Add a test reproducing the bug

4930: Return `UserError::InvalidDocumentId` for primary keys with a length greater than 512 bytes r=curquiza a=flevi29

# Pull Request

## Related issue
Fixes #4843

## PR checklist
Please check if your PR fulfills the following requirements:
- [x] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)?
- [x] Have you read the contributing guidelines?
- [x] Have you made sure that the title is accurate and descriptive of the changes?

Thank you so much for contributing to Meilisearch!


Co-authored-by: F. Levi <55688616+flevi29@users.noreply.github.com>

.github/workflows/bench-manual.yml

@@ -18,7 +18,7 @@ jobs:
         timeout-minutes: 180 # 3h
         steps:
             - uses: actions/checkout@v3
-            - uses: helix-editor/rust-toolchain@v1
+            - uses: dtolnay/rust-toolchain@1.79
               with:
                 profile: minimal
 

.github/workflows/bench-pr.yml

@@ -16,6 +16,37 @@ jobs:
       runs-on: benchmarks
       timeout-minutes: 180 # 3h
       steps:
+        - name: Check permissions
+          id: permission
+          env:
+            PR_AUTHOR: ${{github.event.issue.user.login }}
+            COMMENT_AUTHOR: ${{github.event.comment.user.login }}
+            REPOSITORY: ${{github.repository}}
+            PR_ID: ${{github.event.issue.number}}
+          run: |
+            PR_REPOSITORY=$(gh api /repos/"$REPOSITORY"/pulls/"$PR_ID" --jq .head.repo.full_name)
+            if $(gh api /repos/"$REPOSITORY"/collaborators/"$PR_AUTHOR"/permission --jq .user.permissions.push)
+            then
+              echo "::notice title=Authentication success::PR author authenticated"
+            else
+              echo "::error title=Authentication error::PR author doesn't have push permission on this repository"
+              exit 1
+            fi
+            if $(gh api /repos/"$REPOSITORY"/collaborators/"$COMMENT_AUTHOR"/permission --jq .user.permissions.push)
+            then
+              echo "::notice title=Authentication success::Comment author authenticated"
+            else
+              echo "::error title=Authentication error::Comment author doesn't have push permission on this repository"
+              exit 1
+            fi
+            if [ "$PR_REPOSITORY" = "$REPOSITORY" ]
+            then
+              echo "::notice title=Authentication success::PR started from main repository"
+            else
+              echo "::error title=Authentication error::PR started from a fork"
+              exit 1
+            fi
+
         - name: Check for Command
           id: command
           uses: xt0rted/slash-command-action@v2
@@ -35,7 +66,7 @@ jobs:
             fetch-depth: 0 # fetch full history to be able to get main commit sha
             ref: ${{ steps.comment-branch.outputs.head_ref }}
 
-        - uses: helix-editor/rust-toolchain@v1
+        - uses: dtolnay/rust-toolchain@1.79
           with:
             profile: minimal
 

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

@@ -12,7 +12,7 @@ jobs:
         timeout-minutes: 180 # 3h
         steps:
           - uses: actions/checkout@v3
-          - uses: helix-editor/rust-toolchain@v1
+          - uses: dtolnay/rust-toolchain@1.79
             with:
               profile: minimal
 

.github/workflows/benchmarks-manual.yml

@@ -18,7 +18,7 @@ jobs:
     timeout-minutes: 4320 # 72h
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
 

.github/workflows/benchmarks-pr.yml

@@ -13,7 +13,38 @@ jobs:
     runs-on: benchmarks
     timeout-minutes: 4320 # 72h
     steps:
-      - uses: helix-editor/rust-toolchain@v1
+      - name: Check permissions
+        id: permission
+        env:
+          PR_AUTHOR: ${{github.event.issue.user.login }}
+          COMMENT_AUTHOR: ${{github.event.comment.user.login }}
+          REPOSITORY: ${{github.repository}}
+          PR_ID: ${{github.event.issue.number}}
+        run: |
+          PR_REPOSITORY=$(gh api /repos/"$REPOSITORY"/pulls/"$PR_ID" --jq .head.repo.full_name)
+          if $(gh api /repos/"$REPOSITORY"/collaborators/"$PR_AUTHOR"/permission --jq .user.permissions.push)
+          then
+            echo "::notice title=Authentication success::PR author authenticated"
+          else
+            echo "::error title=Authentication error::PR author doesn't have push permission on this repository"
+            exit 1
+          fi
+          if $(gh api /repos/"$REPOSITORY"/collaborators/"$COMMENT_AUTHOR"/permission --jq .user.permissions.push)
+          then
+            echo "::notice title=Authentication success::Comment author authenticated"
+          else
+            echo "::error title=Authentication error::Comment author doesn't have push permission on this repository"
+            exit 1
+          fi
+          if [ "$PR_REPOSITORY" = "$REPOSITORY" ]
+          then
+            echo "::notice title=Authentication success::PR started from main repository"
+          else
+            echo "::error title=Authentication error::PR started from a fork"
+            exit 1
+          fi
+
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
 

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

@@ -16,7 +16,7 @@ jobs:
     timeout-minutes: 4320 # 72h
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
 

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

@@ -15,7 +15,7 @@ jobs:
     runs-on: benchmarks
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
 

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

@@ -15,7 +15,7 @@ jobs:
     runs-on: benchmarks
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
 

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

@@ -15,7 +15,7 @@ jobs:
     runs-on: benchmarks
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
 

.github/workflows/flaky-tests.yml

@@ -1,6 +1,5 @@
 name: Look for flaky tests
-env:
-  ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
+
 on:
   workflow_dispatch:
   schedule:
@@ -10,15 +9,15 @@ jobs:
   flaky:
     runs-on: ubuntu-latest
     container:
-      # Use ubuntu-18.04 to compile with glibc 2.27, which are the production expectations
-      image: ubuntu:18.04
+      # Use ubuntu-20.04 to compile with glibc 2.28
+      image: ubuntu:20.04
     steps:
     - uses: actions/checkout@v3
     - name: Install needed dependencies
       run: |
         apt-get update && apt-get install -y curl
         apt-get install build-essential -y
-    - uses: helix-editor/rust-toolchain@v1
+    - uses: dtolnay/rust-toolchain@1.79
     - name: Install cargo-flaky
       run: cargo install cargo-flaky
     - name: Run cargo flaky in the dumps

.github/workflows/fuzzer-indexing.yml

@@ -1,6 +1,5 @@
 name: Run the indexing fuzzer
-env:
-  ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
+
 on:
   push:
     branches:
@@ -13,7 +12,7 @@ jobs:
     timeout-minutes: 4320 # 72h
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
 

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

@@ -15,19 +15,17 @@ jobs:
 
   debian:
     name: Publish debian packagge
-    env:
-      ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
     runs-on: ubuntu-latest
     needs: check-version
     container:
-      # Use ubuntu-18.04 to compile with glibc 2.27
-      image: ubuntu:18.04
+      # Use ubuntu-20.04 to compile with glibc 2.28
+      image: ubuntu:20.04
     steps:
     - name: Install needed dependencies
       run: |
         apt-get update && apt-get install -y curl
         apt-get install build-essential -y
-    - uses: helix-editor/rust-toolchain@v1
+    - uses: dtolnay/rust-toolchain@1.79
     - name: Install cargo-deb
       run: cargo install cargo-deb
     - uses: actions/checkout@v3

.github/workflows/publish-binaries.yml

@@ -35,19 +35,17 @@ jobs:
   publish-linux:
     name: Publish binary for Linux
     runs-on: ubuntu-latest
-    env:
-      ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
     needs: check-version
     container:
-      # Use ubuntu-18.04 to compile with glibc 2.27
-      image: ubuntu:18.04
+      # Use ubuntu-20.04 to compile with glibc 2.28
+      image: ubuntu:20.04
     steps:
     - uses: actions/checkout@v3
     - name: Install needed dependencies
       run: |
         apt-get update && apt-get install -y curl
         apt-get install build-essential -y
-    - uses: helix-editor/rust-toolchain@v1
+    - uses: dtolnay/rust-toolchain@1.79
     - name: Build
       run: cargo build --release --locked
     # No need to upload binaries for dry run (cron)
@@ -77,7 +75,7 @@ jobs:
             asset_name: meilisearch-windows-amd64.exe
     steps:
     - uses: actions/checkout@v3
-    - uses: helix-editor/rust-toolchain@v1
+    - uses: dtolnay/rust-toolchain@1.79
     - name: Build
       run: cargo build --release --locked
     # No need to upload binaries for dry run (cron)
@@ -103,7 +101,7 @@ jobs:
       - name: Checkout repository
         uses: actions/checkout@v3
       - name: Installing Rust toolchain
-        uses: helix-editor/rust-toolchain@v1
+        uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
           target: ${{ matrix.target }}
@@ -127,10 +125,10 @@ jobs:
     runs-on: ubuntu-latest
     needs: check-version
     env:
-      ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
+      DEBIAN_FRONTEND: noninteractive
     container:
-      # Use ubuntu-18.04 to compile with glibc 2.27
-      image: ubuntu:18.04
+      # Use ubuntu-20.04 to compile with glibc 2.28
+      image: ubuntu:20.04
     strategy:
       matrix:
         include:
@@ -150,7 +148,7 @@ jobs:
           add-apt-repository "deb [arch=$(dpkg --print-architecture)] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
           apt-get update -y && apt-get install -y docker-ce
       - name: Installing Rust toolchain
-        uses: helix-editor/rust-toolchain@v1
+        uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
           target: ${{ matrix.target }}
@@ -164,6 +162,9 @@ jobs:
           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:

.github/workflows/test-suite.yml

@@ -19,13 +19,11 @@ env:
 
 jobs:
   test-linux:
-    name: Tests on ubuntu-18.04
+    name: Tests on ubuntu-20.04
     runs-on: ubuntu-latest
-    env:
-      ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
     container:
-      # Use ubuntu-18.04 to compile with glibc 2.27, which are the production expectations
-      image: ubuntu:18.04
+      # Use ubuntu-20.04 to compile with glibc 2.28
+      image: ubuntu:20.04
     steps:
       - uses: actions/checkout@v3
       - name: Install needed dependencies
@@ -33,7 +31,7 @@ jobs:
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
       - name: Setup test with Rust stable
-        uses: helix-editor/rust-toolchain@v1
+        uses: dtolnay/rust-toolchain@1.79
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.7.1
       - name: Run cargo check without any default features
@@ -58,7 +56,7 @@ jobs:
       - uses: actions/checkout@v3
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.7.1
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
       - name: Run cargo check without any default features
         uses: actions-rs/cargo@v1
         with:
@@ -73,11 +71,9 @@ jobs:
   test-all-features:
     name: Tests almost all features
     runs-on: ubuntu-latest
-    env:
-      ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
     container:
-      # Use ubuntu-18.04 to compile with glibc 2.27, which are the production expectations
-      image: ubuntu:18.04
+      # Use ubuntu-20.04 to compile with glibc 2.28
+      image: ubuntu:20.04
     if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
     steps:
       - uses: actions/checkout@v3
@@ -85,7 +81,7 @@ jobs:
         run: |
           apt-get update
           apt-get install --assume-yes build-essential curl
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
       - name: Run cargo build with almost all features
         run: |
           cargo build --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda)"
@@ -95,11 +91,9 @@ jobs:
 
   test-disabled-tokenization:
     name: Test disabled tokenization
-    env:
-      ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
     runs-on: ubuntu-latest
     container:
-      image: ubuntu:18.04
+      image: ubuntu:20.04
     if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
     steps:
       - uses: actions/checkout@v3
@@ -107,7 +101,7 @@ jobs:
         run: |
           apt-get update
           apt-get install --assume-yes build-essential curl
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
       - name: Run cargo tree without default features and check lindera is not present
         run: |
           if cargo tree -f '{p} {f}' -e normal --no-default-features | grep -qz lindera; then
@@ -121,19 +115,17 @@ jobs:
   # We run tests in debug also, to make sure that the debug_assertions are hit
   test-debug:
     name: Run tests in debug
-    env:
-      ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true
     runs-on: ubuntu-latest
     container:
-      # Use ubuntu-18.04 to compile with glibc 2.27, which are the production expectations
-      image: ubuntu:18.04
+      # Use ubuntu-20.04 to compile with glibc 2.28
+      image: ubuntu:20.04
     steps:
       - uses: actions/checkout@v3
       - name: Install needed dependencies
         run: |
           apt-get update && apt-get install -y curl
           apt-get install build-essential -y
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
       - name: Cache dependencies
         uses: Swatinem/rust-cache@v2.7.1
       - name: Run tests in debug
@@ -147,7 +139,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
           components: clippy
@@ -164,7 +156,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
           toolchain: nightly-2024-07-09

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

@@ -18,7 +18,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v3
-      - uses: helix-editor/rust-toolchain@v1
+      - uses: dtolnay/rust-toolchain@1.79
         with:
           profile: minimal
       - name: Install sd

