATCHES_PROGRESS_TRACE_MS
+                        .with_label_values(&[&batch_uid, &step_name])
+                        .set(duration.as_millis() as i64);
+                }
+                Err(e) => tracing::error!("Failed to parse duration: {e}"),
+            }
+        }
+    }
+
     if let Some(last_update) = response.last_update {
         crate::metrics::MEILISEARCH_LAST_UPDATE.set(last_update.unix_timestamp());
     }

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

@@ -18,10 +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;
 
@@ -145,6 +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>,
@@ -188,6 +190,7 @@ pub async fn multi_search_with_post(
         err
     })?;
 
+    let include_metadata = parse_include_metadata_header(&req);
     let response = match federation {
         Some(federation) => {
             debug!(
@@ -209,6 +212,7 @@ pub async fn multi_search_with_post(
                 features,
                 is_proxy,
                 request_uid,
+                include_metadata,
             )
             .await;
             permit.drop().await;
@@ -233,7 +237,7 @@ 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()
@@ -266,6 +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(
@@ -277,23 +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,
-                            request_uid,
                         )
                     })
                     .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)

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/network.rs

@@ -7,7 +7,6 @@ use deserr::Deserr;
 use index_scheduler::IndexScheduler;
 use itertools::{EitherOrBoth, Itertools};
 use meilisearch_types::deserr::DeserrJsonError;
-use meilisearch_types::enterprise_edition::network::{Network as DbNetwork, Remote as DbRemote};
 use meilisearch_types::error::deserr_codes::{
     InvalidNetworkRemotes, InvalidNetworkSearchApiKey, InvalidNetworkSelf, InvalidNetworkSharding,
     InvalidNetworkUrl, InvalidNetworkWriteApiKey,
@@ -15,6 +14,7 @@ use meilisearch_types::error::deserr_codes::{
 use meilisearch_types::error::ResponseError;
 use meilisearch_types::keys::actions;
 use meilisearch_types::milli::update::Setting;
+use meilisearch_types::network::{Network as DbNetwork, Remote as DbRemote};
 use serde::Serialize;
 use tracing::debug;
 use utoipa::{OpenApi, ToSchema};
@@ -211,6 +211,16 @@ async fn patch_network(
     let old_network = index_scheduler.network();
     debug!(parameters = ?new_network, "Patch network");
 
+    #[cfg(not(feature = "enterprise"))]
+    if new_network.sharding.set().is_some() {
+        use meilisearch_types::error::Code;
+
+        return Err(ResponseError::from_msg(
+            "Meilisearch Enterprise Edition is required to set `network.sharding`".into(),
+            Code::RequiresEnterpriseEdition,
+        ));
+    }
+
     let merged_self = match new_network.local {
         Setting::Set(new_self) => Some(new_self),
         Setting::Reset => None,
@@ -312,6 +322,7 @@ async fn patch_network(
 
     let merged_network =
         DbNetwork { local: merged_self, remotes: merged_remotes, sharding: merged_sharding };
+
     index_scheduler.put_network(merged_network.clone())?;
     debug!(returns = ?merged_network, "Patch network");
     Ok(HttpResponse::Ok().json(merged_network))

crates/meilisearch/src/routes/webhooks.rs

@@ -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

@@ -9,12 +9,12 @@ use std::vec::{IntoIter, Vec};
 use actix_http::StatusCode;
 use index_scheduler::{IndexScheduler, RoFeatures};
 use itertools::Itertools;
-use meilisearch_types::enterprise_edition::network::{Network, Remote};
 use meilisearch_types::error::ResponseError;
 use meilisearch_types::milli::order_by_map::OrderByMap;
 use meilisearch_types::milli::score_details::{ScoreDetails, WeightedScoreValue};
 use meilisearch_types::milli::vector::Embedding;
 use meilisearch_types::milli::{self, DocumentId, OrderBy, TimeBudget, DEFAULT_VALUES_PER_FACET};
+use meilisearch_types::network::{Network, Remote};
 use roaring::RoaringBitmap;
 use tokio::task::JoinHandle;
 use uuid::Uuid;
@@ -22,7 +22,8 @@ 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::{
@@ -41,32 +42,57 @@ pub async fn perform_federated_search(
     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 {
@@ -108,11 +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)
@@ -127,7 +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() {
@@ -146,7 +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);
 
@@ -173,6 +213,7 @@ pub async fn perform_federated_search(
         facets_by_index,
         remote_errors: partitioned_queries.has_remote.then_some(remote_errors),
         request_uid: Some(request_uid),
+        metadata: query_metadata,
     })
 }
 
@@ -396,6 +437,7 @@ struct SearchHitByIndex {
 
 struct SearchResultByIndex {
     index: String,
+    primary_key: Option<String>,
     hits: Vec<SearchHitByIndex>,
     estimated_total_hits: usize,
     degraded: bool,
@@ -403,6 +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>,
@@ -414,6 +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,
@@ -442,6 +540,7 @@ fn merge_metadata(
         degraded: degraded_for_host,
         used_negative_operator: host_used_negative_operator,
         remote_errors: _,
+        metadata: _,
         request_uid: _,
     } in remote_results
     {
@@ -502,6 +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();
@@ -570,7 +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))
@@ -590,7 +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
+                    }
                 }),
             );
         }
@@ -634,6 +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()),
@@ -729,6 +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<_>> =
@@ -755,6 +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
 
@@ -824,17 +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,
                 )?;
 
@@ -981,6 +1102,7 @@ impl SearchByIndex {
             })?;
         self.results_by_index.push(SearchResultByIndex {
             index: index_uid,
+            primary_key,
             hits: merged_result,
             estimated_total_hits,
             degraded,