CONTRIBUTING.md

@@ -62,6 +62,10 @@ We recommend using the standard `$HOME/.cache/lindera` directory:
 export LINDERA_CACHE=$HOME/.cache/lindera
 ```
 
+Furthermore, you can improve incremental compilation by setting the `MEILI_NO_VERGEN` environment variable.
+Setting this variable will prevent the Meilisearch binary from being rebuilt each time the directory that hosts the Meilisearch repository changes.
+Do not enable this environment variable for production builds (as it will break the `version` route, among other things).
+
 #### Snapshot-based tests
 
 We are using [insta](https://insta.rs) to perform snapshot-based testing.

Cargo.lock

@@ -55,7 +55,7 @@ dependencies = [
  "encoding_rs",
  "flate2",
  "futures-core",
- "h2",
+ "h2 0.3.26",
  "http 0.2.11",
  "httparse",
  "httpdate",
@@ -149,11 +149,11 @@ dependencies = [
  "futures-core",
  "impl-more",
  "pin-project-lite",
+ "rustls-pki-types",
  "tokio",
- "tokio-rustls 0.24.1",
+ "tokio-rustls",
  "tokio-util",
  "tracing",
- "webpki-roots 0.25.3",
 ]
 
 [[package]]
@@ -387,14 +387,14 @@ checksum = "96d30a06541fbafbc7f82ed10c06164cfbd2c401138f6addd8404629c4b16711"
 [[package]]
 name = "arroy"
 version = "0.4.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2ece9e5347e7fdaaea3181dec7f916677ad5f3fcbac183648ce1924eb4aeef9a"
+source = "git+https://github.com/meilisearch/arroy/?rev=2386594dfb009ce08821a925ccc89fb8e30bf73d#2386594dfb009ce08821a925ccc89fb8e30bf73d"
 dependencies = [
  "bytemuck",
  "byteorder",
  "heed",
  "log",
  "memmap2",
+ "nohash",
  "ordered-float",
  "rand",
  "rayon",
@@ -403,6 +403,16 @@ dependencies = [
  "thiserror",
 ]
 
+[[package]]
+name = "assert-json-diff"
+version = "2.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "47e4f2b81832e72834d7518d8487a0396a28cc408186a2e8854c0f98011faf12"
+dependencies = [
+ "serde",
+ "serde_json",
+]
+
 [[package]]
 name = "async-trait"
 version = "0.1.81"
@@ -414,6 +424,12 @@ dependencies = [
  "syn 2.0.60",
 ]
 
+[[package]]
+name = "atomic-waker"
+version = "1.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1505bd5d3d116872e7271a6d4e16d81d0c8570876c8de68093a09ac269d8aac0"
+
 [[package]]
 name = "autocfg"
 version = "1.2.0"
@@ -455,7 +471,7 @@ checksum = "72b3254f16251a8381aa12e40e3c4d2f0199f8c6508fbecb9d91f575e0fbb8c6"
 
 [[package]]
 name = "benchmarks"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "anyhow",
  "bytes",
@@ -511,7 +527,7 @@ dependencies = [
  "proc-macro2",
  "quote",
  "regex",
- "rustc-hash",
+ "rustc-hash 1.1.0",
  "shlex",
  "syn 2.0.60",
 ]
@@ -636,7 +652,7 @@ dependencies = [
 
 [[package]]
 name = "build-info"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "anyhow",
  "time",
@@ -917,20 +933,17 @@ dependencies = [
 
 [[package]]
 name = "charabia"
-version = "0.8.12"
+version = "0.9.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9868a22f10dee80498a8a2b6c641d80bf28ea4495fcf71c2dc4836c2dd23958c"
+checksum = "55ff52497324e7d168505a16949ae836c14595606fab94687238d2f6c8d4c798"
 dependencies = [
  "aho-corasick",
- "cow-utils",
  "csv",
- "deunicode",
  "either",
  "fst",
  "irg-kvariants",
  "jieba-rs",
  "lindera",
- "litemap",
  "once_cell",
  "pinyin",
  "serde",
@@ -938,7 +951,6 @@ dependencies = [
  "unicode-normalization",
  "wana_kana",
  "whatlang",
- "zerovec",
 ]
 
 [[package]]
@@ -1129,12 +1141,6 @@ version = "0.8.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e496a50fda8aacccc86d7529e2c1e0892dbd0f898a6b5645b5561b89c3210efa"
 
-[[package]]
-name = "cow-utils"
-version = "0.1.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "79bb3adfaf5f75d24b01aee375f7555907840fa2800e5ec8fa3b9e2031830173"
-
 [[package]]
 name = "cpufeatures"
 version = "0.2.12"
@@ -1377,6 +1383,24 @@ dependencies = [
  "syn 2.0.60",
 ]
 
+[[package]]
+name = "deadpool"
+version = "0.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fb84100978c1c7b37f09ed3ce3e5f843af02c2a2c431bae5b19230dad2c1b490"
+dependencies = [
+ "async-trait",
+ "deadpool-runtime",
+ "num_cpus",
+ "tokio",
+]
+
+[[package]]
+name = "deadpool-runtime"
+version = "0.1.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "092966b41edc516079bdf31ec78a2e0588d1d0c08f78b91d8307215928642b2b"
+
 [[package]]
 name = "debugid"
 version = "0.8.0"
@@ -1517,12 +1541,6 @@ dependencies = [
  "syn 2.0.60",
 ]
 
-[[package]]
-name = "deunicode"
-version = "1.6.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "339544cc9e2c4dc3fc7149fd630c5f22263a4fdf18a98afd0075784968b5cf00"
-
 [[package]]
 name = "digest"
 version = "0.10.7"
@@ -1604,7 +1622,7 @@ dependencies = [
 
 [[package]]
 name = "dump"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "anyhow",
  "big_s",
@@ -1816,7 +1834,7 @@ checksum = "9fc0510504f03c51ada170672ac806f1f105a88aa97a5281117e1ddc3368e51a"
 
 [[package]]
 name = "file-store"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "tempfile",
  "thiserror",
@@ -1838,7 +1856,7 @@ dependencies = [
 
 [[package]]
 name = "filter-parser"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "insta",
  "nom",
@@ -1858,7 +1876,7 @@ dependencies = [
 
 [[package]]
 name = "flatten-serde-json"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "criterion",
  "serde_json",
@@ -1982,7 +2000,7 @@ dependencies = [
 
 [[package]]
 name = "fuzzers"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "arbitrary",
  "clap",
@@ -2231,6 +2249,25 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "h2"
+version = "0.4.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fa82e28a107a8cc405f0839610bdc9b15f1e25ec7d696aa5cf173edbcb1486ab"
+dependencies = [
+ "atomic-waker",
+ "bytes",
+ "fnv",
+ "futures-core",
+ "futures-sink",
+ "http 1.1.0",
+ "indexmap",
+ "slab",
+ "tokio",
+ "tokio-util",
+ "tracing",
+]
+
 [[package]]
 name = "half"
 version = "1.8.2"
@@ -2441,9 +2478,11 @@ dependencies = [
  "bytes",
  "futures-channel",
  "futures-util",
+ "h2 0.4.5",
  "http 1.1.0",
  "http-body",
  "httparse",
+ "httpdate",
  "itoa",
  "pin-project-lite",
  "smallvec",
@@ -2461,12 +2500,12 @@ dependencies = [
  "http 1.1.0",
  "hyper",
  "hyper-util",
- "rustls 0.23.11",
+ "rustls",
  "rustls-pki-types",
  "tokio",
- "tokio-rustls 0.26.0",
+ "tokio-rustls",
  "tower-service",
- "webpki-roots 0.26.1",
+ "webpki-roots",
 ]
 
 [[package]]
@@ -2513,7 +2552,7 @@ checksum = "206ca75c9c03ba3d4ace2460e57b189f39f43de612c2f85836e65c929701bb2d"
 
 [[package]]
 name = "index-scheduler"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "anyhow",
  "arroy",
@@ -2707,7 +2746,7 @@ dependencies = [
 
 [[package]]
 name = "json-depth-checker"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "criterion",
  "serde_json",
@@ -3223,12 +3262,6 @@ dependencies = [
  "unicode-segmentation",
 ]
 
-[[package]]
-name = "litemap"
-version = "0.7.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "643cb0b8d4fcc284004d5fd0d67ccf61dfffadb7f75e1e71bc420f4688a3a704"
-
 [[package]]
 name = "lmdb-master-sys"
 version = "0.2.2"
@@ -3332,7 +3365,7 @@ checksum = "490cc448043f947bae3cbee9c203358d62dbee0db12107a74be5c30ccfd09771"
 
 [[package]]
 name = "meili-snap"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "insta",
  "md5",
@@ -3341,7 +3374,7 @@ dependencies = [
 
 [[package]]
 name = "meilisearch"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "actix-cors",
  "actix-http",
@@ -3395,8 +3428,9 @@ dependencies = [
  "regex",
  "reqwest",
  "roaring",
- "rustls 0.21.12",
- "rustls-pemfile 1.0.4",
+ "rustls",
+ "rustls-pemfile",
+ "rustls-pki-types",
  "segment",
  "serde",
  "serde_json",
@@ -3422,13 +3456,14 @@ dependencies = [
  "url",
  "urlencoding",
  "uuid",
+ "wiremock",
  "yaup",
  "zip 2.1.3",
 ]
 
 [[package]]
 name = "meilisearch-auth"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "base64 0.22.1",
  "enum-iterator",
@@ -3447,7 +3482,7 @@ dependencies = [
 
 [[package]]
 name = "meilisearch-types"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "actix-web",
  "anyhow",
@@ -3477,7 +3512,7 @@ dependencies = [
 
 [[package]]
 name = "meilitool"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "anyhow",
  "clap",
@@ -3485,6 +3520,7 @@ dependencies = [
  "file-store",
  "meilisearch-auth",
  "meilisearch-types",
+ "serde",
  "time",
  "uuid",
 ]
@@ -3507,7 +3543,7 @@ dependencies = [
 
 [[package]]
 name = "milli"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "arroy",
  "big_s",
@@ -3542,6 +3578,7 @@ dependencies = [
  "maplit",
  "md5",
  "meili-snap",
+ "memchr",
  "memmap2",
  "mimalloc",
  "obkv",
@@ -3649,6 +3686,12 @@ version = "0.0.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6d02c0b00610773bb7fc61d85e13d86c7858cbdf00e1a120bfc41bc055dbaa0e"
 
+[[package]]
+name = "nohash"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a0f889fb66f7acdf83442c35775764b51fed3c606ab9cee51500dbde2cf528ca"
+
 [[package]]
 name = "nom"
 version = "7.1.3"
@@ -3940,7 +3983,7 @@ checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e"
 
 [[package]]
 name = "permissive-json-pointer"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "big_s",
  "serde_json",
@@ -4271,8 +4314,8 @@ dependencies = [
  "pin-project-lite",
  "quinn-proto",
  "quinn-udp",
- "rustc-hash",
- "rustls 0.23.11",
+ "rustc-hash 1.1.0",
+ "rustls",
  "thiserror",
  "tokio",
  "tracing",
@@ -4280,15 +4323,15 @@ dependencies = [
 
 [[package]]
 name = "quinn-proto"
-version = "0.11.3"
+version = "0.11.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ddf517c03a109db8100448a4be38d498df8a210a99fe0e1b9eaf39e78c640efe"
+checksum = "fadfaed2cd7f389d0161bb73eeb07b7b78f8691047a6f3e73caaeae55310a4a6"
 dependencies = [
  "bytes",
  "rand",
  "ring",
- "rustc-hash",
- "rustls 0.23.11",
+ "rustc-hash 2.0.0",
+ "rustls",
  "slab",
  "thiserror",
  "tinyvec",
@@ -4516,15 +4559,15 @@ dependencies = [
  "percent-encoding",
  "pin-project-lite",
  "quinn",
- "rustls 0.23.11",
- "rustls-pemfile 2.1.2",
+ "rustls",
+ "rustls-pemfile",
  "rustls-pki-types",
  "serde",
  "serde_json",
  "serde_urlencoded",
  "sync_wrapper",
  "tokio",
- "tokio-rustls 0.26.0",
+ "tokio-rustls",
  "tokio-util",
  "tower-service",
  "url",
@@ -4532,7 +4575,7 @@ dependencies = [
  "wasm-bindgen-futures",
  "wasm-streams",
  "web-sys",
- "webpki-roots 0.26.1",
+ "webpki-roots",
  "winreg",
 ]
 
@@ -4660,6 +4703,12 @@ version = "1.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
 
+[[package]]
+name = "rustc-hash"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "583034fd73374156e66797ed8e5b0d5690409c9226b22d87cb7f19821c05d152"
+
 [[package]]
 name = "rustc_version"
 version = "0.4.0"
@@ -4682,18 +4731,6 @@ dependencies = [
  "windows-sys 0.52.0",
 ]
 
-[[package]]
-name = "rustls"
-version = "0.21.12"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3f56a14d1f48b391359b22f731fd4bd7e43c97f3c50eee276f3aa09c94784d3e"
-dependencies = [
- "log",
- "ring",
- "rustls-webpki 0.101.7",
- "sct",
-]
-
 [[package]]
 name = "rustls"
 version = "0.23.11"
@@ -4704,20 +4741,11 @@ dependencies = [
  "once_cell",
  "ring",
  "rustls-pki-types",
- "rustls-webpki 0.102.5",
+ "rustls-webpki",
  "subtle",
  "zeroize",
 ]
 
-[[package]]
-name = "rustls-pemfile"
-version = "1.0.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1c74cae0a4cf6ccbbf5f359f08efdf8ee7e1dc532573bf0db71968cb56b1448c"
-dependencies = [
- "base64 0.21.7",
-]
-
 [[package]]
 name = "rustls-pemfile"
 version = "2.1.2"
@@ -4734,16 +4762,6 @@ version = "1.7.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "976295e77ce332211c0d24d92c0e83e50f5c5f046d11082cea19f3df13a3562d"
 
-[[package]]
-name = "rustls-webpki"
-version = "0.101.7"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8b6275d1ee7a1cd780b64aca7726599a1dbc893b1e64144529e55c3c2f745765"
-dependencies = [
- "ring",
- "untrusted",
-]
-
 [[package]]
 name = "rustls-webpki"
 version = "0.102.5"
@@ -4792,16 +4810,6 @@ version = "1.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
 
-[[package]]
-name = "sct"
-version = "0.7.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "da046153aa2352493d6cb7da4b6e5c0c057d8a1d0a9aa8560baffdd945acd414"
-dependencies = [
- "ring",
- "untrusted",
-]
-
 [[package]]
 name = "seahash"
 version = "4.1.0"
@@ -4839,9 +4847,9 @@ checksum = "a3f0bf26fd526d2a95683cd0f87bf103b8539e2ca1ef48ce002d67aad59aa0b4"
 
 [[package]]
 name = "serde"
-version = "1.0.204"
+version = "1.0.209"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bc76f558e0cbb2a839d37354c575f1dc3fdc6546b5be373ba43d95f231bf7c12"
+checksum = "99fce0ffe7310761ca6bf9faf5115afbc19688edd00171d81b1bb1b116c63e09"
 dependencies = [
  "serde_derive",
 ]
@@ -4857,9 +4865,9 @@ dependencies = [
 
 [[package]]
 name = "serde_derive"
-version = "1.0.204"
+version = "1.0.209"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e0cd7e117be63d3c3678776753929474f3b04a43a080c744d6b0ae2a8c28e222"
+checksum = "a5831b979fd7b5439637af1752d535ff49f4860c0f341d1baeb6faf0f4242170"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -5352,7 +5360,7 @@ dependencies = [
  "fancy-regex 0.12.0",
  "lazy_static",
  "parking_lot",
- "rustc-hash",
+ "rustc-hash 1.1.0",
 ]
 
 [[package]]
@@ -5482,23 +5490,13 @@ dependencies = [
  "syn 2.0.60",
 ]
 
-[[package]]
-name = "tokio-rustls"
-version = "0.24.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c28327cf380ac148141087fbfb9de9d7bd4e84ab5d2c28fbc911d753de8a7081"
-dependencies = [
- "rustls 0.21.12",
- "tokio",
-]
-
 [[package]]
 name = "tokio-rustls"
 version = "0.26.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0c7bc40d0e5a97695bb96e27995cd3a08538541b0a846f65bba7a359f36700d4"
 dependencies = [
- "rustls 0.23.11",
+ "rustls",
  "rustls-pki-types",
  "tokio",
 ]
@@ -5804,13 +5802,13 @@ dependencies = [
  "flate2",
  "log",
  "once_cell",
- "rustls 0.23.11",
+ "rustls",
  "rustls-pki-types",
  "serde",
  "serde_json",
  "socks",
  "url",
- "webpki-roots 0.26.1",
+ "webpki-roots",
 ]
 
 [[package]]
@@ -6035,12 +6033,6 @@ dependencies = [
  "wasm-bindgen",
 ]
 
-[[package]]
-name = "webpki-roots"
-version = "0.25.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1778a42e8b3b90bff8d0f5032bf22250792889a5cdc752aa0020c84abe3aaf10"
-
 [[package]]
 name = "webpki-roots"
 version = "0.26.1"
@@ -6336,6 +6328,30 @@ dependencies = [
  "windows-sys 0.48.0",
 ]
 
+[[package]]
+name = "wiremock"
+version = "0.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ec874e1eef0df2dcac546057fe5e29186f09c378181cd7b635b4b7bcc98e9d81"
+dependencies = [
+ "assert-json-diff",
+ "async-trait",
+ "base64 0.21.7",
+ "deadpool",
+ "futures",
+ "http 1.1.0",
+ "http-body-util",
+ "hyper",
+ "hyper-util",
+ "log",
+ "once_cell",
+ "regex",
+ "serde",
+ "serde_json",
+ "tokio",
+ "url",
+]
+
 [[package]]
 name = "wyz"
 version = "0.5.1"
@@ -6358,7 +6374,7 @@ dependencies = [
 
 [[package]]
 name = "xtask"
-version = "1.9.0"
+version = "1.11.0"
 dependencies = [
  "anyhow",
  "build-info",
@@ -6481,15 +6497,6 @@ dependencies = [
  "syn 2.0.60",
 ]
 
-[[package]]
-name = "zerovec"
-version = "0.10.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "aa2b893d79df23bfb12d5461018d408ea19dfafe76c2c7ef6d4eba614f8ff079"
-dependencies = [
- "zerofrom",
-]
-
 [[package]]
 name = "zip"
 version = "1.1.4"

Cargo.toml

@@ -22,7 +22,7 @@ members = [
 ]
 
 [workspace.package]
-version = "1.9.0"
+version = "1.11.0"
 authors = [
     "Quentin de Quelen <quentin@dequelen.me>",
     "Clément Renault <clement@meilisearch.com>",

README.md

@@ -45,14 +45,14 @@ See the list of all our example apps in our [demos repository](https://github.co
 ## ✨ Features
 - **Hybrid search:** Combine the best of both [semantic](https://www.meilisearch.com/docs/learn/experimental/vector_search?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features) & full-text search to get the most relevant results
 - **Search-as-you-type:** Find & display results in less than 50 milliseconds to provide an intuitive experience
-- **[Typo tolerance](https://www.meilisearch.com/docs/learn/configuration/typo_tolerance?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** get relevant matches even when queries contain typos and misspellings
+- **[Typo tolerance](https://www.meilisearch.com/docs/learn/relevancy/typo_tolerance_settings?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** get relevant matches even when queries contain typos and misspellings
 - **[Filtering](https://www.meilisearch.com/docs/learn/fine_tuning_results/filtering?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features) and [faceted search](https://www.meilisearch.com/docs/learn/fine_tuning_results/faceted_search?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** enhance your users' search experience with custom filters and build a faceted search interface in a few lines of code
 - **[Sorting](https://www.meilisearch.com/docs/learn/fine_tuning_results/sorting?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** sort results based on price, date, or pretty much anything else your users need
-- **[Synonym support](https://www.meilisearch.com/docs/learn/configuration/synonyms?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** configure synonyms to include more relevant content in your search results
+- **[Synonym support](https://www.meilisearch.com/docs/learn/relevancy/synonyms?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** configure synonyms to include more relevant content in your search results
 - **[Geosearch](https://www.meilisearch.com/docs/learn/fine_tuning_results/geosearch?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** filter and sort documents based on geographic data
 - **[Extensive language support](https://www.meilisearch.com/docs/learn/what_is_meilisearch/language?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** search datasets in any language, with optimized support for Chinese, Japanese, Hebrew, and languages using the Latin alphabet
 - **[Security management](https://www.meilisearch.com/docs/learn/security/master_api_keys?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** control which users can access what data with API keys that allow fine-grained permissions handling
-- **[Multi-Tenancy](https://www.meilisearch.com/docs/learn/security/tenant_tokens?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** personalize search results for any number of application tenants
+- **[Multi-Tenancy](https://www.meilisearch.com/docs/learn/security/multitenancy_tenant_tokens?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** personalize search results for any number of application tenants
 - **Highly Customizable:** customize Meilisearch to your specific needs or use our out-of-the-box and hassle-free presets
 - **[RESTful API](https://www.meilisearch.com/docs/reference/api/overview?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** integrate Meilisearch in your technical stack with our plugins and SDKs
 - **Easy to install, deploy, and maintain**

bors.toml

@@ -1,5 +1,5 @@
 status = [
-    'Tests on ubuntu-18.04',
+    'Tests on ubuntu-20.04',
     'Tests on macos-12',
     'Tests on windows-2022',
     'Run Clippy',

build-info/build.rs

@@ -5,6 +5,13 @@ fn main() {
 }
 
 fn emit_git_variables() -> anyhow::Result<()> {
+    println!("cargo::rerun-if-env-changed=MEILI_NO_VERGEN");
+
+    let has_vergen =
+        !matches!(std::env::var_os("MEILI_NO_VERGEN"), Some(x) if x != "false" && x != "0");
+
+    anyhow::ensure!(has_vergen, "disabled via `MEILI_NO_VERGEN`");
+
     // 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.

dump/src/lib.rs

@@ -286,6 +286,7 @@ pub(crate) mod test {
             pagination: Setting::NotSet,
             embedders: Setting::NotSet,
             search_cutoff_ms: Setting::NotSet,
+            localized_attributes: Setting::NotSet,
             _kind: std::marker::PhantomData,
         };
         settings.check()

dump/src/reader/compat/v5_to_v6.rs

@@ -379,6 +379,7 @@ impl<T> From<v5::Settings<T>> for v6::Settings<v6::Unchecked> {
                 v5::Setting::NotSet => v6::Setting::NotSet,
             },
             embedders: v6::Setting::NotSet,
+            localized_attributes: v6::Setting::NotSet,
             search_cutoff_ms: v6::Setting::NotSet,
             _kind: std::marker::PhantomData,
         }

dump/src/reader/mod.rs

@@ -255,6 +255,8 @@ pub(crate) mod test {
         }
         "###);
 
+        insta::assert_json_snapshot!(vector_index.settings().unwrap());
+
         {
             let documents: Result<Vec<_>> = vector_index.documents().unwrap().collect();
             let mut documents = documents.unwrap();

dump/src/reader/snapshots/dump__reader__test__import_dump_v6_with_vectors-5.snap

@@ -1,783 +1,56 @@
 ---
 source: dump/src/reader/mod.rs
-expression: document
+expression: vector_index.settings().unwrap()
 ---
 {
-  "id": "e3",
-  "desc": "overriden vector + map",
-  "_vectors": {
-    "default": [
-      0.2,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1,
-      0.1
-    ],
-    "toto": [
-      0.1
-    ]
-  }
+  "displayedAttributes": [
+    "*"
+  ],
+  "searchableAttributes": [
+    "*"
+  ],
+  "filterableAttributes": [],
+  "sortableAttributes": [],
+  "rankingRules": [
+    "words",
+    "typo",
+    "proximity",
+    "attribute",
+    "sort",
+    "exactness"
+  ],
+  "stopWords": [],
+  "nonSeparatorTokens": [],
+  "separatorTokens": [],
+  "dictionary": [],
+  "synonyms": {},
+  "distinctAttribute": null,
+  "proximityPrecision": "byWord",
+  "typoTolerance": {
+    "enabled": true,
+    "minWordSizeForTypos": {
+      "oneTypo": 5,
+      "twoTypos": 9
+    },
+    "disableOnWords": [],
+    "disableOnAttributes": []
+  },
+  "faceting": {
+    "maxValuesPerFacet": 100,
+    "sortFacetValuesBy": {
+      "*": "alpha"
+    }
+  },
+  "pagination": {
+    "maxTotalHits": 1000
+  },
+  "embedders": {
+    "default": {
+      "source": "huggingFace",
+      "model": "BAAI/bge-base-en-v1.5",
+      "revision": "617ca489d9e86b49b8167676d8220688b99db36e",
+      "documentTemplate": "{% for field in fields %} {{ field.name }}: {{ field.value }}\n{% endfor %}"
+    }
+  },
+  "searchCutoffMs": null
 }

dump/src/reader/snapshots/dump__reader__test__import_dump_v6_with_vectors-9.snap

@@ -0,0 +1,780 @@
+---
+source: dump/src/reader/mod.rs
+expression: document
+---
+{
+  "id": "e0",
+  "desc": "overriden vector",
+  "_vectors": {
+    "default": [
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1
+    ]
+  }
+}

dump/src/reader/v4/meta.rs

@@ -74,7 +74,8 @@ impl Display for IndexUidFormatError {
             f,
             "invalid index uid `{}`, the uid must be an integer \
             or a string containing only alphanumeric characters \
-            a-z A-Z 0-9, hyphens - and underscores _.",
+            a-z A-Z 0-9, hyphens - and underscores _, \
+            and can not be more than 400 bytes.",
             self.invalid_uid,
         )
     }

dump/src/reader/v5/meta.rs

@@ -74,7 +74,8 @@ impl Display for IndexUidFormatError {
             f,
             "invalid index uid `{}`, the uid must be an integer \
             or a string containing only alphanumeric characters \
-            a-z A-Z 0-9, hyphens - and underscores _.",
+            a-z A-Z 0-9, hyphens - and underscores _, \
+            and can not be more than 400 bytes.",
             self.invalid_uid,
         )
     }

filter-parser/src/condition.rs

@@ -26,6 +26,8 @@ pub enum Condition<'a> {
     LowerThan(Token<'a>),
     LowerThanOrEqual(Token<'a>),
     Between { from: Token<'a>, to: Token<'a> },
+    Contains { keyword: Token<'a>, word: Token<'a> },
+    StartsWith { keyword: Token<'a>, word: Token<'a> },
 }
 
 /// condition      = value ("==" | ">" ...) value
@@ -92,6 +94,62 @@ pub fn parse_not_exists(input: Span) -> IResult<FilterCondition> {
     Ok((input, FilterCondition::Not(Box::new(FilterCondition::Condition { fid: key, op: Exists }))))
 }
 
+/// contains        = value "CONTAINS" value
+pub fn parse_contains(input: Span) -> IResult<FilterCondition> {
+    let (input, (fid, contains, value)) =
+        tuple((parse_value, tag("CONTAINS"), cut(parse_value)))(input)?;
+    Ok((
+        input,
+        FilterCondition::Condition {
+            fid,
+            op: Contains { keyword: Token { span: contains, value: None }, word: value },
+        },
+    ))
+}
+
+/// contains        = value "NOT" WS+ "CONTAINS" value
+pub fn parse_not_contains(input: Span) -> IResult<FilterCondition> {
+    let keyword = tuple((tag("NOT"), multispace1, tag("CONTAINS")));
+    let (input, (fid, (_not, _spaces, contains), value)) =
+        tuple((parse_value, keyword, cut(parse_value)))(input)?;
+
+    Ok((
+        input,
+        FilterCondition::Not(Box::new(FilterCondition::Condition {
+            fid,
+            op: Contains { keyword: Token { span: contains, value: None }, word: value },
+        })),
+    ))
+}
+
+/// starts with        = value "CONTAINS" value
+pub fn parse_starts_with(input: Span) -> IResult<FilterCondition> {
+    let (input, (fid, starts_with, value)) =
+        tuple((parse_value, tag("STARTS WITH"), cut(parse_value)))(input)?;
+    Ok((
+        input,
+        FilterCondition::Condition {
+            fid,
+            op: StartsWith { keyword: Token { span: starts_with, value: None }, word: value },
+        },
+    ))
+}
+
+/// starts with        = value "NOT" WS+ "CONTAINS" value
+pub fn parse_not_starts_with(input: Span) -> IResult<FilterCondition> {
+    let keyword = tuple((tag("NOT"), multispace1, tag("STARTS WITH")));
+    let (input, (fid, (_not, _spaces, starts_with), value)) =
+        tuple((parse_value, keyword, cut(parse_value)))(input)?;
+
+    Ok((
+        input,
+        FilterCondition::Not(Box::new(FilterCondition::Condition {
+            fid,
+            op: StartsWith { keyword: Token { span: starts_with, value: None }, word: value },
+        })),
+    ))
+}
+
 /// to             = value value "TO" WS+ value
 pub fn parse_to(input: Span) -> IResult<FilterCondition> {
     let (input, (key, from, _, _, to)) =

filter-parser/src/error.rs

@@ -146,7 +146,7 @@ impl<'a> Display for Error<'a> {
             }
             ErrorKind::InvalidPrimary => {
                 let text = if input.trim().is_empty() { "but instead got nothing.".to_string() } else { format!("at `{}`.", escaped_input) };
-                writeln!(f, "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` {}", text)?
+                writeln!(f, "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` {}", text)?
             }
             ErrorKind::InvalidEscapedNumber => {
                 writeln!(f, "Found an invalid escaped sequence number: `{}`.", escaped_input)?

filter-parser/src/lib.rs

@@ -48,8 +48,8 @@ use std::fmt::Debug;
 
 pub use condition::{parse_condition, parse_to, Condition};
 use condition::{
-    parse_exists, parse_is_empty, parse_is_not_empty, parse_is_not_null, parse_is_null,
-    parse_not_exists,
+    parse_contains, parse_exists, parse_is_empty, parse_is_not_empty, parse_is_not_null,
+    parse_is_null, parse_not_contains, parse_not_exists, parse_not_starts_with, parse_starts_with,
 };
 use error::{cut_with_err, ExpectedValueKind, NomErrorExt};
 pub use error::{Error, ErrorKind};
@@ -147,7 +147,38 @@ pub enum FilterCondition<'a> {
     GeoBoundingBox { top_right_point: [Token<'a>; 2], bottom_left_point: [Token<'a>; 2] },
 }
 
+pub enum TraversedElement<'a> {
+    FilterCondition(&'a FilterCondition<'a>),
+    Condition(&'a Condition<'a>),
+}
+
 impl<'a> FilterCondition<'a> {
+    pub fn use_contains_operator(&self) -> Option<&Token> {
+        match self {
+            FilterCondition::Condition { fid: _, op } => match op {
+                Condition::GreaterThan(_)
+                | Condition::GreaterThanOrEqual(_)
+                | Condition::Equal(_)
+                | Condition::NotEqual(_)
+                | Condition::Null
+                | Condition::Empty
+                | Condition::Exists
+                | Condition::LowerThan(_)
+                | Condition::LowerThanOrEqual(_)
+                | Condition::Between { .. } => None,
+                Condition::Contains { keyword, word: _ }
+                | Condition::StartsWith { keyword, word: _ } => Some(keyword),
+            },
+            FilterCondition::Not(this) => this.use_contains_operator(),
+            FilterCondition::Or(seq) | FilterCondition::And(seq) => {
+                seq.iter().find_map(|filter| filter.use_contains_operator())
+            }
+            FilterCondition::GeoLowerThan { .. }
+            | FilterCondition::GeoBoundingBox { .. }
+            | FilterCondition::In { .. } => None,
+        }
+    }
+
     /// Returns the first token found at the specified depth, `None` if no token at this depth.
     pub fn token_at_depth(&self, depth: usize) -> Option<&Token> {
         match self {
@@ -452,6 +483,10 @@ fn parse_primary(input: Span, depth: usize) -> IResult<FilterCondition> {
         parse_exists,
         parse_not_exists,
         parse_to,
+        parse_contains,
+        parse_not_contains,
+        parse_starts_with,
+        parse_not_starts_with,
         // the next lines are only for error handling and are written at the end to have the less possible performance impact
         parse_geo,
         parse_geo_distance,
@@ -534,6 +569,8 @@ impl<'a> std::fmt::Display for Condition<'a> {
             Condition::LowerThan(token) => write!(f, "< {token}"),
             Condition::LowerThanOrEqual(token) => write!(f, "<= {token}"),
             Condition::Between { from, to } => write!(f, "{from} TO {to}"),
+            Condition::Contains { word, keyword: _ } => write!(f, "CONTAINS {word}"),
+            Condition::StartsWith { word, keyword: _ } => write!(f, "STARTS WITH {word}"),
         }
     }
 }
@@ -558,6 +595,7 @@ pub mod tests {
         unsafe { Span::new_from_raw_offset(offset, lines as u32, value, "") }.into()
     }
 
+    #[track_caller]
     fn p(s: &str) -> impl std::fmt::Display + '_ {
         Fc::parse(s).unwrap().unwrap()
     }
@@ -639,6 +677,20 @@ pub mod tests {
         insta::assert_snapshot!(p("NOT subscribers NOT EXISTS"), @"{subscribers} EXISTS");
         insta::assert_snapshot!(p("subscribers NOT   EXISTS"), @"NOT ({subscribers} EXISTS)");
 
+        // Test CONTAINS + NOT CONTAINS
+        insta::assert_snapshot!(p("subscribers CONTAINS 'hello'"), @"{subscribers} CONTAINS {hello}");
+        insta::assert_snapshot!(p("NOT subscribers CONTAINS 'hello'"), @"NOT ({subscribers} CONTAINS {hello})");
+        insta::assert_snapshot!(p("subscribers NOT CONTAINS hello"), @"NOT ({subscribers} CONTAINS {hello})");
+        insta::assert_snapshot!(p("NOT subscribers NOT CONTAINS 'hello'"), @"{subscribers} CONTAINS {hello}");
+        insta::assert_snapshot!(p("subscribers NOT   CONTAINS 'hello'"), @"NOT ({subscribers} CONTAINS {hello})");
+
+        // Test STARTS WITH + NOT STARTS WITH
+        insta::assert_snapshot!(p("subscribers STARTS WITH 'hel'"), @"{subscribers} STARTS WITH {hel}");
+        insta::assert_snapshot!(p("NOT subscribers STARTS WITH 'hel'"), @"NOT ({subscribers} STARTS WITH {hel})");
+        insta::assert_snapshot!(p("subscribers NOT STARTS WITH hel"), @"NOT ({subscribers} STARTS WITH {hel})");
+        insta::assert_snapshot!(p("NOT subscribers NOT STARTS WITH 'hel'"), @"{subscribers} STARTS WITH {hel}");
+        insta::assert_snapshot!(p("subscribers NOT   STARTS WITH 'hel'"), @"NOT ({subscribers} STARTS WITH {hel})");
+
         // Test nested NOT
         insta::assert_snapshot!(p("NOT NOT NOT NOT x = 5"), @"{x} = {5}");
         insta::assert_snapshot!(p("NOT NOT (NOT NOT x = 5)"), @"{x} = {5}");
@@ -710,7 +762,7 @@ pub mod tests {
         "###);
 
         insta::assert_snapshot!(p("'OR'"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `\'OR\'`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `\'OR\'`.
         1:5 'OR'
         "###);
 
@@ -720,12 +772,12 @@ pub mod tests {
         "###);
 
         insta::assert_snapshot!(p("channel Ponce"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `channel Ponce`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `channel Ponce`.
         1:14 channel Ponce
         "###);
 
         insta::assert_snapshot!(p("channel = Ponce OR"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` but instead got nothing.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` but instead got nothing.
         19:19 channel = Ponce OR
         "###);
 
@@ -810,12 +862,12 @@ pub mod tests {
         "###);
 
         insta::assert_snapshot!(p("colour NOT EXIST"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `colour NOT EXIST`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `colour NOT EXIST`.
         1:17 colour NOT EXIST
         "###);
 
         insta::assert_snapshot!(p("subscribers 100 TO1000"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `subscribers 100 TO1000`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `subscribers 100 TO1000`.
         1:23 subscribers 100 TO1000
         "###);
 
@@ -878,35 +930,35 @@ pub mod tests {
         "###);
 
         insta::assert_snapshot!(p(r#"value NULL"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value NULL`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value NULL`.
         1:11 value NULL
         "###);
         insta::assert_snapshot!(p(r#"value NOT NULL"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value NOT NULL`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value NOT NULL`.
         1:15 value NOT NULL
         "###);
         insta::assert_snapshot!(p(r#"value EMPTY"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value EMPTY`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value EMPTY`.
         1:12 value EMPTY
         "###);
         insta::assert_snapshot!(p(r#"value NOT EMPTY"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value NOT EMPTY`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value NOT EMPTY`.
         1:16 value NOT EMPTY
         "###);
         insta::assert_snapshot!(p(r#"value IS"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value IS`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS`.
         1:9 value IS
         "###);
         insta::assert_snapshot!(p(r#"value IS NOT"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value IS NOT`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS NOT`.
         1:13 value IS NOT
         "###);
         insta::assert_snapshot!(p(r#"value IS EXISTS"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value IS EXISTS`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS EXISTS`.
         1:16 value IS EXISTS
         "###);
         insta::assert_snapshot!(p(r#"value IS NOT EXISTS"#), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `value IS NOT EXISTS`.
+        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `value IS NOT EXISTS`.
         1:20 value IS NOT EXISTS
         "###);
     }

filter-parser/src/value.rs

@@ -211,6 +211,9 @@ fn is_keyword(s: &str) -> bool {
             | "IS"
             | "NULL"
             | "EMPTY"
+            | "CONTAINS"
+            | "STARTS"
+            | "WITH"
             | "_geoRadius"
             | "_geoBoundingBox"
     )

index-scheduler/Cargo.toml

@@ -40,7 +40,7 @@ ureq = "2.10.0"
 uuid = { version = "1.10.0", features = ["serde", "v4"] }
 
 [dev-dependencies]
-arroy = "0.4.0"
+arroy = { git = "https://github.com/meilisearch/arroy/", rev = "2386594dfb009ce08821a925ccc89fb8e30bf73d" }
 big_s = "1.0.2"
 crossbeam = "0.8.4"
 insta = { version = "1.39.0", features = ["json", "redactions"] }

index-scheduler/src/autobatcher.rs

@@ -25,8 +25,9 @@ enum AutobatchKind {
         primary_key: Option<String>,
     },
     DocumentEdition,
-    DocumentDeletion,
-    DocumentDeletionByFilter,
+    DocumentDeletion {
+        by_filter: bool,
+    },
     DocumentClear,
     Settings {
         allow_index_creation: bool,
@@ -65,10 +66,12 @@ impl From<KindWithContent> for AutobatchKind {
                 ..
             } => AutobatchKind::DocumentImport { method, allow_index_creation, primary_key },
             KindWithContent::DocumentEdition { .. } => AutobatchKind::DocumentEdition,
-            KindWithContent::DocumentDeletion { .. } => AutobatchKind::DocumentDeletion,
+            KindWithContent::DocumentDeletion { .. } => {
+                AutobatchKind::DocumentDeletion { by_filter: false }
+            }
             KindWithContent::DocumentClear { .. } => AutobatchKind::DocumentClear,
             KindWithContent::DocumentDeletionByFilter { .. } => {
-                AutobatchKind::DocumentDeletionByFilter
+                AutobatchKind::DocumentDeletion { by_filter: true }
             }
             KindWithContent::SettingsUpdate { allow_index_creation, is_deletion, .. } => {
                 AutobatchKind::Settings {
@@ -105,9 +108,7 @@ pub enum BatchKind {
     },
     DocumentDeletion {
         deletion_ids: Vec<TaskId>,
-    },
-    DocumentDeletionByFilter {
-        id: TaskId,
+        includes_by_filter: bool,
     },
     ClearAndSettings {
         other: Vec<TaskId>,
@@ -205,12 +206,13 @@ impl BatchKind {
                 allow_index_creation,
             ),
             K::DocumentEdition => (Break(BatchKind::DocumentEdition { id: task_id }), false),
-            K::DocumentDeletion => {
-                (Continue(BatchKind::DocumentDeletion { deletion_ids: vec![task_id] }), false)
-            }
-            K::DocumentDeletionByFilter => {
-                (Break(BatchKind::DocumentDeletionByFilter { id: task_id }), false)
-            }
+            K::DocumentDeletion { by_filter: includes_by_filter } => (
+                Continue(BatchKind::DocumentDeletion {
+                    deletion_ids: vec![task_id],
+                    includes_by_filter,
+                }),
+                false,
+            ),
             K::Settings { allow_index_creation } => (
                 Continue(BatchKind::Settings { allow_index_creation, settings_ids: vec![task_id] }),
                 allow_index_creation,
@@ -228,7 +230,7 @@ impl BatchKind {
 
         match (self, kind) {
             // We don't batch any of these operations
-            (this, K::IndexCreation | K::IndexUpdate | K::IndexSwap | K::DocumentEdition | K::DocumentDeletionByFilter) => Break(this),
+            (this, K::IndexCreation | K::IndexUpdate | K::IndexSwap | K::DocumentEdition) => Break(this),
             // We must not batch tasks that don't have the same index creation rights if the index doesn't already exists.
             (this, kind) if !index_already_exists && this.allow_index_creation() == Some(false) && kind.allow_index_creation() == Some(true) => {
                 Break(this)
@@ -264,7 +266,7 @@ impl BatchKind {
             // The index deletion can batch with everything but must stop after
             (
                 BatchKind::DocumentClear { mut ids }
-                | BatchKind::DocumentDeletion { deletion_ids: mut ids }
+                | BatchKind::DocumentDeletion { deletion_ids: mut ids, includes_by_filter: _ }
                 | BatchKind::DocumentOperation { method: _, allow_index_creation: _, primary_key: _, operation_ids: mut ids }
                 | BatchKind::Settings { allow_index_creation: _, settings_ids: mut ids },
                 K::IndexDeletion,
@@ -284,7 +286,7 @@ impl BatchKind {
 
             (
                 BatchKind::DocumentClear { mut ids },
-                K::DocumentClear | K::DocumentDeletion,
+                K::DocumentClear | K::DocumentDeletion { by_filter: _ },
             ) => {
                 ids.push(id);
                 Continue(BatchKind::DocumentClear { ids })
@@ -328,7 +330,7 @@ impl BatchKind {
             }
             (
                 BatchKind::DocumentOperation { method, allow_index_creation, primary_key, mut operation_ids },
-                K::DocumentDeletion,
+                K::DocumentDeletion { by_filter: false },
             ) => {
                 operation_ids.push(id);
 
@@ -339,6 +341,13 @@ impl BatchKind {
                     operation_ids,
                 })
             }
+            // We can't batch a document operation with a delete by filter
+            (
+                this @ BatchKind::DocumentOperation { .. },
+                K::DocumentDeletion { by_filter: true },
+            ) => {
+                Break(this)
+            }
             // but we can't autobatch documents if it's not the same kind
             // this match branch MUST be AFTER the previous one
             (
@@ -357,13 +366,18 @@ impl BatchKind {
                 operation_ids,
             }),
 
-            (BatchKind::DocumentDeletion { mut deletion_ids }, K::DocumentClear) => {
+            (BatchKind::DocumentDeletion { mut deletion_ids, includes_by_filter: _ }, K::DocumentClear) => {
                 deletion_ids.push(id);
                 Continue(BatchKind::DocumentClear { ids: deletion_ids })
             }
+            // we can't autobatch the deletion and import if the document deletion contained a filter
+            (
+                this @ BatchKind::DocumentDeletion { deletion_ids: _, includes_by_filter: true },
+                K::DocumentImport { .. }
+            ) => Break(this),
             // we can autobatch the deletion and import if the index already exists
             (
-                BatchKind::DocumentDeletion { mut deletion_ids },
+                BatchKind::DocumentDeletion { mut deletion_ids, includes_by_filter: false },
                 K::DocumentImport { method, allow_index_creation, primary_key }
             ) if index_already_exists => {
                 deletion_ids.push(id);
@@ -377,7 +391,7 @@ impl BatchKind {
             }
             // we can autobatch the deletion and import if both can't create an index
             (
-                BatchKind::DocumentDeletion { mut deletion_ids },
+                BatchKind::DocumentDeletion { mut deletion_ids, includes_by_filter: false },
                 K::DocumentImport { method, allow_index_creation, primary_key }
             ) if !allow_index_creation => {
                 deletion_ids.push(id);
@@ -396,9 +410,9 @@ impl BatchKind {
             ) => {
                 Break(this)
             }
-            (BatchKind::DocumentDeletion { mut deletion_ids }, K::DocumentDeletion) => {
+            (BatchKind::DocumentDeletion { mut deletion_ids, includes_by_filter }, K::DocumentDeletion { by_filter }) => {
                 deletion_ids.push(id);
-                Continue(BatchKind::DocumentDeletion { deletion_ids })
+                Continue(BatchKind::DocumentDeletion { deletion_ids, includes_by_filter: includes_by_filter | by_filter })
             }
             (this @ BatchKind::DocumentDeletion { .. }, K::Settings { .. }) => Break(this),
 
@@ -412,7 +426,7 @@ impl BatchKind {
             }),
             (
                 this @ BatchKind::Settings { .. },
-                K::DocumentImport { .. } | K::DocumentDeletion,
+                K::DocumentImport { .. } | K::DocumentDeletion { .. },
             ) => Break(this),
             (
                 BatchKind::Settings { mut settings_ids, allow_index_creation },
@@ -443,7 +457,7 @@ impl BatchKind {
                     settings_ids,
                     allow_index_creation,
                 },
-                K::DocumentDeletion,
+                K::DocumentDeletion { .. },
             ) => {
                 other.push(id);
                 Continue(BatchKind::ClearAndSettings {
@@ -505,7 +519,7 @@ impl BatchKind {
             // this MUST be AFTER the two previous branch
             (
                 this @ BatchKind::SettingsAndDocumentOperation { .. },
-                K::DocumentDeletion | K::DocumentImport { .. },
+                K::DocumentDeletion { .. } | K::DocumentImport { .. },
             ) => Break(this),
             (
                 BatchKind::SettingsAndDocumentOperation { mut settings_ids, method, allow_index_creation,primary_key, operation_ids },
@@ -525,8 +539,7 @@ impl BatchKind {
                 | BatchKind::IndexDeletion { .. }
                 | BatchKind::IndexUpdate { .. }
                 | BatchKind::IndexSwap { .. }
-                | BatchKind::DocumentEdition { .. }
-                | BatchKind::DocumentDeletionByFilter { .. },
+                | BatchKind::DocumentEdition { .. },
                 _,
             ) => {
                 unreachable!()
@@ -616,6 +629,13 @@ mod tests {
         }
     }
 
+    fn doc_del_fil() -> KindWithContent {
+        KindWithContent::DocumentDeletionByFilter {
+            index_uid: String::from("doggo"),
+            filter_expr: serde_json::json!("cuteness > 100"),
+        }
+    }
+
     fn doc_clr() -> KindWithContent {
         KindWithContent::DocumentClear { index_uid: String::from("doggo") }
     }
@@ -676,10 +696,16 @@ mod tests {
         debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1, 2] }, false))");
 
         // we can autobatch one or multiple DocumentDeletion together
-        debug_snapshot!(autobatch_from(true, None, [doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_del(), doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0, 1, 2] }, false))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_del(), doc_del(), doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0, 1, 2] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_del(), doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0, 1, 2], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(false,None,  [doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(false,None,  [doc_del(), doc_del(), doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0, 1, 2], includes_by_filter: false }, false))");
+
+        // we can autobatch one or multiple DocumentDeletionByFilter together
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_del_fil(), doc_del_fil()]), @"Some((DocumentDeletion { deletion_ids: [0, 1, 2], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(false,None,  [doc_del_fil()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(false,None,  [doc_del_fil(), doc_del_fil(), doc_del_fil()]), @"Some((DocumentDeletion { deletion_ids: [0, 1, 2], includes_by_filter: true }, false))");
 
         // we can autobatch one or multiple Settings together
         debug_snapshot!(autobatch_from(true, None, [settings(true)]), @"Some((Settings { allow_index_creation: true, settings_ids: [0] }, true))");
@@ -722,25 +748,63 @@ mod tests {
         debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
         debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+
+        // But we can't autobatch document addition with document deletion by filter
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+        // And the other way around
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(UpdateDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(UpdateDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del_fil(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del_fil(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del_fil(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del_fil(), doc_imp(UpdateDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
     }
 
     #[test]
     fn simple_document_operation_dont_autobatch_with_other() {
-        // addition, updates and deletion can't batch together
+        // addition, updates and deletion by filter can't batch together
         debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
         debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
 
         debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_create()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
         debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_create()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_create()]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_create()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), idx_create()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
 
         debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_update()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
         debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_update()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_update()]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_update()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), idx_update()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
 
         debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_swap()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
         debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_swap()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_swap()]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_swap()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), idx_swap()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
     }
 
     #[test]
@@ -807,6 +871,7 @@ mod tests {
         debug_snapshot!(autobatch_from(true, None, [idx_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(true, None, [idx_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(true, None, [idx_del(), doc_del()]), @"Some((IndexDeletion { ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [idx_del(), doc_del_fil()]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(true, None, [idx_del(), doc_clr()]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(true, None, [idx_del(), settings(true)]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(true, None, [idx_del(), settings(false)]), @"Some((IndexDeletion { ids: [0] }, false))");
@@ -816,6 +881,7 @@ mod tests {
         debug_snapshot!(autobatch_from(false,None,  [idx_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [idx_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [idx_del(), doc_del()]), @"Some((IndexDeletion { ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(false,None,  [idx_del(), doc_del_fil()]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [idx_del(), doc_clr()]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [idx_del(), settings(true)]), @"Some((IndexDeletion { ids: [0] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [idx_del(), settings(false)]), @"Some((IndexDeletion { ids: [0] }, false))");
@@ -827,6 +893,7 @@ mod tests {
         debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(true, None, [doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(true, None, [settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, true))");
         debug_snapshot!(autobatch_from(true, None, [settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
@@ -836,6 +903,7 @@ mod tests {
         debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [doc_del(), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(false,None,  [doc_del_fil(), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, true))");
         debug_snapshot!(autobatch_from(false,None,  [settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
@@ -901,10 +969,10 @@ mod tests {
         debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
 
         // batch deletion and addition
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
     }
 
     #[test]

index-scheduler/src/batch.rs

@@ -110,9 +110,9 @@ pub(crate) enum IndexOperation {
         index_uid: String,
         task: Task,
     },
-    IndexDocumentDeletionByFilter {
+    DocumentDeletion {
         index_uid: String,
-        task: Task,
+        tasks: Vec<Task>,
     },
     DocumentClear {
         index_uid: String,
@@ -165,11 +165,11 @@ impl Batch {
             Batch::IndexOperation { op, .. } => match op {
                 IndexOperation::DocumentOperation { tasks, .. }
                 | IndexOperation::Settings { tasks, .. }
+                | IndexOperation::DocumentDeletion { tasks, .. }
                 | IndexOperation::DocumentClear { tasks, .. } => {
                     RoaringBitmap::from_iter(tasks.iter().map(|task| task.uid))
                 }
-                IndexOperation::DocumentEdition { task, .. }
-                | IndexOperation::IndexDocumentDeletionByFilter { task, .. } => {
+                IndexOperation::DocumentEdition { task, .. } => {
                     RoaringBitmap::from_sorted_iter(std::iter::once(task.uid)).unwrap()
                 }
                 IndexOperation::SettingsAndDocumentOperation {
@@ -234,7 +234,7 @@ impl IndexOperation {
         match self {
             IndexOperation::DocumentOperation { index_uid, .. }
             | IndexOperation::DocumentEdition { index_uid, .. }
-            | IndexOperation::IndexDocumentDeletionByFilter { index_uid, .. }
+            | IndexOperation::DocumentDeletion { index_uid, .. }
             | IndexOperation::DocumentClear { index_uid, .. }
             | IndexOperation::Settings { index_uid, .. }
             | IndexOperation::DocumentClearAndSetting { index_uid, .. }
@@ -252,8 +252,8 @@ impl fmt::Display for IndexOperation {
             IndexOperation::DocumentEdition { .. } => {
                 f.write_str("IndexOperation::DocumentEdition")
             }
-            IndexOperation::IndexDocumentDeletionByFilter { .. } => {
-                f.write_str("IndexOperation::IndexDocumentDeletionByFilter")
+            IndexOperation::DocumentDeletion { .. } => {
+                f.write_str("IndexOperation::DocumentDeletion")
             }
             IndexOperation::DocumentClear { .. } => f.write_str("IndexOperation::DocumentClear"),
             IndexOperation::Settings { .. } => f.write_str("IndexOperation::Settings"),
@@ -289,21 +289,6 @@ impl IndexScheduler {
                 },
                 must_create_index,
             })),
-            BatchKind::DocumentDeletionByFilter { id } => {
-                let task = self.get_task(rtxn, id)?.ok_or(Error::CorruptedTaskQueue)?;
-                match &task.kind {
-                    KindWithContent::DocumentDeletionByFilter { index_uid, .. } => {
-                        Ok(Some(Batch::IndexOperation {
-                            op: IndexOperation::IndexDocumentDeletionByFilter {
-                                index_uid: index_uid.clone(),
-                                task,
-                            },
-                            must_create_index: false,
-                        }))
-                    }
-                    _ => unreachable!(),
-                }
-            }
             BatchKind::DocumentEdition { id } => {
                 let task = self.get_task(rtxn, id)?.ok_or(Error::CorruptedTaskQueue)?;
                 match &task.kind {
@@ -366,30 +351,11 @@ impl IndexScheduler {
                     must_create_index,
                 }))
             }
-            BatchKind::DocumentDeletion { deletion_ids } => {
+            BatchKind::DocumentDeletion { deletion_ids, includes_by_filter: _ } => {
                 let tasks = self.get_existing_tasks(rtxn, deletion_ids)?;
 
-                let mut operations = Vec::with_capacity(tasks.len());
-                let mut documents_counts = Vec::with_capacity(tasks.len());
-                for task in &tasks {
-                    match task.kind {
-                        KindWithContent::DocumentDeletion { ref documents_ids, .. } => {
-                            operations.push(DocumentOperation::Delete(documents_ids.clone()));
-                            documents_counts.push(documents_ids.len() as u64);
-                        }
-                        _ => unreachable!(),
-                    }
-                }
-
                 Ok(Some(Batch::IndexOperation {
-                    op: IndexOperation::DocumentOperation {
-                        index_uid,
-                        primary_key: None,
-                        method: IndexDocumentsMethod::ReplaceDocuments,
-                        documents_counts,
-                        operations,
-                        tasks,
-                    },
+                    op: IndexOperation::DocumentDeletion { index_uid, tasks },
                     must_create_index,
                 }))
             }
@@ -1281,6 +1247,7 @@ impl IndexScheduler {
                 operations,
                 mut tasks,
             } => {
+                let started_processing_at = std::time::Instant::now();
                 let mut primary_key_has_been_set = false;
                 let must_stop_processing = self.must_stop_processing.clone();
                 let indexer_config = self.index_mapper.indexer_config();
@@ -1395,7 +1362,7 @@ impl IndexScheduler {
 
                 if !tasks.iter().all(|res| res.error.is_some()) {
                     let addition = builder.execute()?;
-                    tracing::info!(indexing_result = ?addition, "document indexing done");
+                    tracing::info!(indexing_result = ?addition, processed_in = ?started_processing_at.elapsed(), "document indexing done");
                 } else if primary_key_has_been_set {
                     // Everything failed but we've set a primary key.
                     // We need to remove it.
@@ -1438,7 +1405,7 @@ impl IndexScheduler {
                 {
                     (original_filter, context, function)
                 } else {
-                    // In the case of a `documentDeleteByFilter` the details MUST be set
+                    // In the case of a `documentEdition` the details MUST be set
                     unreachable!();
                 };
 
@@ -1468,52 +1435,102 @@ impl IndexScheduler {
 
                 Ok(vec![task])
             }
-            IndexOperation::IndexDocumentDeletionByFilter { mut task, index_uid: _ } => {
-                let filter =
-                    if let KindWithContent::DocumentDeletionByFilter { filter_expr, .. } =
-                        &task.kind
-                    {
-                        filter_expr
-                    } else {
-                        unreachable!()
-                    };
-                let deleted_documents = delete_document_by_filter(
-                    index_wtxn,
-                    filter,
-                    self.index_mapper.indexer_config(),
-                    self.must_stop_processing.clone(),
-                    index,
-                );
-                let original_filter = if let Some(Details::DocumentDeletionByFilter {
-                    original_filter,
-                    deleted_documents: _,
-                }) = task.details
-                {
-                    original_filter
-                } else {
-                    // In the case of a `documentDeleteByFilter` the details MUST be set
-                    unreachable!();
-                };
+            IndexOperation::DocumentDeletion { mut tasks, index_uid: _ } => {
+                let mut to_delete = RoaringBitmap::new();
+                let external_documents_ids = index.external_documents_ids();
 
-                match deleted_documents {
-                    Ok(deleted_documents) => {
-                        task.status = Status::Succeeded;
-                        task.details = Some(Details::DocumentDeletionByFilter {
-                            original_filter,
-                            deleted_documents: Some(deleted_documents),
-                        });
-                    }
-                    Err(e) => {
-                        task.status = Status::Failed;
-                        task.details = Some(Details::DocumentDeletionByFilter {
-                            original_filter,
-                            deleted_documents: Some(0),
-                        });
-                        task.error = Some(e.into());
+                for task in tasks.iter_mut() {
+                    let before = to_delete.len();
+                    task.status = Status::Succeeded;
+
+                    match &task.kind {
+                        KindWithContent::DocumentDeletion { index_uid: _, documents_ids } => {
+                            for id in documents_ids {
+                                if let Some(id) = external_documents_ids.get(index_wtxn, id)? {
+                                    to_delete.insert(id);
+                                }
+                            }
+                            let will_be_removed = to_delete.len() - before;
+                            task.details = Some(Details::DocumentDeletion {
+                                provided_ids: documents_ids.len(),
+                                deleted_documents: Some(will_be_removed),
+                            });
+                        }
+                        KindWithContent::DocumentDeletionByFilter { index_uid: _, filter_expr } => {
+                            let before = to_delete.len();
+                            let filter = match Filter::from_json(filter_expr) {
+                                Ok(filter) => filter,
+                                Err(err) => {
+                                    // theorically, this should be catched by deserr before reaching the index-scheduler and cannot happens
+                                    task.status = Status::Failed;
+                                    task.error = match err {
+                                        milli::Error::UserError(
+                                            milli::UserError::InvalidFilterExpression { .. },
+                                        ) => Some(
+                                            Error::from(err)
+                                                .with_custom_error_code(Code::InvalidDocumentFilter)
+                                                .into(),
+                                        ),
+                                        e => Some(e.into()),
+                                    };
+                                    None
+                                }
+                            };
+                            if let Some(filter) = filter {
+                                let candidates =
+                                    filter.evaluate(index_wtxn, index).map_err(|err| match err {
+                                        milli::Error::UserError(
+                                            milli::UserError::InvalidFilter(_),
+                                        ) => Error::from(err)
+                                            .with_custom_error_code(Code::InvalidDocumentFilter),
+                                        e => e.into(),
+                                    });
+                                match candidates {
+                                    Ok(candidates) => to_delete |= candidates,
+                                    Err(err) => {
+                                        task.status = Status::Failed;
+                                        task.error = Some(err.into());
+                                    }
+                                };
+                            }
+                            let will_be_removed = to_delete.len() - before;
+                            if let Some(Details::DocumentDeletionByFilter {
+                                original_filter: _,
+                                deleted_documents,
+                            }) = &mut task.details
+                            {
+                                *deleted_documents = Some(will_be_removed);
+                            } else {
+                                // In the case of a `documentDeleteByFilter` the details MUST be set
+                                unreachable!()
+                            }
+                        }
+                        _ => unreachable!(),
                     }
                 }
 
-                Ok(vec![task])
+                let config = IndexDocumentsConfig {
+                    update_method: IndexDocumentsMethod::ReplaceDocuments,
+                    ..Default::default()
+                };
+
+                let must_stop_processing = self.must_stop_processing.clone();
+                let mut builder = milli::update::IndexDocuments::new(
+                    index_wtxn,
+                    index,
+                    self.index_mapper.indexer_config(),
+                    config,
+                    |indexing_step| tracing::debug!(update = ?indexing_step),
+                    || must_stop_processing.get(),
+                )?;
+
+                let (new_builder, _count) =
+                    builder.remove_documents_from_db_no_batch(&to_delete)?;
+                builder = new_builder;
+
+                let _ = builder.execute()?;
+
+                Ok(tasks)
             }
             IndexOperation::Settings { index_uid: _, settings, mut tasks } => {
                 let indexer_config = self.index_mapper.indexer_config();
@@ -1717,46 +1734,6 @@ impl IndexScheduler {
     }
 }
 
-fn delete_document_by_filter<'a>(
-    wtxn: &mut RwTxn<'a>,
-    filter: &serde_json::Value,
-    indexer_config: &IndexerConfig,
-    must_stop_processing: MustStopProcessing,
-    index: &'a Index,
-) -> Result<u64> {
-    let filter = Filter::from_json(filter)?;
-    Ok(if let Some(filter) = filter {
-        let candidates = filter.evaluate(wtxn, index).map_err(|err| match err {
-            milli::Error::UserError(milli::UserError::InvalidFilter(_)) => {
-                Error::from(err).with_custom_error_code(Code::InvalidDocumentFilter)
-            }
-            e => e.into(),
-        })?;
-
-        let config = IndexDocumentsConfig {
-            update_method: IndexDocumentsMethod::ReplaceDocuments,
-            ..Default::default()
-        };
-
-        let mut builder = milli::update::IndexDocuments::new(
-            wtxn,
-            index,
-            indexer_config,
-            config,
-            |indexing_step| tracing::debug!(update = ?indexing_step),
-            || must_stop_processing.get(),
-        )?;
-
-        let (new_builder, count) = builder.remove_documents_from_db_no_batch(&candidates)?;
-        builder = new_builder;
-
-        let _ = builder.execute()?;
-        count
-    } else {
-        0
-    })
-}
-
 fn edit_documents_by_function<'a>(
     wtxn: &mut RwTxn<'a>,
     filter: &Option<serde_json::Value>,

index-scheduler/src/error.rs

@@ -101,7 +101,7 @@ pub enum Error {
     )]
     InvalidTaskCanceledBy { canceled_by: String },
     #[error(
-        "{index_uid} is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_)."
+        "{index_uid} is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_), and can not be more than 512 bytes."
     )]
     InvalidIndexUid { index_uid: String },
     #[error("Task `{0}` not found.")]

index-scheduler/src/features.rs

@@ -81,6 +81,19 @@ impl RoFeatures {
             .into())
         }
     }
+
+    pub fn check_contains_filter(&self) -> Result<()> {
+        if self.runtime.contains_filter {
+            Ok(())
+        } else {
+            Err(FeatureNotEnabledError {
+                disabled_action: "Using `CONTAINS` or `STARTS WITH` in a filter",
+                feature: "contains filter",
+                issue_link: "https://github.com/orgs/meilisearch/discussions/763",
+            }
+            .into())
+        }
+    }
 }
 
 impl FeatureData {
@@ -92,9 +105,11 @@ impl FeatureData {
         let txn = env.read_txn()?;
         let persisted_features: RuntimeTogglableFeatures =
             runtime_features_db.get(&txn, EXPERIMENTAL_FEATURES)?.unwrap_or_default();
+        let InstanceTogglableFeatures { metrics, logs_route, contains_filter } = instance_features;
         let runtime = Arc::new(RwLock::new(RuntimeTogglableFeatures {
-            metrics: instance_features.metrics || persisted_features.metrics,
-            logs_route: instance_features.logs_route || persisted_features.logs_route,
+            metrics: metrics || persisted_features.metrics,
+            logs_route: logs_route || persisted_features.logs_route,
+            contains_filter: contains_filter || persisted_features.contains_filter,
             ..persisted_features
         }));
 

index-scheduler/src/index_mapper/mod.rs

@@ -108,8 +108,10 @@ pub struct IndexStats {
     /// Association of every field name with the number of times it occurs in the documents.
     pub field_distribution: FieldDistribution,
     /// Creation date of the index.
+    #[serde(with = "time::serde::rfc3339")]
     pub created_at: OffsetDateTime,
     /// Date of the last update of the index.
+    #[serde(with = "time::serde::rfc3339")]
     pub updated_at: OffsetDateTime,
 }
 

index-scheduler/src/insta_snapshot.rs

@@ -11,6 +11,9 @@ use crate::index_mapper::IndexMapper;
 use crate::{IndexScheduler, Kind, Status, BEI128};
 
 pub fn snapshot_index_scheduler(scheduler: &IndexScheduler) -> String {
+    // Since we'll snapshot the index right afterward, we don't need to ensure it's internally consistent for every run.
+    // We can only do it for the release run, where the function runs way faster.
+    #[cfg(not(debug_assertions))]
     scheduler.assert_internally_consistent();
 
     let IndexScheduler {

index-scheduler/src/lib.rs

@@ -35,6 +35,7 @@ pub type TaskId = u32;
 use std::collections::{BTreeMap, HashMap};
 use std::io::{self, BufReader, Read};
 use std::ops::{Bound, RangeBounds};
+use std::panic::{catch_unwind, AssertUnwindSafe};
 use std::path::{Path, PathBuf};
 use std::sync::atomic::Ordering::{self, Relaxed};
 use std::sync::atomic::{AtomicBool, AtomicU32};
@@ -612,19 +613,24 @@ impl IndexScheduler {
                 #[cfg(test)]
                 run.breakpoint(Breakpoint::Init);
 
-                run.wake_up.wait();
+                run.wake_up.wait_timeout(std::time::Duration::from_secs(60));
 
                 loop {
-                    match run.tick() {
-                        Ok(TickOutcome::TickAgain(_)) => (),
-                        Ok(TickOutcome::WaitForSignal) => run.wake_up.wait(),
-                        Err(e) => {
+                    let ret = catch_unwind(AssertUnwindSafe(|| run.tick()));
+                    match ret {
+                        Ok(Ok(TickOutcome::TickAgain(_))) => (),
+                        Ok(Ok(TickOutcome::WaitForSignal)) => run.wake_up.wait(),
+                        Ok(Err(e)) => {
                             tracing::error!("{e}");
                             // Wait one second when an irrecoverable error occurs.
                             if !e.is_recoverable() {
                                 std::thread::sleep(Duration::from_secs(1));
                             }
                         }
+                        Err(_panic) => {
+                            tracing::error!("Internal error: Unexpected panic in the `IndexScheduler::run` method.");
+
+                        }
                     }
                 }
             })
@@ -662,7 +668,11 @@ impl IndexScheduler {
         let rtxn = self.env.read_txn()?;
         self.index_mapper.index(&rtxn, name)
     }
-
+    /// Return the boolean referring if index exists.
+    pub fn index_exists(&self, name: &str) -> Result<bool> {
+        let rtxn = self.env.read_txn()?;
+        self.index_mapper.index_exists(&rtxn, name)
+    }
     /// Return the name of all indexes without opening them.
     pub fn index_names(&self) -> Result<Vec<String>> {
         let rtxn = self.env.read_txn()?;
@@ -1467,7 +1477,7 @@ impl IndexScheduler {
             .map(
                 |IndexEmbeddingConfig {
                      name,
-                     config: milli::vector::EmbeddingConfig { embedder_options, prompt },
+                     config: milli::vector::EmbeddingConfig { embedder_options, prompt, quantized },
                      ..
                  }| {
                     let prompt =
@@ -1476,7 +1486,10 @@ impl IndexScheduler {
                     {
                         let embedders = self.embedders.read().unwrap();
                         if let Some(embedder) = embedders.get(&embedder_options) {
-                            return Ok((name, (embedder.clone(), prompt)));
+                            return Ok((
+                                name,
+                                (embedder.clone(), prompt, quantized.unwrap_or_default()),
+                            ));
                         }
                     }
 
@@ -1490,7 +1503,7 @@ impl IndexScheduler {
                         let mut embedders = self.embedders.write().unwrap();
                         embedders.insert(embedder_options, embedder.clone());
                     }
-                    Ok((name, (embedder, prompt)))
+                    Ok((name, (embedder, prompt, quantized.unwrap_or_default())))
                 },
             )
             .collect();
@@ -1754,6 +1767,7 @@ mod tests {
     use crossbeam::channel::RecvTimeoutError;
     use file_store::File;
     use insta::assert_json_snapshot;
+    use maplit::btreeset;
     use meili_snap::{json_string, snapshot};
     use meilisearch_auth::AuthFilter;
     use meilisearch_types::document_formats::DocumentFormatError;
@@ -1996,11 +2010,13 @@ mod tests {
         fn advance_till(&mut self, breakpoints: impl IntoIterator<Item = Breakpoint>) {
             for breakpoint in breakpoints {
                 let b = self.advance();
-                let state = snapshot_index_scheduler(&self.index_scheduler);
                 assert_eq!(
-                    b, breakpoint,
-                    "Was expecting the breakpoint `{:?}` but instead got `{:?}`.\n{state}",
-                    breakpoint, b
+                    b,
+                    breakpoint,
+                    "Was expecting the breakpoint `{:?}` but instead got `{:?}`.\n{}",
+                    breakpoint,
+                    b,
+                    snapshot_index_scheduler(&self.index_scheduler)
                 );
             }
         }
@@ -2024,7 +2040,6 @@ mod tests {
         // Wait for one successful batch.
         #[track_caller]
         fn advance_one_successful_batch(&mut self) {
-            self.index_scheduler.assert_internally_consistent();
             self.advance_till([Start, BatchCreated]);
             loop {
                 match self.advance() {
@@ -2043,7 +2058,6 @@ mod tests {
             }
 
             self.advance_till([AfterProcessing]);
-            self.index_scheduler.assert_internally_consistent();
         }
 
         // Wait for one failed batch.
@@ -2543,6 +2557,117 @@ mod tests {
         snapshot!(serde_json::to_string_pretty(&documents).unwrap(), name: "documents");
     }
 
+    #[test]
+    fn fail_in_process_batch_for_document_deletion() {
+        let (index_scheduler, mut handle) = IndexScheduler::test(true, vec![]);
+
+        use meilisearch_types::settings::{Settings, Unchecked};
+        let mut new_settings: Box<Settings<Unchecked>> = Box::default();
+        new_settings.filterable_attributes = Setting::Set(btreeset!(S("catto")));
+
+        index_scheduler
+            .register(
+                KindWithContent::SettingsUpdate {
+                    index_uid: S("doggos"),
+                    new_settings,
+                    is_deletion: false,
+                    allow_index_creation: true,
+                },
+                None,
+                false,
+            )
+            .unwrap();
+
+        let content = r#"[
+            { "id": 1, "doggo": "jean bob" },
+            { "id": 2, "catto": "jorts" },
+            { "id": 3, "doggo": "bork" }
+        ]"#;
+
+        let (uuid, mut file) = index_scheduler.create_update_file_with_uuid(0).unwrap();
+        let documents_count = read_json(content.as_bytes(), &mut file).unwrap();
+        file.persist().unwrap();
+        index_scheduler
+            .register(
+                KindWithContent::DocumentAdditionOrUpdate {
+                    index_uid: S("doggos"),
+                    primary_key: Some(S("id")),
+                    method: ReplaceDocuments,
+                    content_file: uuid,
+                    documents_count,
+                    allow_index_creation: true,
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "registered_the_setting_and_document_addition");
+
+        handle.advance_one_successful_batch();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_adding_the_settings");
+        handle.advance_one_successful_batch();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_adding_the_documents");
+
+        index_scheduler
+            .register(
+                KindWithContent::DocumentDeletion {
+                    index_uid: S("doggos"),
+                    documents_ids: vec![S("1")],
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        // This one should not be catched by Meilisearch but it's still nice to handle it because if one day we break the filters it could happens
+        index_scheduler
+            .register(
+                KindWithContent::DocumentDeletionByFilter {
+                    index_uid: S("doggos"),
+                    filter_expr: serde_json::json!(true),
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        // Should fail because the ids are not filterable
+        index_scheduler
+            .register(
+                KindWithContent::DocumentDeletionByFilter {
+                    index_uid: S("doggos"),
+                    filter_expr: serde_json::json!("id = 2"),
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        index_scheduler
+            .register(
+                KindWithContent::DocumentDeletionByFilter {
+                    index_uid: S("doggos"),
+                    filter_expr: serde_json::json!("catto EXISTS"),
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "registered_the_document_deletions");
+
+        // Everything should be batched together
+        handle.advance_one_successful_batch();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_removing_the_documents");
+
+        let index = index_scheduler.index("doggos").unwrap();
+        let rtxn = index.read_txn().unwrap();
+        let field_ids_map = index.fields_ids_map(&rtxn).unwrap();
+        let field_ids = field_ids_map.ids().collect::<Vec<_>>();
+        let documents = index
+            .all_documents(&rtxn)
+            .unwrap()
+            .map(|ret| obkv_to_json(&field_ids, &field_ids_map, ret.unwrap().1).unwrap())
+            .collect::<Vec<_>>();
+        snapshot!(serde_json::to_string_pretty(&documents).unwrap(), name: "documents_remaining_should_only_be_bork");
+    }
+
     #[test]
     fn do_not_batch_task_of_different_indexes() {
         let (index_scheduler, mut handle) = IndexScheduler::test(true, vec![]);
@@ -3043,6 +3168,8 @@ mod tests {
             api_key: Setting::Set(S("My super secret")),
             url: Setting::Set(S("http://localhost:7777")),
             dimensions: Setting::Set(4),
+            request: Setting::Set(serde_json::json!("{{text}}")),
+            response: Setting::Set(serde_json::json!("{{embedding}}")),
             ..Default::default()
         };
         embedders.insert(S("default"), Setting::Set(embedding_settings));
@@ -3795,15 +3922,15 @@ mod tests {
         ]);
         snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_processing_the_10_tasks");
 
-        // The index should not exists.
-        snapshot!(format!("{}", index_scheduler.index("doggos").map(|_| ()).unwrap_err()), @"Index `doggos` not found.");
+        // The index should not exist.
+        snapshot!(matches!(index_scheduler.index_exists("doggos"), Ok(true)), @"false");
     }
 
     #[test]
     fn test_document_addition_cant_create_index_without_index_without_autobatching() {
         // We're going to execute multiple document addition that don't have
         // the right to create an index while there is no index currently.
-        // Since the autobatching is disabled, every tasks should be processed
+        // Since the auto-batching is disabled, every task should be processed
         // sequentially and throw an IndexDoesNotExists.
         let (index_scheduler, mut handle) = IndexScheduler::test(false, vec![]);
 
@@ -3845,8 +3972,8 @@ mod tests {
         handle.advance_n_failed_batches(5);
         snapshot!(snapshot_index_scheduler(&index_scheduler), name: "all_tasks_processed");
 
-        // The index should not exists.
-        snapshot!(format!("{}", index_scheduler.index("doggos").map(|_| ()).unwrap_err()), @"Index `doggos` not found.");
+        // The index should not exist.
+        snapshot!(matches!(index_scheduler.index_exists("doggos"), Ok(true)), @"false");
     }
 
     #[test]
@@ -5002,6 +5129,8 @@ mod tests {
             api_key: Setting::Set(S("My super secret")),
             url: Setting::Set(S("http://localhost:7777")),
             dimensions: Setting::Set(384),
+            request: Setting::Set(serde_json::json!("{{text}}")),
+            response: Setting::Set(serde_json::json!("{{embedding}}")),
             ..Default::default()
         };
         embedders.insert(S("A_fakerest"), Setting::Set(embedding_settings));
@@ -5071,7 +5200,7 @@ mod tests {
             let simple_hf_name = name.clone();
 
             let configs = index_scheduler.embedders(configs).unwrap();
-            let (hf_embedder, _) = configs.get(&simple_hf_name).unwrap();
+            let (hf_embedder, _, _) = configs.get(&simple_hf_name).unwrap();
             let beagle_embed = hf_embedder.embed_one(S("Intel the beagle best doggo")).unwrap();
             let lab_embed = hf_embedder.embed_one(S("Max the lab best doggo")).unwrap();
             let patou_embed = hf_embedder.embed_one(S("kefir the patou best doggo")).unwrap();
@@ -5389,7 +5518,11 @@ mod tests {
                     ),
                     prompt: PromptData {
                         template: "{{doc.doggo}}",
+                        max_bytes: Some(
+                            400,
+                        ),
                     },
+                    quantized: None,
                 },
                 user_provided: RoaringBitmap<[1, 2]>,
             },
@@ -5402,28 +5535,8 @@ mod tests {
 
         // the document with the id 3 should keep its original embedding
         let docid = index.external_documents_ids.get(&rtxn, "3").unwrap().unwrap();
-        let mut embeddings = Vec::new();
-
-        'vectors: for i in 0..=u8::MAX {
-            let reader = arroy::Reader::open(&rtxn, i as u16, index.vector_arroy)
-                .map(Some)
-                .or_else(|e| match e {
-                    arroy::Error::MissingMetadata(_) => Ok(None),
-                    e => Err(e),
-                })
-                .transpose();
-
-            let Some(reader) = reader else {
-                break 'vectors;
-            };
-
-            let embedding = reader.unwrap().item_vector(&rtxn, docid).unwrap();
-            if let Some(embedding) = embedding {
-                embeddings.push(embedding)
-            } else {
-                break 'vectors;
-            }
-        }
+        let embeddings = index.embeddings(&rtxn, docid).unwrap();
+        let embeddings = &embeddings["my_doggo_embedder"];
 
         snapshot!(embeddings.len(), @"1");
         assert!(embeddings[0].iter().all(|i| *i == 3.0), "{:?}", embeddings[0]);
@@ -5603,8 +5716,12 @@ mod tests {
                         },
                     ),
                     prompt: PromptData {
-                        template: "{% for field in fields %} {{ field.name }}: {{ field.value }}\n{% endfor %}",
+                        template: "{% for field in fields %}{% if field.is_searchable and field.value != nil %}{{ field.name }}: {{ field.value }}\n{% endif %}{% endfor %}",
+                        max_bytes: Some(
+                            400,
+                        ),
                     },
+                    quantized: None,
                 },
                 user_provided: RoaringBitmap<[0]>,
             },
@@ -5643,8 +5760,12 @@ mod tests {
                         },
                     ),
                     prompt: PromptData {
-                        template: "{% for field in fields %} {{ field.name }}: {{ field.value }}\n{% endfor %}",
+                        template: "{% for field in fields %}{% if field.is_searchable and field.value != nil %}{{ field.name }}: {{ field.value }}\n{% endif %}{% endfor %}",
+                        max_bytes: Some(
+                            400,
+                        ),
                     },
+                    quantized: None,
                 },
                 user_provided: RoaringBitmap<[]>,
             },

index-scheduler/src/snapshots/index_scheduler__tests__import_vectors-2.snap

@@ -8,7 +8,9 @@ expression: task.details
       "source": "rest",
       "apiKey": "MyXXXX...",
       "dimensions": 384,
-      "url": "http://localhost:7777"
+      "url": "http://localhost:7777",
+      "request": "{{text}}",
+      "response": "{{embedding}}"
     },
     "B_small_hf": {
       "source": "huggingFace",

index-scheduler/src/snapshots/index_scheduler__tests__import_vectors-5.snap

@@ -8,16 +8,8 @@ expression: fakerest_config.embedder_options
     "distribution": null,
     "dimensions": 384,
     "url": "http://localhost:7777",
-    "query": null,
-    "input_field": [
-      "input"
-    ],
-    "path_to_embeddings": [
-      "data"
-    ],
-    "embedding_object": [
-      "embedding"
-    ],
-    "input_type": "text"
+    "request": "{{text}}",
+    "response": "{{embedding}}",
+    "headers": {}
   }
 }

index-scheduler/src/snapshots/index_scheduler__tests__import_vectors.snap

@@ -8,7 +8,9 @@ expression: task.details
       "source": "rest",
       "apiKey": "MyXXXX...",
       "dimensions": 384,
-      "url": "http://localhost:7777"
+      "url": "http://localhost:7777",
+      "request": "{{text}}",
+      "response": "{{embedding}}"
     },
     "B_small_hf": {
       "source": "huggingFace",

index-scheduler/src/snapshots/index_scheduler__tests__settings_update-2.snap

@@ -8,7 +8,9 @@ expression: task.details
       "source": "rest",
       "apiKey": "MyXXXX...",
       "dimensions": 4,
-      "url": "http://localhost:7777"
+      "url": "http://localhost:7777",
+      "request": "{{text}}",
+      "response": "{{embedding}}"
     }
   }
 }

index-scheduler/src/snapshots/index_scheduler__tests__settings_update-5.snap

@@ -1,6 +1,6 @@
 ---
 source: index-scheduler/src/lib.rs
-expression: embedding_config.embedder_options
+expression: config.embedder_options
 ---
 {
   "Rest": {
@@ -8,16 +8,8 @@ expression: embedding_config.embedder_options
     "distribution": null,
     "dimensions": 4,
     "url": "http://localhost:7777",
-    "query": null,
-    "input_field": [
-      "input"
-    ],
-    "path_to_embeddings": [
-      "data"
-    ],
-    "embedding_object": [
-      "embedding"
-    ],
-    "input_type": "text"
+    "request": "{{text}}",
+    "response": "{{embedding}}",
+    "headers": {}
   }
 }

index-scheduler/src/snapshots/index_scheduler__tests__settings_update.snap

@@ -8,7 +8,9 @@ expression: task.details
       "source": "rest",
       "apiKey": "MyXXXX...",
       "dimensions": 4,
-      "url": "http://localhost:7777"
+      "url": "http://localhost:7777",
+      "request": "{{text}}",
+      "response": "{{embedding}}"
     }
   }
 }

index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/after_adding_the_documents.snap

@@ -0,0 +1,44 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: succeeded, details: { received_documents: 3, indexed_documents: Some(3) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued []
+succeeded [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 3, field_distribution: {"catto": 1, "doggo": 2, "id": 3} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------

index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/after_adding_the_settings.snap

@@ -0,0 +1,43 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: enqueued, details: { received_documents: 3, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued [1,]
+succeeded [0,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 0, field_distribution: {} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------

index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/after_adding_the_settings_and_documents.snap

@@ -0,0 +1,43 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: enqueued, details: { received_documents: 3, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued [1,]
+succeeded [0,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 0, field_distribution: {} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------

index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/after_removing_the_documents.snap

@@ -0,0 +1,56 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: succeeded, details: { received_documents: 3, indexed_documents: Some(3) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
+2 {uid: 2, status: succeeded, details: { received_document_ids: 1, deleted_documents: Some(1) }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1"] }}
+3 {uid: 3, status: failed, error: ResponseError { code: 200, message: "Invalid type for filter subexpression: expected: String, Array, found: true.", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: true, deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: Bool(true) }}
+4 {uid: 4, status: failed, error: ResponseError { code: 200, message: "Attribute `id` is not filterable. Available filterable attributes are: `catto`.\n1:3 id = 2", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: "id = 2", deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("id = 2") }}
+5 {uid: 5, status: succeeded, details: { original_filter: "catto EXISTS", deleted_documents: Some(1) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("catto EXISTS") }}
+----------------------------------------------------------------------
+### Status:
+enqueued []
+succeeded [0,1,2,5,]
+failed [3,4,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"documentDeletion" [2,3,4,5,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,2,3,4,5,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 1, field_distribution: {"doggo": 1, "id": 1} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,]
+[timestamp] [3,]
+[timestamp] [4,]
+[timestamp] [5,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,3,4,5,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,3,4,5,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------

index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/documents_remaining_should_only_be_bork.snap

@@ -0,0 +1,9 @@
+---
+source: index-scheduler/src/lib.rs
+---
+[
+  {
+    "id": 3,
+    "doggo": "bork"
+  }
+]

index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/registered_the_document_deletions.snap

@@ -0,0 +1,53 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: succeeded, details: { received_documents: 3, indexed_documents: Some(3) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
+2 {uid: 2, status: enqueued, details: { received_document_ids: 1, deleted_documents: None }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1"] }}
+3 {uid: 3, status: enqueued, details: { original_filter: true, deleted_documents: None }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: Bool(true) }}
+4 {uid: 4, status: enqueued, details: { original_filter: "id = 2", deleted_documents: None }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("id = 2") }}
+5 {uid: 5, status: enqueued, details: { original_filter: "catto EXISTS", deleted_documents: None }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("catto EXISTS") }}
+----------------------------------------------------------------------
+### Status:
+enqueued [2,3,4,5,]
+succeeded [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"documentDeletion" [2,3,4,5,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,2,3,4,5,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 3, field_distribution: {"catto": 1, "doggo": 2, "id": 3} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,]
+[timestamp] [3,]
+[timestamp] [4,]
+[timestamp] [5,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------

index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/registered_the_setting_and_document_addition.snap

@@ -0,0 +1,39 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: enqueued, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: enqueued, details: { received_documents: 3, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+----------------------------------------------------------------------
+### Finished At:
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------

index-scheduler/src/snapshots/lib.rs/import_vectors/Intel to kefir succeeds.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 1 {uid: 1, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
 2 {uid: 2, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: None, method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000001, documents_count: 1, allow_index_creation: true }}
 ----------------------------------------------------------------------
@@ -46,4 +46,3 @@ doggos: { number_of_documents: 1, field_distribution: {"_vectors": 1, "breed": 1
 ### File Store:
 
 ----------------------------------------------------------------------
-

index-scheduler/src/snapshots/lib.rs/import_vectors/Intel to kefir.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 1 {uid: 1, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
 2 {uid: 2, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: None, method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000001, documents_count: 1, allow_index_creation: true }}
 ----------------------------------------------------------------------
@@ -45,4 +45,3 @@ doggos: { number_of_documents: 1, field_distribution: {"_vectors": 1, "breed": 1
 00000000-0000-0000-0000-000000000001
 
 ----------------------------------------------------------------------
-

index-scheduler/src/snapshots/lib.rs/import_vectors/adding Intel succeeds.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 1 {uid: 1, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
 ----------------------------------------------------------------------
 ### Status:
@@ -42,4 +42,3 @@ doggos: { number_of_documents: 1, field_distribution: {"_vectors": 1, "breed": 1
 ### File Store:
 
 ----------------------------------------------------------------------
-

index-scheduler/src/snapshots/lib.rs/import_vectors/after adding Intel.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 1 {uid: 1, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
 ----------------------------------------------------------------------
 ### Status:
@@ -41,4 +41,3 @@ doggos: { number_of_documents: 0, field_distribution: {} }
 00000000-0000-0000-0000-000000000000
 
 ----------------------------------------------------------------------
-

index-scheduler/src/snapshots/lib.rs/import_vectors/after_registering_settings_task_vectors.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: enqueued, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: enqueued, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 ----------------------------------------------------------------------
 ### Status:
 enqueued [0,]
@@ -33,4 +33,3 @@ doggos [0,]
 ### File Store:
 
 ----------------------------------------------------------------------
-

index-scheduler/src/snapshots/lib.rs/import_vectors/settings_update_processed_vectors.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, binary_quantized: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), document_template_max_bytes: NotSet, url: NotSet, request: NotSet, response: NotSet, headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 ----------------------------------------------------------------------
 ### Status:
 enqueued []
@@ -37,4 +37,3 @@ doggos: { number_of_documents: 0, field_distribution: {} }
 ### File Store:
 
 ----------------------------------------------------------------------
-

index-scheduler/src/snapshots/lib.rs/test_settings_update/after_registering_settings_task.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: enqueued, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: enqueued, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 ----------------------------------------------------------------------
 ### Status:
 enqueued [0,]
@@ -33,4 +33,3 @@ doggos [0,]
 ### File Store:
 
 ----------------------------------------------------------------------
-

index-scheduler/src/snapshots/lib.rs/test_settings_update/settings_update_processed.snap

@@ -6,7 +6,7 @@ source: index-scheduler/src/lib.rs
 []
 ----------------------------------------------------------------------
 ### All Tasks:
-0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"default": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(4), binary_quantized: NotSet, document_template: NotSet, document_template_max_bytes: NotSet, url: Set("http://localhost:7777"), request: Set(String("{{text}}")), response: Set(String("{{embedding}}")), headers: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, localized_attributes: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 ----------------------------------------------------------------------
 ### Status:
 enqueued []
@@ -37,4 +37,3 @@ doggos: { number_of_documents: 0, field_distribution: {} }
 ### File Store:
 
 ----------------------------------------------------------------------
-

meilisearch-types/Cargo.toml

@@ -66,3 +66,5 @@ khmer = ["milli/khmer"]
 vietnamese = ["milli/vietnamese"]
 # force swedish character recomposition
 swedish-recomposition = ["milli/swedish-recomposition"]
+# force german character recomposition
+german = ["milli/german"]

meilisearch-types/src/error.rs

@@ -238,8 +238,14 @@ InvalidIndexLimit                     , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexOffset                    , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexPrimaryKey                , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexUid                       , InvalidRequest       , BAD_REQUEST ;
+InvalidMultiSearchFacets              , InvalidRequest       , BAD_REQUEST ;
+InvalidMultiSearchFacetsByIndex       , InvalidRequest       , BAD_REQUEST ;
+InvalidMultiSearchFacetOrder          , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFederated           , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFederationOptions   , InvalidRequest       , BAD_REQUEST ;
+InvalidMultiSearchMaxValuesPerFacet   , InvalidRequest       , BAD_REQUEST ;
+InvalidMultiSearchMergeFacets         , InvalidRequest       , BAD_REQUEST ;
+InvalidMultiSearchQueryFacets         , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchQueryPagination     , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchQueryRankingRules   , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchWeight              , InvalidRequest       , BAD_REQUEST ;
@@ -256,6 +262,7 @@ InvalidSearchCropLength               , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchCropMarker               , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchFacets                   , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchSemanticRatio            , InvalidRequest       , BAD_REQUEST ;
+InvalidSearchLocales                  , InvalidRequest       , BAD_REQUEST ;
 InvalidFacetSearchFacetName           , InvalidRequest       , BAD_REQUEST ;
 InvalidSimilarId                      , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchFilter                   , InvalidRequest       , BAD_REQUEST ;
@@ -297,6 +304,7 @@ InvalidSettingsSeparatorTokens        , InvalidRequest       , BAD_REQUEST ;
 InvalidSettingsDictionary             , InvalidRequest       , BAD_REQUEST ;
 InvalidSettingsSynonyms               , InvalidRequest       , BAD_REQUEST ;
 InvalidSettingsTypoTolerance          , InvalidRequest       , BAD_REQUEST ;
+InvalidSettingsLocalizedAttributes    , InvalidRequest       , BAD_REQUEST ;
 InvalidState                          , Internal             , INTERNAL_SERVER_ERROR ;
 InvalidStoreFile                      , Internal             , INTERNAL_SERVER_ERROR ;
 InvalidSwapDuplicateIndexFound        , InvalidRequest       , BAD_REQUEST ;
@@ -386,7 +394,11 @@ impl ErrorCode for milli::Error {
                     | UserError::InvalidOpenAiModelDimensionsMax { .. }
                     | UserError::InvalidSettingsDimensions { .. }
                     | UserError::InvalidUrl { .. }
-                    | UserError::InvalidPrompt(_) => Code::InvalidSettingsEmbedders,
+                    | UserError::InvalidSettingsDocumentTemplateMaxBytes { .. }
+                    | UserError::InvalidPrompt(_)
+                    | UserError::InvalidDisableBinaryQuantization { .. } => {
+                        Code::InvalidSettingsEmbedders
+                    }
                     UserError::TooManyEmbedders(_) => Code::InvalidSettingsEmbedders,
                     UserError::InvalidPromptForEmbeddings(..) => Code::InvalidSettingsEmbedders,
                     UserError::NoPrimaryKeyCandidateFound => Code::IndexPrimaryKeyNoCandidateFound,
@@ -415,7 +427,9 @@ impl ErrorCode for milli::Error {
                         Code::InvalidSettingsTypoTolerance
                     }
                     UserError::InvalidEmbedder(_) => Code::InvalidEmbedder,
-                    UserError::VectorEmbeddingError(_) => Code::VectorEmbeddingError,
+                    UserError::VectorEmbeddingError(_) | UserError::DocumentEmbeddingError(_) => {
+                        Code::VectorEmbeddingError
+                    }
                     UserError::DocumentEditionCannotModifyPrimaryKey
                     | UserError::DocumentEditionDocumentMustBeObject
                     | UserError::DocumentEditionRuntimeError(_)
@@ -529,7 +543,8 @@ impl fmt::Display for deserr_codes::InvalidSimilarId {
             f,
             "the value of `id` is invalid. \
             A document identifier can be of type integer or string, \
-            only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_)."
+            only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), \
+            and can not be more than 512 bytes."
         )
     }
 }

meilisearch-types/src/features.rs

@@ -7,10 +7,12 @@ pub struct RuntimeTogglableFeatures {
     pub metrics: bool,
     pub logs_route: bool,
     pub edit_documents_by_function: bool,
+    pub contains_filter: bool,
 }
 
 #[derive(Default, Debug, Clone, Copy)]
 pub struct InstanceTogglableFeatures {
     pub metrics: bool,
     pub logs_route: bool,
+    pub contains_filter: bool,
 }

meilisearch-types/src/index_uid.rs

@@ -1,3 +1,4 @@
+use std::borrow::Borrow;
 use std::error::Error;
 use std::fmt;
 use std::str::FromStr;
@@ -8,7 +9,7 @@ use crate::error::{Code, ErrorCode};
 
 /// An index uid is composed of only ascii alphanumeric characters, - and _, between 1 and 400
 /// bytes long
-#[derive(Debug, Clone, PartialEq, Eq, Deserr)]
+#[derive(Debug, Clone, PartialEq, Eq, Deserr, PartialOrd, Ord)]
 #[deserr(try_from(String) = IndexUid::try_from -> IndexUidFormatError)]
 pub struct IndexUid(String);
 
@@ -70,6 +71,12 @@ impl From<IndexUid> for String {
     }
 }
 
+impl Borrow<String> for IndexUid {
+    fn borrow(&self) -> &String {
+        &self.0
+    }
+}
+
 #[derive(Debug)]
 pub struct IndexUidFormatError {
     pub invalid_uid: String,
@@ -81,7 +88,8 @@ impl fmt::Display for IndexUidFormatError {
             f,
             "`{}` is not a valid index uid. Index uid can be an \
             integer or a string containing only alphanumeric \
-            characters, hyphens (-) and underscores (_).",
+            characters, hyphens (-) and underscores (_), \
+            and can not be more than 512 bytes.",
             self.invalid_uid,
         )
     }

meilisearch-types/src/lib.rs

@@ -7,6 +7,7 @@ pub mod features;
 pub mod index_uid;
 pub mod index_uid_pattern;
 pub mod keys;
+pub mod locales;
 pub mod settings;
 pub mod star_or;
 pub mod task_view;

meilisearch-types/src/locales.rs

@@ -0,0 +1,166 @@
+use deserr::Deserr;
+use milli::LocalizedAttributesRule;
+use serde::{Deserialize, Serialize};
+
+#[derive(Debug, Clone, PartialEq, Eq, Deserr, Serialize, Deserialize)]
+#[deserr(rename_all = camelCase)]
+#[serde(rename_all = "camelCase")]
+pub struct LocalizedAttributesRuleView {
+    pub attribute_patterns: Vec<String>,
+    pub locales: Vec<Locale>,
+}
+
+impl From<LocalizedAttributesRule> for LocalizedAttributesRuleView {
+    fn from(rule: LocalizedAttributesRule) -> Self {
+        Self {
+            attribute_patterns: rule.attribute_patterns,
+            locales: rule.locales.into_iter().map(|l| l.into()).collect(),
+        }
+    }
+}
+
+impl From<LocalizedAttributesRuleView> for LocalizedAttributesRule {
+    fn from(view: LocalizedAttributesRuleView) -> Self {
+        Self {
+            attribute_patterns: view.attribute_patterns,
+            locales: view.locales.into_iter().map(|l| l.into()).collect(),
+        }
+    }
+}
+
+/// Generate a Locale enum and its From and Into implementations for milli::tokenizer::Language.
+///
+/// this enum implements `Deserr` in order to be used in the API.
+macro_rules! make_locale {
+    ($(($iso_639_1:ident, $iso_639_1_str:expr) => ($iso_639_3:ident, $iso_639_3_str:expr),)+) => {
+        #[derive(Debug, Copy, Clone, PartialEq, Eq, Deserr, Serialize, Deserialize, Ord, PartialOrd)]
+        #[deserr(rename_all = camelCase)]
+        #[serde(rename_all = "camelCase")]
+        pub enum Locale {
+            $($iso_639_1,)+
+            $($iso_639_3,)+
+            Cmn,
+        }
+
+        impl From<milli::tokenizer::Language> for Locale {
+            fn from(other: milli::tokenizer::Language) -> Locale {
+                match other {
+                    $(milli::tokenizer::Language::$iso_639_3 => Locale::$iso_639_3,)+
+                    milli::tokenizer::Language::Cmn => Locale::Cmn,
+                }
+            }
+        }
+
+        impl From<Locale> for milli::tokenizer::Language {
+            fn from(other: Locale) -> milli::tokenizer::Language {
+                match other {
+                    $(Locale::$iso_639_1 => milli::tokenizer::Language::$iso_639_3,)+
+                    $(Locale::$iso_639_3 => milli::tokenizer::Language::$iso_639_3,)+
+                    Locale::Cmn => milli::tokenizer::Language::Cmn,
+                }
+            }
+        }
+
+        impl std::str::FromStr for Locale {
+            type Err = LocaleFormatError;
+
+            fn from_str(s: &str) -> Result<Self, Self::Err> {
+                let locale = match s {
+                    $($iso_639_1_str => Locale::$iso_639_1,)+
+                    $($iso_639_3_str => Locale::$iso_639_3,)+
+                    "cmn" => Locale::Cmn,
+                    _ => return Err(LocaleFormatError { invalid_locale: s.to_string() }),
+                };
+
+                Ok(locale)
+            }
+        }
+
+        #[derive(Debug)]
+        pub struct LocaleFormatError {
+            pub invalid_locale: String,
+        }
+
+        impl std::fmt::Display for LocaleFormatError {
+            fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+                let mut valid_locales = [$($iso_639_1_str),+,$($iso_639_3_str),+,"cmn"];
+                valid_locales.sort_by(|left, right| left.len().cmp(&right.len()).then(left.cmp(right)));
+                write!(f, "Unsupported locale `{}`, expected one of {}", self.invalid_locale, valid_locales.join(", "))
+            }
+        }
+
+        impl std::error::Error for LocaleFormatError {}
+    };
+}
+
+make_locale!(
+    (Af, "af") => (Afr, "afr"),
+    (Ak, "ak") => (Aka, "aka"),
+    (Am, "am") => (Amh, "amh"),
+    (Ar, "ar") => (Ara, "ara"),
+    (Az, "az") => (Aze, "aze"),
+    (Be, "be") => (Bel, "bel"),
+    (Bn, "bn") => (Ben, "ben"),
+    (Bg, "bg") => (Bul, "bul"),
+    (Ca, "ca") => (Cat, "cat"),
+    (Cs, "cs") => (Ces, "ces"),
+    (Da, "da") => (Dan, "dan"),
+    (De, "de") => (Deu, "deu"),
+    (El, "el") => (Ell, "ell"),
+    (En, "en") => (Eng, "eng"),
+    (Eo, "eo") => (Epo, "epo"),
+    (Et, "et") => (Est, "est"),
+    (Fi, "fi") => (Fin, "fin"),
+    (Fr, "fr") => (Fra, "fra"),
+    (Gu, "gu") => (Guj, "guj"),
+    (He, "he") => (Heb, "heb"),
+    (Hi, "hi") => (Hin, "hin"),
+    (Hr, "hr") => (Hrv, "hrv"),
+    (Hu, "hu") => (Hun, "hun"),
+    (Hy, "hy") => (Hye, "hye"),
+    (Id, "id") => (Ind, "ind"),
+    (It, "it") => (Ita, "ita"),
+    (Jv, "jv") => (Jav, "jav"),
+    (Ja, "ja") => (Jpn, "jpn"),
+    (Kn, "kn") => (Kan, "kan"),
+    (Ka, "ka") => (Kat, "kat"),
+    (Km, "km") => (Khm, "khm"),
+    (Ko, "ko") => (Kor, "kor"),
+    (La, "la") => (Lat, "lat"),
+    (Lv, "lv") => (Lav, "lav"),
+    (Lt, "lt") => (Lit, "lit"),
+    (Ml, "ml") => (Mal, "mal"),
+    (Mr, "mr") => (Mar, "mar"),
+    (Mk, "mk") => (Mkd, "mkd"),
+    (My, "my") => (Mya, "mya"),
+    (Ne, "ne") => (Nep, "nep"),
+    (Nl, "nl") => (Nld, "nld"),
+    (Nb, "nb") => (Nob, "nob"),
+    (Or, "or") => (Ori, "ori"),
+    (Pa, "pa") => (Pan, "pan"),
+    (Fa, "fa") => (Pes, "pes"),
+    (Pl, "pl") => (Pol, "pol"),
+    (Pt, "pt") => (Por, "por"),
+    (Ro, "ro") => (Ron, "ron"),
+    (Ru, "ru") => (Rus, "rus"),
+    (Si, "si") => (Sin, "sin"),
+    (Sk, "sk") => (Slk, "slk"),
+    (Sl, "sl") => (Slv, "slv"),
+    (Sn, "sn") => (Sna, "sna"),
+    (Es, "es") => (Spa, "spa"),
+    (Sr, "sr") => (Srp, "srp"),
+    (Sv, "sv") => (Swe, "swe"),
+    (Ta, "ta") => (Tam, "tam"),
+    (Te, "te") => (Tel, "tel"),
+    (Tl, "tl") => (Tgl, "tgl"),
+    (Th, "th") => (Tha, "tha"),
+    (Tk, "tk") => (Tuk, "tuk"),
+    (Tr, "tr") => (Tur, "tur"),
+    (Uk, "uk") => (Ukr, "ukr"),
+    (Ur, "ur") => (Urd, "urd"),
+    (Uz, "uz") => (Uzb, "uzb"),
+    (Vi, "vi") => (Vie, "vie"),
+    (Yi, "yi") => (Yid, "yid"),
+    (Zh, "zh") => (Zho, "zho"),
+    (Zu, "zu") => (Zul, "zul"),
+);

meilisearch-types/src/settings.rs

@@ -17,6 +17,7 @@ use serde::{Deserialize, Serialize, Serializer};
 use crate::deserr::DeserrJsonError;
 use crate::error::deserr_codes::*;
 use crate::facet_values_sort::FacetValuesSort;
+use crate::locales::LocalizedAttributesRuleView;
 
 /// The maximum number of results that the engine
 /// will be able to return in one search call.
@@ -198,6 +199,9 @@ pub struct Settings<T> {
     #[serde(default, skip_serializing_if = "Setting::is_not_set")]
     #[deserr(default, error = DeserrJsonError<InvalidSettingsSearchCutoffMs>)]
     pub search_cutoff_ms: Setting<u64>,
+    #[serde(default, skip_serializing_if = "Setting::is_not_set")]
+    #[deserr(default, error = DeserrJsonError<InvalidSettingsLocalizedAttributes>)]
+    pub localized_attributes: Setting<Vec<LocalizedAttributesRuleView>>,
 
     #[serde(skip)]
     #[deserr(skip)]
@@ -261,6 +265,7 @@ impl Settings<Checked> {
             pagination: Setting::Reset,
             embedders: Setting::Reset,
             search_cutoff_ms: Setting::Reset,
+            localized_attributes: Setting::Reset,
             _kind: PhantomData,
         }
     }
@@ -284,7 +289,8 @@ impl Settings<Checked> {
             pagination,
             embedders,
             search_cutoff_ms,
-            ..
+            localized_attributes: localized_attributes_rules,
+            _kind,
         } = self;
 
         Settings {
@@ -305,6 +311,7 @@ impl Settings<Checked> {
             pagination,
             embedders,
             search_cutoff_ms,
+            localized_attributes: localized_attributes_rules,
             _kind: PhantomData,
         }
     }
@@ -352,6 +359,7 @@ impl Settings<Unchecked> {
             pagination: self.pagination,
             embedders: self.embedders,
             search_cutoff_ms: self.search_cutoff_ms,
+            localized_attributes: self.localized_attributes,
             _kind: PhantomData,
         }
     }
@@ -402,6 +410,7 @@ pub fn apply_settings_to_builder(
         pagination,
         embedders,
         search_cutoff_ms,
+        localized_attributes: localized_attributes_rules,
         _kind,
     } = settings;
 
@@ -485,6 +494,13 @@ pub fn apply_settings_to_builder(
         Setting::NotSet => (),
     }
 
+    match localized_attributes_rules {
+        Setting::Set(ref rules) => builder
+            .set_localized_attributes_rules(rules.iter().cloned().map(|r| r.into()).collect()),
+        Setting::Reset => builder.reset_localized_attributes_rules(),
+        Setting::NotSet => (),
+    }
+
     match typo_tolerance {
         Setting::Set(ref value) => {
             match value.enabled {
@@ -679,6 +695,8 @@ pub fn settings(
 
     let search_cutoff_ms = index.search_cutoff(rtxn)?;
 
+    let localized_attributes_rules = index.localized_attributes_rules(rtxn)?;
+
     let mut settings = Settings {
         displayed_attributes: match displayed_attributes {
             Some(attrs) => Setting::Set(attrs),
@@ -711,6 +729,10 @@ pub fn settings(
             Some(cutoff) => Setting::Set(cutoff),
             None => Setting::Reset,
         },
+        localized_attributes: match localized_attributes_rules {
+            Some(rules) => Setting::Set(rules.into_iter().map(|r| r.into()).collect()),
+            None => Setting::Reset,
+        },
         _kind: PhantomData,
     };
 
@@ -902,6 +924,7 @@ pub(crate) mod test {
             faceting: Setting::NotSet,
             pagination: Setting::NotSet,
             embedders: Setting::NotSet,
+            localized_attributes: Setting::NotSet,
             search_cutoff_ms: Setting::NotSet,
             _kind: PhantomData::<Unchecked>,
         };
@@ -930,6 +953,7 @@ pub(crate) mod test {
             faceting: Setting::NotSet,
             pagination: Setting::NotSet,
             embedders: Setting::NotSet,
+            localized_attributes: Setting::NotSet,
             search_cutoff_ms: Setting::NotSet,
             _kind: PhantomData::<Unchecked>,
         };

meilisearch-types/src/versioning.rs

@@ -10,38 +10,52 @@ static VERSION_MINOR: &str = env!("CARGO_PKG_VERSION_MINOR");
 static VERSION_PATCH: &str = env!("CARGO_PKG_VERSION_PATCH");
 
 /// Persists the version of the current Meilisearch binary to a VERSION file
-pub fn create_version_file(db_path: &Path) -> io::Result<()> {
+pub fn create_current_version_file(db_path: &Path) -> io::Result<()> {
+    create_version_file(db_path, VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH)
+}
+
+pub fn create_version_file(
+    db_path: &Path,
+    major: &str,
+    minor: &str,
+    patch: &str,
+) -> io::Result<()> {
     let version_path = db_path.join(VERSION_FILE_NAME);
-    fs::write(version_path, format!("{}.{}.{}", VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH))
+    fs::write(version_path, format!("{}.{}.{}", major, minor, patch))
 }
 
 /// Ensures Meilisearch version is compatible with the database, returns an error versions mismatch.
 pub fn check_version_file(db_path: &Path) -> anyhow::Result<()> {
-    let version_path = db_path.join(VERSION_FILE_NAME);
+    let (major, minor, patch) = get_version(db_path)?;
 
-    match fs::read_to_string(version_path) {
-        Ok(version) => {
-            let version_components = version.split('.').collect::<Vec<_>>();
-            let (major, minor, patch) = match &version_components[..] {
-                [major, minor, patch] => (major.to_string(), minor.to_string(), patch.to_string()),
-                _ => return Err(VersionFileError::MalformedVersionFile.into()),
-            };
-
-            if major != VERSION_MAJOR || minor != VERSION_MINOR {
-                return Err(VersionFileError::VersionMismatch { major, minor, patch }.into());
-            }
-        }
-        Err(error) => {
-            return match error.kind() {
-                ErrorKind::NotFound => Err(VersionFileError::MissingVersionFile.into()),
-                _ => Err(error.into()),
-            }
-        }
+    if major != VERSION_MAJOR || minor != VERSION_MINOR {
+        return Err(VersionFileError::VersionMismatch { major, minor, patch }.into());
     }
 
     Ok(())
 }
 
+pub fn get_version(db_path: &Path) -> Result<(String, String, String), VersionFileError> {
+    let version_path = db_path.join(VERSION_FILE_NAME);
+
+    match fs::read_to_string(version_path) {
+        Ok(version) => parse_version(&version),
+        Err(error) => match error.kind() {
+            ErrorKind::NotFound => Err(VersionFileError::MissingVersionFile),
+            _ => Err(error.into()),
+        },
+    }
+}
+
+pub fn parse_version(version: &str) -> Result<(String, String, String), VersionFileError> {
+    let version_components = version.split('.').collect::<Vec<_>>();
+    let (major, minor, patch) = match &version_components[..] {
+        [major, minor, patch] => (major.to_string(), minor.to_string(), patch.to_string()),
+        _ => return Err(VersionFileError::MalformedVersionFile),
+    };
+    Ok((major, minor, patch))
+}
+
 #[derive(thiserror::Error, Debug)]
 pub enum VersionFileError {
     #[error(
@@ -58,4 +72,7 @@ pub enum VersionFileError {
         env!("CARGO_PKG_VERSION").to_string()
     )]
     VersionMismatch { major: String, minor: String, patch: String },
+
+    #[error(transparent)]
+    IoError(#[from] std::io::Error),
 }

meilisearch/Cargo.toml

@@ -17,7 +17,7 @@ actix-cors = "0.7.0"
 actix-http = { version = "3.8.0", default-features = false, features = [
     "compress-brotli",
     "compress-gzip",
-    "rustls-0_21",
+    "rustls-0_23",
 ] }
 actix-utils = "3.0.1"
 actix-web = { version = "4.8.0", default-features = false, features = [
@@ -25,7 +25,7 @@ actix-web = { version = "4.8.0", default-features = false, features = [
     "compress-brotli",
     "compress-gzip",
     "cookies",
-    "rustls-0_21",
+    "rustls-0_23",
 ] }
 anyhow = { version = "1.0.86", features = ["backtrace"] }
 async-trait = "0.1.81"
@@ -72,8 +72,9 @@ reqwest = { version = "0.12.5", features = [
     "rustls-tls",
     "json",
 ], default-features = false }
-rustls = "0.21.12"
-rustls-pemfile = "1.0.4"
+rustls = { version = "0.23.11", features = ["ring"], default-features = false }
+rustls-pki-types = { version = "1.7.0", features = ["alloc"] }
+rustls-pemfile = "2.1.2"
 segment = { version = "0.2.4", optional = true }
 serde = { version = "1.0.204", features = ["derive"] }
 serde_json = { version = "1.0.120", features = ["preserve_order"] }
@@ -113,6 +114,7 @@ maplit = "1.0.2"
 meili-snap = { path = "../meili-snap" }
 temp-env = "0.3.6"
 urlencoding = "2.1.3"
+wiremock = "0.6.0"
 yaup = "0.3.1"
 
 [build-dependencies]
@@ -151,6 +153,7 @@ greek = ["meilisearch-types/greek"]
 khmer = ["meilisearch-types/khmer"]
 vietnamese = ["meilisearch-types/vietnamese"]
 swedish-recomposition = ["meilisearch-types/swedish-recomposition"]
+german = ["meilisearch-types/german"]
 
 [package.metadata.mini-dashboard]
 assets-url = "https://github.com/meilisearch/mini-dashboard/releases/download/v0.2.14/build.zip"

meilisearch/src/analytics/mock_analytics.rs

@@ -42,7 +42,7 @@ pub struct MultiSearchAggregator;
 
 #[allow(dead_code)]
 impl MultiSearchAggregator {
-    pub fn from_queries(_: &dyn Any, _: &dyn Any) -> Self {
+    pub fn from_federated_search(_: &dyn Any, _: &dyn Any) -> Self {
         Self
     }
 

meilisearch/src/analytics/mod.rs

@@ -102,7 +102,7 @@ pub trait Analytics: Sync + Send {
     /// This method should be called to aggregate post facet values searches
     fn post_facet_search(&self, aggregate: FacetSearchAggregator);
 
-    // this method should be called to aggregate a add documents request
+    // this method should be called to aggregate an add documents request
     fn add_documents(
         &self,
         documents_query: &UpdateDocumentsQuery,

meilisearch/src/analytics/segment_analytics.rs

@@ -1,4 +1,4 @@
-use std::collections::{BinaryHeap, HashMap, HashSet};
+use std::collections::{BTreeSet, BinaryHeap, HashMap, HashSet};
 use std::fs;
 use std::mem::take;
 use std::path::{Path, PathBuf};
@@ -10,6 +10,7 @@ use actix_web::HttpRequest;
 use byte_unit::Byte;
 use index_scheduler::IndexScheduler;
 use meilisearch_auth::{AuthController, AuthFilter};
+use meilisearch_types::locales::Locale;
 use meilisearch_types::InstanceUid;
 use once_cell::sync::Lazy;
 use regex::Regex;
@@ -34,8 +35,8 @@ use crate::routes::indexes::documents::{DocumentEditionByFunction, UpdateDocumen
 use crate::routes::indexes::facet_search::FacetSearchQuery;
 use crate::routes::{create_all_stats, Stats};
 use crate::search::{
-    FacetSearchResult, MatchingStrategy, SearchQuery, SearchQueryWithIndex, SearchResult,
-    SimilarQuery, SimilarResult, DEFAULT_CROP_LENGTH, DEFAULT_CROP_MARKER,
+    FacetSearchResult, FederatedSearch, MatchingStrategy, SearchQuery, SearchQueryWithIndex,
+    SearchResult, SimilarQuery, SimilarResult, DEFAULT_CROP_LENGTH, DEFAULT_CROP_MARKER,
     DEFAULT_HIGHLIGHT_POST_TAG, DEFAULT_HIGHLIGHT_PRE_TAG, DEFAULT_SEARCH_LIMIT,
     DEFAULT_SEMANTIC_RATIO,
 };
@@ -261,6 +262,7 @@ impl super::Analytics for SegmentAnalytics {
 #[derive(Debug, Clone, Serialize)]
 struct Infos {
     env: String,
+    experimental_contains_filter: bool,
     experimental_enable_metrics: bool,
     experimental_search_queue_size: usize,
     experimental_logs_mode: LogMode,
@@ -303,6 +305,7 @@ impl From<Opt> for Infos {
         // Thus we must not insert `..` at the end.
         let Opt {
             db_path,
+            experimental_contains_filter,
             experimental_enable_metrics,
             experimental_search_queue_size,
             experimental_logs_mode,
@@ -353,6 +356,7 @@ impl From<Opt> for Infos {
         // We consider information sensible if it contains a path, an address, or a key.
         Self {
             env,
+            experimental_contains_filter,
             experimental_enable_metrics,
             experimental_search_queue_size,
             experimental_logs_mode,
@@ -642,14 +646,15 @@ pub struct SearchAggregator {
     max_vector_size: usize,
     // Whether the semantic ratio passed to a hybrid search equals the default ratio.
     semantic_ratio: bool,
-    // Whether a non-default embedder was specified
-    embedder: bool,
     hybrid: bool,
     retrieve_vectors: bool,
 
     // every time a search is done, we increment the counter linked to the used settings
     matching_strategy: HashMap<String, usize>,
 
+    // List of the unique Locales passed as parameter
+    locales: BTreeSet<Locale>,
+
     // pagination
     max_limit: usize,
     max_offset: usize,
@@ -704,6 +709,7 @@ impl SearchAggregator {
             attributes_to_search_on,
             hybrid,
             ranking_score_threshold,
+            locales,
         } = query;
 
         let mut ret = Self::default();
@@ -771,6 +777,10 @@ impl SearchAggregator {
 
         ret.matching_strategy.insert(format!("{:?}", matching_strategy), 1);
 
+        if let Some(locales) = locales {
+            ret.locales = locales.iter().copied().collect();
+        }
+
         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();
@@ -783,7 +793,6 @@ impl SearchAggregator {
 
         if let Some(hybrid) = hybrid {
             ret.semantic_ratio = hybrid.semantic_ratio != DEFAULT_SEMANTIC_RATIO();
-            ret.embedder = hybrid.embedder.is_some();
             ret.hybrid = true;
         }
 
@@ -851,11 +860,11 @@ impl SearchAggregator {
             show_ranking_score,
             show_ranking_score_details,
             semantic_ratio,
-            embedder,
             hybrid,
             total_degraded,
             total_used_negative_operator,
             ranking_score_threshold,
+            ref mut locales,
         } = other;
 
         if self.timestamp.is_none() {
@@ -910,7 +919,6 @@ impl SearchAggregator {
         self.retrieve_vectors |= retrieve_vectors;
         self.semantic_ratio |= semantic_ratio;
         self.hybrid |= hybrid;
-        self.embedder |= embedder;
 
         // pagination
         self.max_limit = self.max_limit.max(max_limit);
@@ -944,6 +952,9 @@ impl SearchAggregator {
         self.show_ranking_score |= show_ranking_score;
         self.show_ranking_score_details |= show_ranking_score_details;
         self.ranking_score_threshold |= ranking_score_threshold;
+
+        // locales
+        self.locales.append(locales);
     }
 
     pub fn into_event(self, user: &User, event_name: &str) -> Option<Track> {
@@ -983,11 +994,11 @@ impl SearchAggregator {
             show_ranking_score,
             show_ranking_score_details,
             semantic_ratio,
-            embedder,
             hybrid,
             total_degraded,
             total_used_negative_operator,
             ranking_score_threshold,
+            locales,
         } = self;
 
         if total_received == 0 {
@@ -1034,7 +1045,6 @@ impl SearchAggregator {
                 "hybrid": {
                     "enabled": hybrid,
                     "semantic_ratio": semantic_ratio,
-                    "embedder": embedder,
                 },
                 "pagination": {
                    "max_limit": max_limit,
@@ -1057,6 +1067,7 @@ impl SearchAggregator {
                 "matching_strategy": {
                     "most_used_strategy": matching_strategy.iter().max_by_key(|(_, v)| *v).map(|(k, _)| json!(k)).unwrap_or_else(|| json!(null)),
                 },
+                "locales": locales,
                 "scoring": {
                     "show_ranking_score": show_ranking_score,
                     "show_ranking_score_details": show_ranking_score_details,
@@ -1095,22 +1106,33 @@ pub struct MultiSearchAggregator {
     show_ranking_score: bool,
     show_ranking_score_details: bool,
 
+    // federation
+    use_federation: bool,
+
     // context
     user_agents: HashSet<String>,
 }
 
 impl MultiSearchAggregator {
-    pub fn from_queries(query: &[SearchQueryWithIndex], request: &HttpRequest) -> Self {
+    pub fn from_federated_search(
+        federated_search: &FederatedSearch,
+        request: &HttpRequest,
+    ) -> Self {
         let timestamp = Some(OffsetDateTime::now_utc());
 
         let user_agents = extract_user_agents(request).into_iter().collect();
 
-        let distinct_indexes: HashSet<_> = query
+        let use_federation = federated_search.federation.is_some();
+
+        let distinct_indexes: HashSet<_> = federated_search
+            .queries
             .iter()
             .map(|query| {
+                let query = &query;
                 // make sure we get a compilation error if a field gets added to / removed from SearchQueryWithIndex
                 let SearchQueryWithIndex {
                     index_uid,
+                    federation_options: _,
                     q: _,
                     vector: _,
                     offset: _,
@@ -1136,14 +1158,17 @@ impl MultiSearchAggregator {
                     attributes_to_search_on: _,
                     hybrid: _,
                     ranking_score_threshold: _,
+                    locales: _,
                 } = query;
 
                 index_uid.as_str()
             })
             .collect();
 
-        let show_ranking_score = query.iter().any(|query| query.show_ranking_score);
-        let show_ranking_score_details = query.iter().any(|query| query.show_ranking_score_details);
+        let show_ranking_score =
+            federated_search.queries.iter().any(|query| query.show_ranking_score);
+        let show_ranking_score_details =
+            federated_search.queries.iter().any(|query| query.show_ranking_score_details);
 
         Self {
             timestamp,
@@ -1151,10 +1176,11 @@ impl MultiSearchAggregator {
             total_succeeded: 0,
             total_distinct_index_count: distinct_indexes.len(),
             total_single_index: if distinct_indexes.len() == 1 { 1 } else { 0 },
-            total_search_count: query.len(),
+            total_search_count: federated_search.queries.len(),
             show_ranking_score,
             show_ranking_score_details,
             user_agents,
+            use_federation,
         }
     }
 
@@ -1180,6 +1206,7 @@ impl MultiSearchAggregator {
         let show_ranking_score_details =
             this.show_ranking_score_details || other.show_ranking_score_details;
         let mut user_agents = this.user_agents;
+        let use_federation = this.use_federation || other.use_federation;
 
         for user_agent in other.user_agents.into_iter() {
             user_agents.insert(user_agent);
@@ -1196,6 +1223,7 @@ impl MultiSearchAggregator {
             user_agents,
             show_ranking_score,
             show_ranking_score_details,
+            use_federation,
             // do not add _ or ..Default::default() here
         };
 
@@ -1214,6 +1242,7 @@ impl MultiSearchAggregator {
             user_agents,
             show_ranking_score,
             show_ranking_score_details,
+            use_federation,
         } = self;
 
         if total_received == 0 {
@@ -1238,6 +1267,9 @@ impl MultiSearchAggregator {
                 "scoring": {
                     "show_ranking_score": show_ranking_score,
                     "show_ranking_score_details": show_ranking_score_details,
+                },
+                "federation": {
+                    "use_federation": use_federation,
                 }
             });
 
@@ -1284,6 +1316,7 @@ impl FacetSearchAggregator {
             attributes_to_search_on,
             hybrid,
             ranking_score_threshold,
+            locales,
         } = query;
 
         let mut ret = Self::default();
@@ -1299,7 +1332,8 @@ impl FacetSearchAggregator {
             || *matching_strategy != MatchingStrategy::default()
             || attributes_to_search_on.is_some()
             || hybrid.is_some()
-            || ranking_score_threshold.is_some();
+            || ranking_score_threshold.is_some()
+            || locales.is_some();
 
         ret
     }
@@ -1741,7 +1775,6 @@ pub struct SimilarAggregator {
     used_syntax: HashMap<String, usize>,
 
     // Whether a non-default embedder was specified
-    embedder: bool,
     retrieve_vectors: bool,
 
     // pagination
@@ -1762,7 +1795,7 @@ impl SimilarAggregator {
     pub fn from_query(query: &SimilarQuery, request: &HttpRequest) -> Self {
         let SimilarQuery {
             id: _,
-            embedder,
+            embedder: _,
             offset,
             limit,
             attributes_to_retrieve: _,
@@ -1810,7 +1843,6 @@ impl SimilarAggregator {
         ret.show_ranking_score_details = *show_ranking_score_details;
         ret.ranking_score_threshold = ranking_score_threshold.is_some();
 
-        ret.embedder = embedder.is_some();
         ret.retrieve_vectors = *retrieve_vectors;
 
         ret
@@ -1842,7 +1874,6 @@ impl SimilarAggregator {
             max_attributes_to_retrieve,
             show_ranking_score,
             show_ranking_score_details,
-            embedder,
             ranking_score_threshold,
             retrieve_vectors,
         } = other;
@@ -1873,7 +1904,6 @@ impl SimilarAggregator {
             *used_syntax = used_syntax.saturating_add(value);
         }
 
-        self.embedder |= embedder;
         self.retrieve_vectors |= retrieve_vectors;
 
         // pagination
@@ -1907,7 +1937,6 @@ impl SimilarAggregator {
             max_attributes_to_retrieve,
             show_ranking_score,
             show_ranking_score_details,
-            embedder,
             ranking_score_threshold,
             retrieve_vectors,
         } = self;
@@ -1939,9 +1968,6 @@ impl SimilarAggregator {
                 "vector": {
                     "retrieve_vectors": retrieve_vectors,
                 },
-                "hybrid": {
-                    "embedder": embedder,
-                },
                 "pagination": {
                    "max_limit": max_limit,
                    "max_offset": max_offset,

meilisearch/src/error.rs

@@ -4,6 +4,7 @@ use byte_unit::{Byte, UnitType};
 use meilisearch_types::document_formats::{DocumentFormatError, PayloadType};
 use meilisearch_types::error::{Code, ErrorCode, ResponseError};
 use meilisearch_types::index_uid::{IndexUid, IndexUidFormatError};
+use meilisearch_types::milli::OrderBy;
 use serde_json::Value;
 use tokio::task::JoinError;
 
@@ -25,12 +26,22 @@ pub enum MeilisearchHttpError {
     DocumentNotFound(String),
     #[error("Sending an empty filter is forbidden.")]
     EmptyFilter,
-    #[error("Using `federationOptions` is not allowed in a non-federated search.\n Hint: remove `federationOptions` from query #{0} or add `federation: {{}}` to the request.")]
-    FederationOptionsInNonFederatedRequest(usize),
-    #[error("Inside `.queries[{0}]`: Using pagination options is not allowed in federated queries.\n Hint: remove `{1}` from query #{0} or remove `federation: {{}}` from the request")]
-    PaginationInFederatedQuery(usize, &'static str),
     #[error("Invalid syntax for the filter parameter: `expected {}, found: {1}`.", .0.join(", "))]
     InvalidExpression(&'static [&'static str], Value),
+    #[error("Using `federationOptions` is not allowed in a non-federated search.\n - Hint: remove `federationOptions` from query #{0} or add `federation` to the request.")]
+    FederationOptionsInNonFederatedRequest(usize),
+    #[error("Inside `.queries[{0}]`: Using pagination options is not allowed in federated queries.\n - Hint: remove `{1}` from query #{0} or remove `federation` from the request\n - Hint: pass `federation.limit` and `federation.offset` for pagination in federated search")]
+    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("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,
+        previous_facet_order: OrderBy,
+        previous_uid: String,
+        index_facet_order: OrderBy,
+        current_uid: String,
+    },
     #[error("A {0} payload is missing.")]
     MissingPayload(PayloadType),
     #[error("Too many search requests running at the same time: {0}. Retry after 10s.")]
@@ -61,7 +72,7 @@ pub enum MeilisearchHttpError {
     DocumentFormat(#[from] DocumentFormatError),
     #[error(transparent)]
     Join(#[from] JoinError),
-    #[error("Invalid request: missing `hybrid` parameter when both `q` and `vector` are present.")]
+    #[error("Invalid request: missing `hybrid` parameter when `vector` is present.")]
     MissingSearchHybrid,
 }
 
@@ -96,6 +107,10 @@ impl ErrorCode for MeilisearchHttpError {
             MeilisearchHttpError::PaginationInFederatedQuery(_, _) => {
                 Code::InvalidMultiSearchQueryPagination
             }
+            MeilisearchHttpError::FacetsInFederatedQuery(..) => Code::InvalidMultiSearchQueryFacets,
+            MeilisearchHttpError::InconsistentFacetOrder { .. } => {
+                Code::InvalidMultiSearchFacetOrder
+            }
         }
     }
 }

meilisearch/src/lib.rs

@@ -13,11 +13,10 @@ pub mod search_queue;
 
 use std::fs::File;
 use std::io::{BufReader, BufWriter};
-use std::num::NonZeroUsize;
 use std::path::Path;
 use std::str::FromStr;
 use std::sync::Arc;
-use std::thread::{self, available_parallelism};
+use std::thread;
 use std::time::Duration;
 
 use actix_cors::Cors;
@@ -37,7 +36,7 @@ use meilisearch_types::milli::documents::{DocumentsBatchBuilder, DocumentsBatchR
 use meilisearch_types::milli::update::{IndexDocumentsConfig, IndexDocumentsMethod};
 use meilisearch_types::settings::apply_settings_to_builder;
 use meilisearch_types::tasks::KindWithContent;
-use meilisearch_types::versioning::{check_version_file, create_version_file};
+use meilisearch_types::versioning::{check_version_file, create_current_version_file};
 use meilisearch_types::{compression, milli, VERSION_FILE_NAME};
 pub use option::Opt;
 use option::ScheduleSnapshot;
@@ -118,6 +117,7 @@ pub type LogStderrType = tracing_subscriber::filter::Filtered<
 pub fn create_app(
     index_scheduler: Data<IndexScheduler>,
     auth_controller: Data<AuthController>,
+    search_queue: Data<SearchQueue>,
     opt: Opt,
     logs: (LogRouteHandle, LogStderrHandle),
     analytics: Arc<dyn Analytics>,
@@ -137,6 +137,7 @@ pub fn create_app(
                 s,
                 index_scheduler.clone(),
                 auth_controller.clone(),
+                search_queue.clone(),
                 &opt,
                 logs,
                 analytics.clone(),
@@ -318,7 +319,7 @@ fn open_or_create_database_unchecked(
     match (
         index_scheduler_builder(),
         auth_controller.map_err(anyhow::Error::from),
-        create_version_file(&opt.db_path).map_err(anyhow::Error::from),
+        create_current_version_file(&opt.db_path).map_err(anyhow::Error::from),
     ) {
         (Ok(i), Ok(a), Ok(())) => Ok((i, a)),
         (Err(e), _, _) | (_, Err(e), _) | (_, _, Err(e)) => {
@@ -469,19 +470,16 @@ 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: Arc<dyn Analytics>,
 ) {
-    let search_queue = SearchQueue::new(
-        opt.experimental_search_queue_size,
-        available_parallelism().unwrap_or(NonZeroUsize::new(2).unwrap()),
-    );
     let http_payload_size_limit = opt.http_payload_size_limit.as_u64() as usize;
     config
         .app_data(index_scheduler)
         .app_data(auth)
-        .app_data(web::Data::new(search_queue))
+        .app_data(search_queue)
         .app_data(web::Data::from(analytics))
         .app_data(web::Data::new(logs_route))
         .app_data(web::Data::new(logs_stderr))

meilisearch/src/main.rs

@@ -1,8 +1,10 @@
 use std::env;
 use std::io::{stderr, LineWriter, Write};
+use std::num::NonZeroUsize;
 use std::path::PathBuf;
 use std::str::FromStr;
 use std::sync::Arc;
+use std::thread::available_parallelism;
 
 use actix_web::http::KeepAlive;
 use actix_web::web::Data;
@@ -11,6 +13,7 @@ use index_scheduler::IndexScheduler;
 use is_terminal::IsTerminal;
 use meilisearch::analytics::Analytics;
 use meilisearch::option::LogMode;
+use meilisearch::search_queue::SearchQueue;
 use meilisearch::{
     analytics, create_app, setup_meilisearch, LogRouteHandle, LogRouteType, LogStderrHandle,
     LogStderrType, Opt, SubscriberForSecondLayer,
@@ -72,6 +75,19 @@ fn on_panic(info: &std::panic::PanicInfo) {
 
 #[actix_web::main]
 async fn main() -> anyhow::Result<()> {
+    try_main().await.inspect_err(|error| {
+        tracing::error!(%error);
+        let mut current = error.source();
+        let mut depth = 0;
+        while let Some(source) = current {
+            tracing::info!(%source, depth, "Error caused by");
+            current = source.source();
+            depth += 1;
+        }
+    })
+}
+
+async fn try_main() -> anyhow::Result<()> {
     let (opt, config_read_from) = Opt::try_build()?;
 
     std::panic::set_hook(Box::new(on_panic));
@@ -135,11 +151,17 @@ async fn run_http(
     let opt_clone = opt.clone();
     let index_scheduler = Data::from(index_scheduler);
     let auth_controller = Data::from(auth_controller);
+    let search_queue = SearchQueue::new(
+        opt.experimental_search_queue_size,
+        available_parallelism().unwrap_or(NonZeroUsize::new(2).unwrap()),
+    );
+    let search_queue = Data::new(search_queue);
 
     let http_server = HttpServer::new(move || {
         create_app(
             index_scheduler.clone(),
             auth_controller.clone(),
+            search_queue.clone(),
             opt.clone(),
             logs.clone(),
             analytics.clone(),
@@ -151,7 +173,7 @@ async fn run_http(
     .keep_alive(KeepAlive::Os);
 
     if let Some(config) = opt_clone.get_ssl_config()? {
-        http_server.bind_rustls_021(opt_clone.http_addr, config)?.run().await?;
+        http_server.bind_rustls_0_23(opt_clone.http_addr, config)?.run().await?;
     } else {
         http_server.bind(&opt_clone.http_addr)?.run().await?;
     }

meilisearch/src/middleware.rs

@@ -55,16 +55,17 @@ where
         let index_scheduler = req.app_data::<Data<IndexScheduler>>().unwrap();
         let features = index_scheduler.features();
 
+        let request_path = req.path();
+        let request_pattern = req.match_pattern();
+        let metric_path = request_pattern.as_ref().map_or(request_path, String::as_str).to_string();
+        let request_method = req.method().to_string();
+
         if features.check_metrics().is_ok() {
-            let request_path = req.path();
             let is_registered_resource = req.resource_map().has_resource(request_path);
             if is_registered_resource {
-                let request_pattern = req.match_pattern();
-                let metric_path = request_pattern.as_ref().map_or(request_path, String::as_str);
-                let request_method = req.method().to_string();
                 histogram_timer = Some(
                     crate::metrics::MEILISEARCH_HTTP_RESPONSE_TIME_SECONDS
-                        .with_label_values(&[&request_method, metric_path])
+                        .with_label_values(&[&request_method, &metric_path])
                         .start_timer(),
                 );
             }
@@ -76,11 +77,7 @@ where
             let res = fut.await?;
 
             crate::metrics::MEILISEARCH_HTTP_REQUESTS_TOTAL
-                .with_label_values(&[
-                    res.request().method().as_str(),
-                    res.request().path(),
-                    res.status().as_str(),
-                ])
+                .with_label_values(&[&request_method, &metric_path, res.status().as_str()])
                 .inc();
 
             if let Some(histogram_timer) = histogram_timer {

meilisearch/src/option.rs

@@ -14,11 +14,9 @@ use clap::Parser;
 use meilisearch_types::features::InstanceTogglableFeatures;
 use meilisearch_types::milli::update::IndexerConfig;
 use meilisearch_types::milli::ThreadPoolNoAbortBuilder;
-use rustls::server::{
-    AllowAnyAnonymousOrAuthenticatedClient, AllowAnyAuthenticatedClient, ServerSessionMemoryCache,
-};
+use rustls::server::{ServerSessionMemoryCache, WebPkiClientVerifier};
 use rustls::RootCertStore;
-use rustls_pemfile::{certs, pkcs8_private_keys, rsa_private_keys};
+use rustls_pemfile::{certs, rsa_private_keys};
 use serde::{Deserialize, Serialize};
 use sysinfo::{MemoryRefreshKind, RefreshKind, System};
 use url::Url;
@@ -54,6 +52,7 @@ const MEILI_LOG_LEVEL: &str = "MEILI_LOG_LEVEL";
 const MEILI_EXPERIMENTAL_LOGS_MODE: &str = "MEILI_EXPERIMENTAL_LOGS_MODE";
 const MEILI_EXPERIMENTAL_REPLICATION_PARAMETERS: &str = "MEILI_EXPERIMENTAL_REPLICATION_PARAMETERS";
 const MEILI_EXPERIMENTAL_ENABLE_LOGS_ROUTE: &str = "MEILI_EXPERIMENTAL_ENABLE_LOGS_ROUTE";
+const MEILI_EXPERIMENTAL_CONTAINS_FILTER: &str = "MEILI_EXPERIMENTAL_CONTAINS_FILTER";
 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_REDUCE_INDEXING_MEMORY_USAGE: &str =
@@ -339,6 +338,13 @@ pub struct Opt {
     #[serde(default)]
     pub log_level: LogLevel,
 
+    /// Experimental contains filter feature. For more information, see: <https://github.com/orgs/meilisearch/discussions/763>
+    ///
+    /// Enables the experimental contains filter operator.
+    #[clap(long, env = MEILI_EXPERIMENTAL_CONTAINS_FILTER)]
+    #[serde(default)]
+    pub experimental_contains_filter: bool,
+
     /// Experimental metrics feature. For more information, see: <https://github.com/meilisearch/meilisearch/discussions/3518>
     ///
     /// Enables the Prometheus metrics on the `GET /metrics` endpoint.
@@ -483,6 +489,7 @@ impl Opt {
             config_file_path: _,
             #[cfg(feature = "analytics")]
             no_analytics,
+            experimental_contains_filter,
             experimental_enable_metrics,
             experimental_search_queue_size,
             experimental_logs_mode,
@@ -540,6 +547,10 @@ impl Opt {
 
         export_to_env_if_not_present(MEILI_DUMP_DIR, dump_dir);
         export_to_env_if_not_present(MEILI_LOG_LEVEL, log_level.to_string());
+        export_to_env_if_not_present(
+            MEILI_EXPERIMENTAL_CONTAINS_FILTER,
+            experimental_contains_filter.to_string(),
+        );
         export_to_env_if_not_present(
             MEILI_EXPERIMENTAL_ENABLE_METRICS,
             experimental_enable_metrics.to_string(),
@@ -569,23 +580,21 @@ impl Opt {
 
     pub fn get_ssl_config(&self) -> anyhow::Result<Option<rustls::ServerConfig>> {
         if let (Some(cert_path), Some(key_path)) = (&self.ssl_cert_path, &self.ssl_key_path) {
-            let config = rustls::ServerConfig::builder().with_safe_defaults();
+            let config = rustls::ServerConfig::builder();
 
             let config = match &self.ssl_auth_path {
                 Some(auth_path) => {
                     let roots = load_certs(auth_path.to_path_buf())?;
                     let mut client_auth_roots = RootCertStore::empty();
                     for root in roots {
-                        client_auth_roots.add(&root).unwrap();
+                        client_auth_roots.add(root).unwrap();
                     }
-                    if self.ssl_require_auth {
-                        let verifier = AllowAnyAuthenticatedClient::new(client_auth_roots);
-                        config.with_client_cert_verifier(Arc::from(verifier))
-                    } else {
-                        let verifier =
-                            AllowAnyAnonymousOrAuthenticatedClient::new(client_auth_roots);
-                        config.with_client_cert_verifier(Arc::from(verifier))
+                    let mut client_verifier =
+                        WebPkiClientVerifier::builder(client_auth_roots.into());
+                    if !self.ssl_require_auth {
+                        client_verifier = client_verifier.allow_unauthenticated();
                     }
+                    config.with_client_cert_verifier(client_verifier.build()?)
                 }
                 None => config.with_no_client_auth(),
             };
@@ -594,7 +603,7 @@ impl Opt {
             let privkey = load_private_key(key_path.to_path_buf())?;
             let ocsp = load_ocsp(&self.ssl_ocsp_path)?;
             let mut config = config
-                .with_single_cert_with_ocsp_and_sct(certs, privkey, ocsp, vec![])
+                .with_single_cert_with_ocsp(certs, privkey, ocsp)
                 .map_err(|_| anyhow::anyhow!("bad certificates/private key"))?;
 
             config.key_log = Arc::new(rustls::KeyLogFile::new());
@@ -604,7 +613,7 @@ impl Opt {
             }
 
             if self.ssl_tickets {
-                config.ticketer = rustls::Ticketer::new().unwrap();
+                config.ticketer = rustls::crypto::ring::Ticketer::new().unwrap();
             }
 
             Ok(Some(config))
@@ -617,6 +626,7 @@ impl Opt {
         InstanceTogglableFeatures {
             metrics: self.experimental_enable_metrics,
             logs_route: self.experimental_enable_logs_route,
+            contains_filter: self.experimental_contains_filter,
         }
     }
 }
@@ -769,21 +779,26 @@ impl Deref for MaxThreads {
     }
 }
 
-fn load_certs(filename: PathBuf) -> anyhow::Result<Vec<rustls::Certificate>> {
+fn load_certs(
+    filename: PathBuf,
+) -> anyhow::Result<Vec<rustls::pki_types::CertificateDer<'static>>> {
     let certfile =
         fs::File::open(filename).map_err(|_| anyhow::anyhow!("cannot open certificate file"))?;
     let mut reader = BufReader::new(certfile);
     certs(&mut reader)
-        .map(|certs| certs.into_iter().map(rustls::Certificate).collect())
+        .collect::<Result<Vec<_>, _>>()
         .map_err(|_| anyhow::anyhow!("cannot read certificate file"))
 }
 
-fn load_private_key(filename: PathBuf) -> anyhow::Result<rustls::PrivateKey> {
+fn load_private_key(
+    filename: PathBuf,
+) -> anyhow::Result<rustls::pki_types::PrivateKeyDer<'static>> {
     let rsa_keys = {
         let keyfile = fs::File::open(filename.clone())
             .map_err(|_| anyhow::anyhow!("cannot open private key file"))?;
         let mut reader = BufReader::new(keyfile);
         rsa_private_keys(&mut reader)
+            .collect::<Result<Vec<_>, _>>()
             .map_err(|_| anyhow::anyhow!("file contains invalid rsa private key"))?
     };
 
@@ -791,19 +806,21 @@ fn load_private_key(filename: PathBuf) -> anyhow::Result<rustls::PrivateKey> {
         let keyfile = fs::File::open(filename)
             .map_err(|_| anyhow::anyhow!("cannot open private key file"))?;
         let mut reader = BufReader::new(keyfile);
-        pkcs8_private_keys(&mut reader).map_err(|_| {
-            anyhow::anyhow!(
-                "file contains invalid pkcs8 private key (encrypted keys not supported)"
-            )
-        })?
+        rustls_pemfile::pkcs8_private_keys(&mut reader).collect::<Result<Vec<_>, _>>().map_err(
+            |_| {
+                anyhow::anyhow!(
+                    "file contains invalid pkcs8 private key (encrypted keys not supported)"
+                )
+            },
+        )?
     };
 
     // prefer to load pkcs8 keys
     if !pkcs8_keys.is_empty() {
-        Ok(rustls::PrivateKey(pkcs8_keys[0].clone()))
+        Ok(rustls::pki_types::PrivateKeyDer::Pkcs8(pkcs8_keys[0].clone_key()))
     } else {
         assert!(!rsa_keys.is_empty());
-        Ok(rustls::PrivateKey(rsa_keys[0].clone()))
+        Ok(rustls::pki_types::PrivateKeyDer::Pkcs1(rsa_keys[0].clone_key()))
     }
 }
 

meilisearch/src/routes/features.rs

@@ -49,6 +49,8 @@ pub struct RuntimeTogglableFeatures {
     pub logs_route: Option<bool>,
     #[deserr(default)]
     pub edit_documents_by_function: Option<bool>,
+    #[deserr(default)]
+    pub contains_filter: Option<bool>,
 }
 
 async fn patch_features(
@@ -72,6 +74,7 @@ async fn patch_features(
             .0
             .edit_documents_by_function
             .unwrap_or(old_features.edit_documents_by_function),
+        contains_filter: new_features.0.contains_filter.unwrap_or(old_features.contains_filter),
     };
 
     // explicitly destructure for analytics rather than using the `Serialize` implementation, because
@@ -82,6 +85,7 @@ async fn patch_features(
         metrics,
         logs_route,
         edit_documents_by_function,
+        contains_filter,
     } = new_features;
 
     analytics.publish(
@@ -91,6 +95,7 @@ async fn patch_features(
             "metrics": metrics,
             "logs_route": logs_route,
             "edit_documents_by_function": edit_documents_by_function,
+            "contains_filter": contains_filter,
         }),
         Some(&req),
     );

meilisearch/src/routes/indexes/documents.rs

@@ -7,7 +7,7 @@ use bstr::ByteSlice as _;
 use deserr::actix_web::{AwebJson, AwebQueryParameter};
 use deserr::Deserr;
 use futures::StreamExt;
-use index_scheduler::{IndexScheduler, TaskId};
+use index_scheduler::{IndexScheduler, RoFeatures, TaskId};
 use meilisearch_types::deserr::query_params::Param;
 use meilisearch_types::deserr::{DeserrJsonError, DeserrQueryParamError};
 use meilisearch_types::document_formats::{read_csv, read_json, read_ndjson, PayloadType};
@@ -260,8 +260,15 @@ fn documents_by_query(
     let retrieve_vectors = RetrieveVectors::new(retrieve_vectors, features)?;
 
     let index = index_scheduler.index(&index_uid)?;
-    let (total, documents) =
-        retrieve_documents(&index, offset, limit, filter, fields, retrieve_vectors)?;
+    let (total, documents) = retrieve_documents(
+        &index,
+        offset,
+        limit,
+        filter,
+        fields,
+        retrieve_vectors,
+        index_scheduler.features(),
+    )?;
 
     let ret = PaginationView::new(offset, limit, total as usize, documents);
 
@@ -305,7 +312,11 @@ pub async fn replace_documents(
     debug!(parameters = ?params, "Replace documents");
     let params = params.into_inner();
 
-    analytics.add_documents(&params, index_scheduler.index(&index_uid).is_err(), &req);
+    analytics.add_documents(
+        &params,
+        index_scheduler.index_exists(&index_uid).map_or(true, |x| !x),
+        &req,
+    );
 
     let allow_index_creation = index_scheduler.filters().allow_index_creation(&index_uid);
     let uid = get_task_id(&req, &opt)?;
@@ -342,7 +353,11 @@ pub async fn update_documents(
     let params = params.into_inner();
     debug!(parameters = ?params, "Update documents");
 
-    analytics.update_documents(&params, index_scheduler.index(&index_uid).is_err(), &req);
+    analytics.add_documents(
+        &params,
+        index_scheduler.index_exists(&index_uid).map_or(true, |x| !x),
+        &req,
+    );
 
     let allow_index_creation = index_scheduler.filters().allow_index_creation(&index_uid);
     let uid = get_task_id(&req, &opt)?;
@@ -557,11 +572,9 @@ pub async fn delete_documents_by_filter(
     analytics.delete_documents(DocumentDeletionKind::PerFilter, &req);
 
     // we ensure the filter is well formed before enqueuing it
-    || -> Result<_, ResponseError> {
-        Ok(crate::search::parse_filter(&filter)?.ok_or(MeilisearchHttpError::EmptyFilter)?)
-    }()
-    // and whatever was the error, the error code should always be an InvalidDocumentFilter
-    .map_err(|err| ResponseError::from_msg(err.message, Code::InvalidDocumentFilter))?;
+    crate::search::parse_filter(&filter, Code::InvalidDocumentFilter, index_scheduler.features())?
+        .ok_or(MeilisearchHttpError::EmptyFilter)?;
+
     let task = KindWithContent::DocumentDeletionByFilter { index_uid, filter_expr: filter };
 
     let uid = get_task_id(&req, &opt)?;
@@ -618,11 +631,12 @@ pub async fn edit_documents_by_function(
 
     if let Some(ref filter) = filter {
         // we ensure the filter is well formed before enqueuing it
-        || -> Result<_, ResponseError> {
-            Ok(crate::search::parse_filter(filter)?.ok_or(MeilisearchHttpError::EmptyFilter)?)
-        }()
-        // and whatever was the error, the error code should always be an InvalidDocumentFilter
-        .map_err(|err| ResponseError::from_msg(err.message, Code::InvalidDocumentFilter))?;
+        crate::search::parse_filter(
+            filter,
+            Code::InvalidDocumentFilter,
+            index_scheduler.features(),
+        )?
+        .ok_or(MeilisearchHttpError::EmptyFilter)?;
     }
     let task = KindWithContent::DocumentEdition {
         index_uid,
@@ -728,12 +742,12 @@ fn retrieve_documents<S: AsRef<str>>(
     filter: Option<Value>,
     attributes_to_retrieve: Option<Vec<S>>,
     retrieve_vectors: RetrieveVectors,
+    features: RoFeatures,
 ) -> Result<(u64, Vec<Document>), ResponseError> {
     let rtxn = index.read_txn()?;
     let filter = &filter;
     let filter = if let Some(filter) = filter {
-        parse_filter(filter)
-            .map_err(|err| ResponseError::from_msg(err.to_string(), Code::InvalidDocumentFilter))?
+        parse_filter(filter, Code::InvalidDocumentFilter, features)?
     } else {
         None
     };

meilisearch/src/routes/indexes/facet_search.rs

@@ -6,6 +6,7 @@ use meilisearch_types::deserr::DeserrJsonError;
 use meilisearch_types::error::deserr_codes::*;
 use meilisearch_types::error::ResponseError;
 use meilisearch_types::index_uid::IndexUid;
+use meilisearch_types::locales::Locale;
 use serde_json::Value;
 use tracing::debug;
 
@@ -48,6 +49,8 @@ pub struct FacetSearchQuery {
     pub attributes_to_search_on: Option<Vec<String>>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchRankingScoreThreshold>, default)]
     pub ranking_score_threshold: Option<RankingScoreThreshold>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchLocales>, default)]
+    pub locales: Option<Vec<Locale>>,
 }
 
 pub async fn search(
@@ -67,6 +70,7 @@ pub async fn search(
 
     let facet_query = query.facet_query.clone();
     let facet_name = query.facet_name.clone();
+    let locales = query.locales.clone().map(|l| l.into_iter().map(Into::into).collect());
     let mut search_query = SearchQuery::from(query);
 
     // Tenant token search_rules.
@@ -77,11 +81,21 @@ pub async fn search(
     let index = index_scheduler.index(&index_uid)?;
     let features = index_scheduler.features();
     let search_kind = search_kind(&search_query, &index_scheduler, &index, features)?;
-    let _permit = search_queue.try_get_search_permit().await?;
+    let permit = search_queue.try_get_search_permit().await?;
     let search_result = tokio::task::spawn_blocking(move || {
-        perform_facet_search(&index, search_query, facet_query, facet_name, search_kind)
+        perform_facet_search(
+            &index,
+            search_query,
+            facet_query,
+            facet_name,
+            search_kind,
+            index_scheduler.features(),
+            locales,
+        )
     })
-    .await?;
+    .await;
+    permit.drop().await;
+    let search_result = search_result?;
 
     if let Ok(ref search_result) = search_result {
         aggregate.succeed(search_result);
@@ -106,6 +120,7 @@ impl From<FacetSearchQuery> for SearchQuery {
             attributes_to_search_on,
             hybrid,
             ranking_score_threshold,
+            locales,
         } = value;
 
         SearchQuery {
@@ -134,6 +149,7 @@ impl From<FacetSearchQuery> for SearchQuery {
             attributes_to_search_on,
             hybrid,
             ranking_score_threshold,
+            locales,
         }
     }
 }

meilisearch/src/routes/indexes/search.rs

@@ -7,6 +7,7 @@ use meilisearch_types::deserr::{DeserrJsonError, DeserrQueryParamError};
 use meilisearch_types::error::deserr_codes::*;
 use meilisearch_types::error::ResponseError;
 use meilisearch_types::index_uid::IndexUid;
+use meilisearch_types::locales::Locale;
 use meilisearch_types::milli;
 use meilisearch_types::serde_cs::vec::CS;
 use serde_json::Value;
@@ -89,6 +90,8 @@ pub struct SearchQueryGet {
     pub hybrid_semantic_ratio: Option<SemanticRatioGet>,
     #[deserr(default, error = DeserrQueryParamError<InvalidSearchRankingScoreThreshold>)]
     pub ranking_score_threshold: Option<RankingScoreThresholdGet>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidSearchLocales>)]
+    pub locales: Option<CS<Locale>>,
 }
 
 #[derive(Debug, Clone, Copy, PartialEq, deserr::Deserr)]
@@ -125,8 +128,10 @@ impl std::ops::Deref for SemanticRatioGet {
     }
 }
 
-impl From<SearchQueryGet> for SearchQuery {
-    fn from(other: SearchQueryGet) -> Self {
+impl TryFrom<SearchQueryGet> for SearchQuery {
+    type Error = ResponseError;
+
+    fn try_from(other: SearchQueryGet) -> Result<Self, Self::Error> {
         let filter = match other.filter {
             Some(f) => match serde_json::from_str(&f) {
                 Ok(v) => Some(v),
@@ -137,19 +142,28 @@ impl From<SearchQueryGet> for SearchQuery {
 
         let hybrid = match (other.hybrid_embedder, other.hybrid_semantic_ratio) {
             (None, None) => None,
-            (None, Some(semantic_ratio)) => {
-                Some(HybridQuery { semantic_ratio: *semantic_ratio, embedder: None })
+            (None, Some(_)) => {
+                return Err(ResponseError::from_msg(
+                    "`hybridEmbedder` is mandatory when `hybridSemanticRatio` is present".into(),
+                    meilisearch_types::error::Code::InvalidHybridQuery,
+                ));
+            }
+            (Some(embedder), None) => {
+                Some(HybridQuery { semantic_ratio: DEFAULT_SEMANTIC_RATIO(), embedder })
             }
-            (Some(embedder), None) => Some(HybridQuery {
-                semantic_ratio: DEFAULT_SEMANTIC_RATIO(),
-                embedder: Some(embedder),
-            }),
             (Some(embedder), Some(semantic_ratio)) => {
-                Some(HybridQuery { semantic_ratio: *semantic_ratio, embedder: Some(embedder) })
+                Some(HybridQuery { semantic_ratio: *semantic_ratio, embedder })
             }
         };
 
-        Self {
+        if other.vector.is_some() && hybrid.is_none() {
+            return Err(ResponseError::from_msg(
+                "`hybridEmbedder` is mandatory when `vector` is present".into(),
+                meilisearch_types::error::Code::MissingSearchHybrid,
+            ));
+        }
+
+        Ok(Self {
             q: other.q,
             vector: other.vector.map(CS::into_inner),
             offset: other.offset.0,
@@ -175,7 +189,8 @@ impl From<SearchQueryGet> for SearchQuery {
             attributes_to_search_on: other.attributes_to_search_on.map(|o| o.into_iter().collect()),
             hybrid,
             ranking_score_threshold: other.ranking_score_threshold.map(|o| o.0),
-        }
+            locales: other.locales.map(|o| o.into_iter().collect()),
+        })
     }
 }
 
@@ -215,7 +230,7 @@ pub async fn search_with_url_query(
     debug!(parameters = ?params, "Search get");
     let index_uid = IndexUid::try_from(index_uid.into_inner())?;
 
-    let mut query: SearchQuery = params.into_inner().into();
+    let mut query: SearchQuery = params.into_inner().try_into()?;
 
     // Tenant token search_rules.
     if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) {
@@ -229,11 +244,13 @@ pub async fn search_with_url_query(
 
     let search_kind = search_kind(&query, index_scheduler.get_ref(), &index, features)?;
     let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features)?;
-    let _permit = search_queue.try_get_search_permit().await?;
+    let permit = search_queue.try_get_search_permit().await?;
     let search_result = tokio::task::spawn_blocking(move || {
-        perform_search(&index, query, search_kind, retrieve_vector)
+        perform_search(&index, query, search_kind, retrieve_vector, index_scheduler.features())
     })
-    .await?;
+    .await;
+    permit.drop().await;
+    let search_result = search_result?;
     if let Ok(ref search_result) = search_result {
         aggregate.succeed(search_result);
     }
@@ -272,11 +289,13 @@ pub async fn search_with_post(
     let search_kind = search_kind(&query, index_scheduler.get_ref(), &index, features)?;
     let retrieve_vectors = RetrieveVectors::new(query.retrieve_vectors, features)?;
 
-    let _permit = search_queue.try_get_search_permit().await?;
+    let permit = search_queue.try_get_search_permit().await?;
     let search_result = tokio::task::spawn_blocking(move || {
-        perform_search(&index, query, search_kind, retrieve_vectors)
+        perform_search(&index, query, search_kind, retrieve_vectors, index_scheduler.features())
     })
-    .await?;
+    .await;
+    permit.drop().await;
+    let search_result = search_result?;
     if let Ok(ref search_result) = search_result {
         aggregate.succeed(search_result);
         if search_result.degraded {
@@ -304,44 +323,36 @@ pub fn search_kind(
         features.check_vector("Passing `hybrid` as a parameter")?;
     }
 
-    // regardless of anything, always do a keyword search when we don't have a vector and the query is whitespace or missing
-    if query.vector.is_none() {
-        match &query.q {
-            Some(q) if q.trim().is_empty() => return Ok(SearchKind::KeywordOnly),
-            None => return Ok(SearchKind::KeywordOnly),
-            _ => {}
+    // handle with care, the order of cases matters, the semantics is subtle
+    match (query.q.as_deref(), &query.hybrid, query.vector.as_deref()) {
+        // empty query, no vector => placeholder search
+        (Some(q), _, None) if q.trim().is_empty() => Ok(SearchKind::KeywordOnly),
+        // no query, no vector => placeholder search
+        (None, _, None) => Ok(SearchKind::KeywordOnly),
+        // hybrid.semantic_ratio == 1.0 => vector
+        (_, Some(HybridQuery { semantic_ratio, embedder }), v) if **semantic_ratio == 1.0 => {
+            SearchKind::semantic(index_scheduler, index, embedder, v.map(|v| v.len()))
         }
-    }
-
-    match &query.hybrid {
-        Some(HybridQuery { semantic_ratio, embedder }) if **semantic_ratio == 1.0 => {
-            Ok(SearchKind::semantic(
-                index_scheduler,
-                index,
-                embedder.as_deref(),
-                query.vector.as_ref().map(Vec::len),
-            )?)
-        }
-        Some(HybridQuery { semantic_ratio, embedder: _ }) if **semantic_ratio == 0.0 => {
+        // hybrid.semantic_ratio == 0.0 => keyword
+        (_, Some(HybridQuery { semantic_ratio, embedder: _ }), _) if **semantic_ratio == 0.0 => {
             Ok(SearchKind::KeywordOnly)
         }
-        Some(HybridQuery { semantic_ratio, embedder }) => Ok(SearchKind::hybrid(
+        // no query, hybrid, vector => semantic
+        (None, Some(HybridQuery { semantic_ratio: _, embedder }), Some(v)) => {
+            SearchKind::semantic(index_scheduler, index, embedder, Some(v.len()))
+        }
+        // query, no hybrid, no vector => keyword
+        (Some(_), None, None) => Ok(SearchKind::KeywordOnly),
+        // query, hybrid, maybe vector => hybrid
+        (Some(_), Some(HybridQuery { semantic_ratio, embedder }), v) => SearchKind::hybrid(
             index_scheduler,
             index,
-            embedder.as_deref(),
+            embedder,
             **semantic_ratio,
-            query.vector.as_ref().map(Vec::len),
-        )?),
-        None => match (query.q.as_deref(), query.vector.as_deref()) {
-            (_query, None) => Ok(SearchKind::KeywordOnly),
-            (None, Some(_vector)) => Ok(SearchKind::semantic(
-                index_scheduler,
-                index,
-                None,
-                query.vector.as_ref().map(Vec::len),
-            )?),
-            (Some(_), Some(_)) => Err(MeilisearchHttpError::MissingSearchHybrid.into()),
-        },
+            v.map(|v| v.len()),
+        ),
+
+        (_, None, Some(_)) => Err(MeilisearchHttpError::MissingSearchHybrid.into()),
     }
 }
 

meilisearch/src/routes/indexes/settings.rs

@@ -474,6 +474,28 @@ make_setting_route!(
     }
 );
 
+make_setting_route!(
+    "/localized-attributes",
+    put,
+    Vec<meilisearch_types::locales::LocalizedAttributesRuleView>,
+    meilisearch_types::deserr::DeserrJsonError<
+        meilisearch_types::error::deserr_codes::InvalidSettingsLocalizedAttributes,
+    >,
+    localized_attributes,
+    "localizedAttributes",
+    analytics,
+    |rules: &Option<Vec<meilisearch_types::locales::LocalizedAttributesRuleView>>, req: &HttpRequest| {
+        use serde_json::json;
+        analytics.publish(
+            "LocalizedAttributesRules Updated".to_string(),
+            json!({
+                "locales": rules.as_ref().map(|rules| rules.iter().flat_map(|rule| rule.locales.iter().cloned()).collect::<std::collections::BTreeSet<_>>())
+            }),
+            Some(req),
+        );
+    }
+);
+
 make_setting_route!(
     "/ranking-rules",
     put,
@@ -614,11 +636,26 @@ fn embedder_analytics(
             .any(|config| config.document_template.set().is_some())
     });
 
+    let document_template_max_bytes = setting.as_ref().and_then(|map| {
+        map.values()
+            .filter_map(|config| config.clone().set())
+            .filter_map(|config| config.document_template_max_bytes.set())
+            .max()
+    });
+
+    let binary_quantization_used = setting.as_ref().map(|map| {
+        map.values()
+            .filter_map(|config| config.clone().set())
+            .any(|config| config.binary_quantized.set().is_some())
+    });
+
     json!(
         {
             "total": setting.as_ref().map(|s| s.len()),
             "sources": sources,
             "document_template_used": document_template_used,
+            "document_template_max_bytes": document_template_max_bytes,
+            "binary_quantization_used": binary_quantization_used,
         }
     )
 }
@@ -660,6 +697,7 @@ generate_configure!(
     filterable_attributes,
     sortable_attributes,
     displayed_attributes,
+    localized_attributes,
     searchable_attributes,
     distinct_attribute,
     proximity_precision,
@@ -786,6 +824,7 @@ pub async fn update_all(
             },
             "embedders": crate::routes::indexes::settings::embedder_analytics(new_settings.embedders.as_ref().set()),
             "search_cutoff_ms": new_settings.search_cutoff_ms.as_ref().set(),
+            "locales": new_settings.localized_attributes.as_ref().set().map(|rules| rules.iter().flat_map(|rule| rule.locales.iter().cloned()).collect::<std::collections::BTreeSet<_>>()),
         }),
         Some(&req),
     );

meilisearch/src/routes/indexes/similar.rs

@@ -102,11 +102,19 @@ async fn similar(
 
     let index = index_scheduler.index(&index_uid)?;
 
-    let (embedder_name, embedder) =
-        SearchKind::embedder(&index_scheduler, &index, query.embedder.as_deref(), None)?;
+    let (embedder_name, embedder, quantized) =
+        SearchKind::embedder(&index_scheduler, &index, &query.embedder, None)?;
 
     tokio::task::spawn_blocking(move || {
-        perform_similar(&index, query, embedder_name, embedder, retrieve_vectors)
+        perform_similar(
+            &index,
+            query,
+            embedder_name,
+            embedder,
+            quantized,
+            retrieve_vectors,
+            index_scheduler.features(),
+        )
     })
     .await?
 }
@@ -132,8 +140,8 @@ pub struct SimilarQueryGet {
     show_ranking_score_details: Param<bool>,
     #[deserr(default, error = DeserrQueryParamError<InvalidSimilarRankingScoreThreshold>, default)]
     pub ranking_score_threshold: Option<RankingScoreThresholdGet>,
-    #[deserr(default, error = DeserrQueryParamError<InvalidEmbedder>)]
-    pub embedder: Option<String>,
+    #[deserr(error = DeserrQueryParamError<InvalidEmbedder>)]
+    pub embedder: String,
 }
 
 #[derive(Debug, Clone, Copy, PartialEq, deserr::Deserr)]

meilisearch/src/routes/multi_search.rs

@@ -10,12 +10,14 @@ use serde::Serialize;
 use tracing::debug;
 
 use crate::analytics::{Analytics, MultiSearchAggregator};
+use crate::error::MeilisearchHttpError;
 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::search::{
-    add_search_rules, perform_search, RetrieveVectors, SearchQueryWithIndex, SearchResultWithIndex,
+    add_search_rules, perform_federated_search, perform_search, FederatedSearch, RetrieveVectors,
+    SearchQueryWithIndex, SearchResultWithIndex,
 };
 use crate::search_queue::SearchQueue;
 
@@ -28,85 +30,44 @@ struct SearchResults {
     results: Vec<SearchResultWithIndex>,
 }
 
-#[derive(Debug, deserr::Deserr)]
-#[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
-pub struct SearchQueries {
-    queries: Vec<SearchQueryWithIndex>,
-}
-
 pub async fn multi_search_with_post(
     index_scheduler: GuardedData<ActionPolicy<{ actions::SEARCH }>, Data<IndexScheduler>>,
     search_queue: Data<SearchQueue>,
-    params: AwebJson<SearchQueries, DeserrJsonError>,
+    params: AwebJson<FederatedSearch, DeserrJsonError>,
     req: HttpRequest,
     analytics: web::Data<dyn Analytics>,
 ) -> Result<HttpResponse, ResponseError> {
-    let queries = params.into_inner().queries;
-
-    let mut multi_aggregate = MultiSearchAggregator::from_queries(&queries, &req);
-    let features = index_scheduler.features();
-
     // Since we don't want to process half of the search requests and then get a permit refused
     // we're going to get one permit for the whole duration of the multi-search request.
-    let _permit = search_queue.try_get_search_permit().await?;
+    let permit = search_queue.try_get_search_permit().await?;
 
-    // Explicitly expect a `(ResponseError, usize)` for the error type rather than `ResponseError` only,
-    // so that `?` doesn't work if it doesn't use `with_index`, ensuring that it is not forgotten in case of code
-    // changes.
-    let search_results: Result<_, (ResponseError, usize)> = async {
-        let mut search_results = Vec::with_capacity(queries.len());
-        for (query_index, (index_uid, mut query)) in
-            queries.into_iter().map(SearchQueryWithIndex::into_index_query).enumerate()
-        {
-            debug!(on_index = query_index, parameters = ?query, "Multi-search");
+    let federated_search = params.into_inner();
 
+    let mut multi_aggregate = MultiSearchAggregator::from_federated_search(&federated_search, &req);
+
+    let FederatedSearch { mut queries, federation } = federated_search;
+
+    let features = index_scheduler.features();
+
+    // regardless of federation, check authorization and apply search rules
+    let auth = 'check_authorization: {
+        for (query_index, federated_query) in queries.iter_mut().enumerate() {
+            let index_uid = federated_query.index_uid.as_str();
             // Check index from API key
-            if !index_scheduler.filters().is_index_authorized(&index_uid) {
-                return Err(AuthenticationError::InvalidToken).with_index(query_index);
+            if !index_scheduler.filters().is_index_authorized(index_uid) {
+                break 'check_authorization Err(AuthenticationError::InvalidToken)
+                    .with_index(query_index);
             }
             // Apply search rules from tenant token
-            if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid)
+            if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(index_uid)
             {
-                add_search_rules(&mut query.filter, search_rules);
+                add_search_rules(&mut federated_query.filter, search_rules);
             }
-
-            let index = index_scheduler
-                .index(&index_uid)
-                .map_err(|err| {
-                    let mut err = ResponseError::from(err);
-                    // Patch the HTTP status code to 400 as it defaults to 404 for `index_not_found`, but
-                    // here the resource not found is not part of the URL.
-                    err.code = StatusCode::BAD_REQUEST;
-                    err
-                })
-                .with_index(query_index)?;
-
-            let search_kind = search_kind(&query, index_scheduler.get_ref(), &index, features)
-                .with_index(query_index)?;
-            let retrieve_vector =
-                RetrieveVectors::new(query.retrieve_vectors, features).with_index(query_index)?;
-
-            let search_result = tokio::task::spawn_blocking(move || {
-                perform_search(&index, query, search_kind, retrieve_vector)
-            })
-            .await
-            .with_index(query_index)?;
-
-            search_results.push(SearchResultWithIndex {
-                index_uid: index_uid.into_inner(),
-                result: search_result.with_index(query_index)?,
-            });
         }
-        Ok(search_results)
-    }
-    .await;
+        Ok(())
+    };
 
-    if search_results.is_ok() {
-        multi_aggregate.succeed();
-    }
-    analytics.post_multi_search(multi_aggregate);
-
-    let search_results = search_results.map_err(|(mut err, query_index)| {
+    auth.map_err(|(mut err, query_index)| {
         // Add the query index that failed as context for the error message.
         // We're doing it only here and not directly in the `WithIndex` trait so that the `with_index` function returns a different type
         // of result and we can benefit from static typing.
@@ -114,9 +75,97 @@ pub async fn multi_search_with_post(
         err
     })?;
 
-    debug!(returns = ?search_results, "Multi-search");
+    let response = match federation {
+        Some(federation) => {
+            let search_result = tokio::task::spawn_blocking(move || {
+                perform_federated_search(&index_scheduler, queries, federation, features)
+            })
+            .await;
+            permit.drop().await;
 
-    Ok(HttpResponse::Ok().json(SearchResults { results: search_results }))
+            if let Ok(Ok(_)) = search_result {
+                multi_aggregate.succeed();
+            }
+
+            analytics.post_multi_search(multi_aggregate);
+            HttpResponse::Ok().json(search_result??)
+        }
+        None => {
+            // Explicitly expect a `(ResponseError, usize)` for the error type rather than `ResponseError` only,
+            // so that `?` doesn't work if it doesn't use `with_index`, ensuring that it is not forgotten in case of code
+            // 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
+                    .into_iter()
+                    .map(SearchQueryWithIndex::into_index_query_federation)
+                    .enumerate()
+                {
+                    debug!(on_index = query_index, parameters = ?query, "Multi-search");
+
+                    if federation_options.is_some() {
+                        return Err((
+                            MeilisearchHttpError::FederationOptionsInNonFederatedRequest(
+                                query_index,
+                            )
+                            .into(),
+                            query_index,
+                        ));
+                    }
+
+                    let index = index_scheduler
+                        .index(&index_uid)
+                        .map_err(|err| {
+                            let mut err = ResponseError::from(err);
+                            // Patch the HTTP status code to 400 as it defaults to 404 for `index_not_found`, but
+                            // here the resource not found is not part of the URL.
+                            err.code = StatusCode::BAD_REQUEST;
+                            err
+                        })
+                        .with_index(query_index)?;
+
+                    let search_kind =
+                        search_kind(&query, index_scheduler.get_ref(), &index, features)
+                            .with_index(query_index)?;
+                    let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features)
+                        .with_index(query_index)?;
+
+                    let search_result = tokio::task::spawn_blocking(move || {
+                        perform_search(&index, query, search_kind, retrieve_vector, features)
+                    })
+                    .await
+                    .with_index(query_index)?;
+
+                    search_results.push(SearchResultWithIndex {
+                        index_uid: index_uid.into_inner(),
+                        result: search_result.with_index(query_index)?,
+                    });
+                }
+                Ok(search_results)
+            }
+            .await;
+            permit.drop().await;
+
+            if search_results.is_ok() {
+                multi_aggregate.succeed();
+            }
+            analytics.post_multi_search(multi_aggregate);
+
+            let search_results = search_results.map_err(|(mut err, query_index)| {
+                // Add the query index that failed as context for the error message.
+                // We're doing it only here and not directly in the `WithIndex` trait so that the `with_index` function returns a different type
+                // of result and we can benefit from static typing.
+                err.message = format!("Inside `.queries[{query_index}]`: {}", err.message);
+                err
+            })?;
+
+            debug!(returns = ?search_results, "Multi-search");
+
+            HttpResponse::Ok().json(SearchResults { results: search_results })
+        }
+    };
+
+    Ok(response)
 }
 
 /// Local `Result` extension trait to avoid `map_err` boilerplate.

meilisearch/src/routes/tasks.rs

@@ -616,7 +616,7 @@ mod tests {
             let err = deserr_query_params::<TaskDeletionOrCancelationQuery>(params).unwrap_err();
             snapshot!(meili_snap::json_string!(err), @r###"
             {
-              "message": "Invalid value in parameter `indexUids[1]`: `hé` is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_).",
+              "message": "Invalid value in parameter `indexUids[1]`: `hé` is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_), and can not be more than 512 bytes.",
               "code": "invalid_index_uid",
               "type": "invalid_request",
               "link": "https://docs.meilisearch.com/errors#invalid_index_uid"
@@ -628,7 +628,7 @@ mod tests {
             let err = deserr_query_params::<TaskDeletionOrCancelationQuery>(params).unwrap_err();
             snapshot!(meili_snap::json_string!(err), @r###"
             {
-              "message": "Invalid value in parameter `indexUids`: `hé` is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_).",
+              "message": "Invalid value in parameter `indexUids`: `hé` is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_), and can not be more than 512 bytes.",
               "code": "invalid_index_uid",
               "type": "invalid_request",
               "link": "https://docs.meilisearch.com/errors#invalid_index_uid"

meilisearch/src/search/federated.rs

@@ -0,0 +1,910 @@
+use std::cmp::Ordering;
+use std::collections::BTreeMap;
+use std::fmt;
+use std::iter::Zip;
+use std::rc::Rc;
+use std::str::FromStr as _;
+use std::time::Duration;
+use std::vec::{IntoIter, Vec};
+
+use actix_http::StatusCode;
+use index_scheduler::{IndexScheduler, RoFeatures};
+use indexmap::IndexMap;
+use meilisearch_types::deserr::DeserrJsonError;
+use meilisearch_types::error::deserr_codes::{
+    InvalidMultiSearchFacetsByIndex, InvalidMultiSearchMaxValuesPerFacet,
+    InvalidMultiSearchMergeFacets, InvalidMultiSearchWeight, InvalidSearchLimit,
+    InvalidSearchOffset,
+};
+use meilisearch_types::error::ResponseError;
+use meilisearch_types::index_uid::IndexUid;
+use meilisearch_types::milli::score_details::{ScoreDetails, ScoreValue};
+use meilisearch_types::milli::{self, DocumentId, OrderBy, TimeBudget};
+use roaring::RoaringBitmap;
+use serde::Serialize;
+
+use super::ranking_rules::{self, RankingRules};
+use super::{
+    compute_facet_distribution_stats, prepare_search, AttributesFormat, ComputedFacets, FacetStats,
+    HitMaker, HitsInfo, RetrieveVectors, SearchHit, SearchKind, SearchQuery, SearchQueryWithIndex,
+};
+use crate::error::MeilisearchHttpError;
+use crate::routes::indexes::search::search_kind;
+
+pub const DEFAULT_FEDERATED_WEIGHT: f64 = 1.0;
+
+#[derive(Debug, Default, Clone, Copy, PartialEq, deserr::Deserr)]
+#[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
+pub struct FederationOptions {
+    #[deserr(default, error = DeserrJsonError<InvalidMultiSearchWeight>)]
+    pub weight: Weight,
+}
+
+#[derive(Debug, Clone, Copy, PartialEq, deserr::Deserr)]
+#[deserr(try_from(f64) = TryFrom::try_from -> InvalidMultiSearchWeight)]
+pub struct Weight(f64);
+
+impl Default for Weight {
+    fn default() -> Self {
+        Weight(DEFAULT_FEDERATED_WEIGHT)
+    }
+}
+
+impl std::convert::TryFrom<f64> for Weight {
+    type Error = InvalidMultiSearchWeight;
+
+    fn try_from(f: f64) -> Result<Self, Self::Error> {
+        if f < 0.0 {
+            Err(InvalidMultiSearchWeight)
+        } else {
+            Ok(Weight(f))
+        }
+    }
+}
+
+impl std::ops::Deref for Weight {
+    type Target = f64;
+
+    fn deref(&self) -> &Self::Target {
+        &self.0
+    }
+}
+
+#[derive(Debug, deserr::Deserr)]
+#[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
+pub struct Federation {
+    #[deserr(default = super::DEFAULT_SEARCH_LIMIT(), error = DeserrJsonError<InvalidSearchLimit>)]
+    pub limit: usize,
+    #[deserr(default = super::DEFAULT_SEARCH_OFFSET(), error = DeserrJsonError<InvalidSearchOffset>)]
+    pub offset: usize,
+    #[deserr(default, error = DeserrJsonError<InvalidMultiSearchFacetsByIndex>)]
+    pub facets_by_index: BTreeMap<IndexUid, Option<Vec<String>>>,
+    #[deserr(default, error = DeserrJsonError<InvalidMultiSearchMergeFacets>)]
+    pub merge_facets: Option<MergeFacets>,
+}
+
+#[derive(Copy, Clone, Debug, deserr::Deserr, Default)]
+#[deserr(error = DeserrJsonError<InvalidMultiSearchMergeFacets>, rename_all = camelCase, deny_unknown_fields)]
+pub struct MergeFacets {
+    #[deserr(default, error = DeserrJsonError<InvalidMultiSearchMaxValuesPerFacet>)]
+    pub max_values_per_facet: Option<usize>,
+}
+
+#[derive(Debug, deserr::Deserr)]
+#[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
+pub struct FederatedSearch {
+    pub queries: Vec<SearchQueryWithIndex>,
+    #[deserr(default)]
+    pub federation: Option<Federation>,
+}
+#[derive(Serialize, Clone)]
+#[serde(rename_all = "camelCase")]
+pub struct FederatedSearchResult {
+    pub hits: Vec<SearchHit>,
+    pub processing_time_ms: u128,
+    #[serde(flatten)]
+    pub hits_info: HitsInfo,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub semantic_hit_count: Option<u32>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub facet_distribution: Option<BTreeMap<String, IndexMap<String, u64>>>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub facet_stats: Option<BTreeMap<String, FacetStats>>,
+    #[serde(skip_serializing_if = "FederatedFacets::is_empty")]
+    pub facets_by_index: FederatedFacets,
+
+    // These fields are only used for analytics purposes
+    #[serde(skip)]
+    pub degraded: bool,
+    #[serde(skip)]
+    pub used_negative_operator: bool,
+}
+
+impl fmt::Debug for FederatedSearchResult {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        let FederatedSearchResult {
+            hits,
+            processing_time_ms,
+            hits_info,
+            semantic_hit_count,
+            degraded,
+            used_negative_operator,
+            facet_distribution,
+            facet_stats,
+            facets_by_index,
+        } = self;
+
+        let mut debug = f.debug_struct("SearchResult");
+        // The most important thing when looking at a search result is the time it took to process
+        debug.field("processing_time_ms", &processing_time_ms);
+        debug.field("hits", &format!("[{} hits returned]", hits.len()));
+        debug.field("hits_info", &hits_info);
+        if *used_negative_operator {
+            debug.field("used_negative_operator", used_negative_operator);
+        }
+        if *degraded {
+            debug.field("degraded", degraded);
+        }
+        if let Some(facet_distribution) = facet_distribution {
+            debug.field("facet_distribution", &facet_distribution);
+        }
+        if let Some(facet_stats) = facet_stats {
+            debug.field("facet_stats", &facet_stats);
+        }
+        if let Some(semantic_hit_count) = semantic_hit_count {
+            debug.field("semantic_hit_count", &semantic_hit_count);
+        }
+        if !facets_by_index.is_empty() {
+            debug.field("facets_by_index", &facets_by_index);
+        }
+
+        debug.finish()
+    }
+}
+
+struct WeightedScore<'a> {
+    details: &'a [ScoreDetails],
+    weight: f64,
+}
+
+impl<'a> WeightedScore<'a> {
+    pub fn new(details: &'a [ScoreDetails], weight: f64) -> Self {
+        Self { details, weight }
+    }
+
+    pub fn weighted_global_score(&self) -> f64 {
+        ScoreDetails::global_score(self.details.iter()) * self.weight
+    }
+
+    pub fn compare_weighted_global_scores(&self, other: &Self) -> Ordering {
+        self.weighted_global_score()
+            .partial_cmp(&other.weighted_global_score())
+            // both are numbers, possibly infinite
+            .unwrap()
+    }
+
+    pub fn compare(&self, other: &Self) -> Ordering {
+        let mut left_it = ScoreDetails::score_values(self.details.iter());
+        let mut right_it = ScoreDetails::score_values(other.details.iter());
+
+        loop {
+            let left = left_it.next();
+            let right = right_it.next();
+
+            match (left, right) {
+                (None, None) => return Ordering::Equal,
+                (None, Some(_)) => return Ordering::Less,
+                (Some(_), None) => return Ordering::Greater,
+                (Some(ScoreValue::Score(left)), Some(ScoreValue::Score(right))) => {
+                    let left = left * self.weight;
+                    let right = right * other.weight;
+                    if (left - right).abs() <= f64::EPSILON {
+                        continue;
+                    }
+                    return left.partial_cmp(&right).unwrap();
+                }
+                (Some(ScoreValue::Sort(left)), Some(ScoreValue::Sort(right))) => {
+                    match left.partial_cmp(right) {
+                        Some(Ordering::Equal) => continue,
+                        Some(order) => return order,
+                        None => return self.compare_weighted_global_scores(other),
+                    }
+                }
+                (Some(ScoreValue::GeoSort(left)), Some(ScoreValue::GeoSort(right))) => {
+                    match left.partial_cmp(right) {
+                        Some(Ordering::Equal) => continue,
+                        Some(order) => return order,
+                        None => {
+                            return self.compare_weighted_global_scores(other);
+                        }
+                    }
+                }
+                // not comparable details, use global
+                (Some(ScoreValue::Score(_)), Some(_))
+                | (Some(_), Some(ScoreValue::Score(_)))
+                | (Some(ScoreValue::GeoSort(_)), Some(ScoreValue::Sort(_)))
+                | (Some(ScoreValue::Sort(_)), Some(ScoreValue::GeoSort(_))) => {
+                    let left_count = left_it.count();
+                    let right_count = right_it.count();
+                    // compare how many remaining groups of rules each side has.
+                    // the group with the most remaining groups wins.
+                    return left_count
+                        .cmp(&right_count)
+                        // breaks ties with the global ranking score
+                        .then_with(|| self.compare_weighted_global_scores(other));
+                }
+            }
+        }
+    }
+}
+
+struct QueryByIndex {
+    query: SearchQuery,
+    federation_options: FederationOptions,
+    query_index: usize,
+}
+
+struct SearchResultByQuery<'a> {
+    documents_ids: Vec<DocumentId>,
+    document_scores: Vec<Vec<ScoreDetails>>,
+    federation_options: FederationOptions,
+    hit_maker: HitMaker<'a>,
+    query_index: usize,
+}
+
+struct SearchResultByQueryIter<'a> {
+    it: Zip<IntoIter<DocumentId>, IntoIter<Vec<ScoreDetails>>>,
+    federation_options: FederationOptions,
+    hit_maker: Rc<HitMaker<'a>>,
+    query_index: usize,
+}
+
+impl<'a> SearchResultByQueryIter<'a> {
+    fn new(
+        SearchResultByQuery {
+            documents_ids,
+            document_scores,
+            federation_options,
+            hit_maker,
+            query_index,
+        }: SearchResultByQuery<'a>,
+    ) -> Self {
+        let it = documents_ids.into_iter().zip(document_scores);
+        Self { it, federation_options, hit_maker: Rc::new(hit_maker), query_index }
+    }
+}
+
+struct SearchResultByQueryIterItem<'a> {
+    docid: DocumentId,
+    score: Vec<ScoreDetails>,
+    federation_options: FederationOptions,
+    hit_maker: Rc<HitMaker<'a>>,
+    query_index: usize,
+}
+
+fn merge_index_local_results(
+    results_by_query: Vec<SearchResultByQuery<'_>>,
+) -> impl Iterator<Item = SearchResultByQueryIterItem> + '_ {
+    itertools::kmerge_by(
+        results_by_query.into_iter().map(SearchResultByQueryIter::new),
+        |left: &SearchResultByQueryIterItem, right: &SearchResultByQueryIterItem| {
+            let left_score = WeightedScore::new(&left.score, *left.federation_options.weight);
+            let right_score = WeightedScore::new(&right.score, *right.federation_options.weight);
+
+            match left_score.compare(&right_score) {
+                // the biggest score goes first
+                Ordering::Greater => true,
+                // break ties using query index
+                Ordering::Equal => left.query_index < right.query_index,
+                Ordering::Less => false,
+            }
+        },
+    )
+}
+
+fn merge_index_global_results(
+    results_by_index: Vec<SearchResultByIndex>,
+) -> impl Iterator<Item = SearchHitByIndex> {
+    itertools::kmerge_by(
+        results_by_index.into_iter().map(|result_by_index| result_by_index.hits.into_iter()),
+        |left: &SearchHitByIndex, right: &SearchHitByIndex| {
+            let left_score = WeightedScore::new(&left.score, *left.federation_options.weight);
+            let right_score = WeightedScore::new(&right.score, *right.federation_options.weight);
+
+            match left_score.compare(&right_score) {
+                // the biggest score goes first
+                Ordering::Greater => true,
+                // break ties using query index
+                Ordering::Equal => left.query_index < right.query_index,
+                Ordering::Less => false,
+            }
+        },
+    )
+}
+
+impl<'a> Iterator for SearchResultByQueryIter<'a> {
+    type Item = SearchResultByQueryIterItem<'a>;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        let (docid, score) = self.it.next()?;
+        Some(SearchResultByQueryIterItem {
+            docid,
+            score,
+            federation_options: self.federation_options,
+            hit_maker: Rc::clone(&self.hit_maker),
+            query_index: self.query_index,
+        })
+    }
+}
+
+struct SearchHitByIndex {
+    hit: SearchHit,
+    score: Vec<ScoreDetails>,
+    federation_options: FederationOptions,
+    query_index: usize,
+}
+
+struct SearchResultByIndex {
+    index: String,
+    hits: Vec<SearchHitByIndex>,
+    estimated_total_hits: usize,
+    degraded: bool,
+    used_negative_operator: bool,
+    facets: Option<ComputedFacets>,
+}
+
+#[derive(Debug, Clone, Default, Serialize)]
+pub struct FederatedFacets(pub BTreeMap<String, ComputedFacets>);
+
+impl FederatedFacets {
+    pub fn insert(&mut self, index: String, facets: Option<ComputedFacets>) {
+        if let Some(facets) = facets {
+            self.0.insert(index, facets);
+        }
+    }
+
+    pub fn is_empty(&self) -> bool {
+        self.0.is_empty()
+    }
+
+    pub fn merge(
+        self,
+        MergeFacets { max_values_per_facet }: MergeFacets,
+        facet_order: BTreeMap<String, (String, OrderBy)>,
+    ) -> Option<ComputedFacets> {
+        if self.is_empty() {
+            return None;
+        }
+
+        let mut distribution: BTreeMap<String, _> = Default::default();
+        let mut stats: BTreeMap<String, FacetStats> = Default::default();
+
+        for facets_by_index in self.0.into_values() {
+            for (facet, index_distribution) in facets_by_index.distribution {
+                match distribution.entry(facet) {
+                    std::collections::btree_map::Entry::Vacant(entry) => {
+                        entry.insert(index_distribution);
+                    }
+                    std::collections::btree_map::Entry::Occupied(mut entry) => {
+                        let distribution = entry.get_mut();
+
+                        for (value, index_count) in index_distribution {
+                            distribution
+                                .entry(value)
+                                .and_modify(|count| *count += index_count)
+                                .or_insert(index_count);
+                        }
+                    }
+                }
+            }
+
+            for (facet, index_stats) in facets_by_index.stats {
+                match stats.entry(facet) {
+                    std::collections::btree_map::Entry::Vacant(entry) => {
+                        entry.insert(index_stats);
+                    }
+                    std::collections::btree_map::Entry::Occupied(mut entry) => {
+                        let stats = entry.get_mut();
+
+                        stats.min = f64::min(stats.min, index_stats.min);
+                        stats.max = f64::max(stats.max, index_stats.max);
+                    }
+                }
+            }
+        }
+
+        // fixup order
+        for (facet, values) in &mut distribution {
+            let order_by = facet_order.get(facet).map(|(_, order)| *order).unwrap_or_default();
+
+            match order_by {
+                OrderBy::Lexicographic => {
+                    values.sort_unstable_by(|left, _, right, _| left.cmp(right))
+                }
+                OrderBy::Count => {
+                    values.sort_unstable_by(|_, left, _, right| {
+                        left.cmp(right)
+                            // biggest first
+                            .reverse()
+                    })
+                }
+            }
+
+            if let Some(max_values_per_facet) = max_values_per_facet {
+                values.truncate(max_values_per_facet)
+            };
+        }
+
+        Some(ComputedFacets { distribution, stats })
+    }
+}
+
+pub fn perform_federated_search(
+    index_scheduler: &IndexScheduler,
+    queries: Vec<SearchQueryWithIndex>,
+    mut federation: Federation,
+    features: RoFeatures,
+) -> Result<FederatedSearchResult, ResponseError> {
+    let before_search = std::time::Instant::now();
+
+    // 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 index
+    let mut queries_by_index: BTreeMap<String, Vec<QueryByIndex>> = Default::default();
+    for (query_index, federated_query) in queries.into_iter().enumerate() {
+        if let Some(pagination_field) = federated_query.has_pagination() {
+            return Err(MeilisearchHttpError::PaginationInFederatedQuery(
+                query_index,
+                pagination_field,
+            )
+            .into());
+        }
+
+        if let Some(facets) = federated_query.has_facets() {
+            let facets = facets.to_owned();
+            return Err(MeilisearchHttpError::FacetsInFederatedQuery(
+                query_index,
+                federated_query.index_uid.into_inner(),
+                facets,
+            )
+            .into());
+        }
+
+        let (index_uid, query, federation_options) = federated_query.into_index_query_federation();
+
+        queries_by_index.entry(index_uid.into_inner()).or_default().push(QueryByIndex {
+            query,
+            federation_options: federation_options.unwrap_or_default(),
+            query_index,
+        })
+    }
+
+    // 2. perform queries, merge and make hits index by index
+    let required_hit_count = federation.limit + federation.offset;
+
+    // In step (2), semantic_hit_count will be set to Some(0) if any search kind uses semantic
+    // Then in step (3), we'll update its value if there is any semantic search
+    let mut semantic_hit_count = None;
+    let mut results_by_index = Vec::with_capacity(queries_by_index.len());
+    let mut previous_query_data: Option<(RankingRules, usize, String)> = None;
+
+    // remember the order and name of first index for each facet when merging with index settings
+    // to detect if the order is inconsistent for a facet.
+    let mut facet_order: Option<BTreeMap<String, (String, OrderBy)>> = match federation.merge_facets
+    {
+        Some(MergeFacets { .. }) => Some(Default::default()),
+        _ => None,
+    };
+
+    for (index_uid, queries) in queries_by_index {
+        let first_query_index = queries.first().map(|query| query.query_index);
+
+        let index = match index_scheduler.index(&index_uid) {
+            Ok(index) => index,
+            Err(err) => {
+                let mut err = ResponseError::from(err);
+                // Patch the HTTP status code to 400 as it defaults to 404 for `index_not_found`, but
+                // here the resource not found is not part of the URL.
+                err.code = StatusCode::BAD_REQUEST;
+                if let Some(query_index) = first_query_index {
+                    err.message = format!("Inside `.queries[{}]`: {}", query_index, err.message);
+                }
+                return Err(err);
+            }
+        };
+
+        // Important: this is the only transaction we'll use for this index during this federated search
+        let rtxn = index.read_txn()?;
+
+        let criteria = index.criteria(&rtxn)?;
+
+        let dictionary = index.dictionary(&rtxn)?;
+        let dictionary: Option<Vec<_>> =
+            dictionary.as_ref().map(|x| x.iter().map(String::as_str).collect());
+        let separators = index.allowed_separators(&rtxn)?;
+        let separators: Option<Vec<_>> =
+            separators.as_ref().map(|x| x.iter().map(String::as_str).collect());
+
+        // each query gets its individual cutoff
+        let cutoff = index.search_cutoff(&rtxn)?;
+
+        let mut degraded = false;
+        let mut used_negative_operator = false;
+        let mut candidates = RoaringBitmap::new();
+
+        let facets_by_index = federation.facets_by_index.remove(&index_uid).flatten();
+
+        // TODO: recover the max size + facets_by_index as return value of this function so as not to ask it for all queries
+        if let Err(mut error) =
+            check_facet_order(&mut facet_order, &index_uid, &facets_by_index, &index, &rtxn)
+        {
+            error.message = format!(
+                "Inside `.federation.facetsByIndex.{index_uid}`: {error}{}",
+                if let Some(query_index) = first_query_index {
+                    format!("\n - Note: index `{index_uid}` used in `.queries[{query_index}]`")
+                } else {
+                    Default::default()
+                }
+            );
+            return Err(error);
+        }
+
+        // 2.1. Compute all candidates for each query in the index
+        let mut results_by_query = Vec::with_capacity(queries.len());
+
+        for QueryByIndex { query, federation_options, query_index } in queries {
+            // use an immediately invoked lambda to capture the result without returning from the function
+
+            let res: Result<(), ResponseError> = (|| {
+                let search_kind = search_kind(&query, index_scheduler, &index, features)?;
+
+                let canonicalization_kind = match (&search_kind, &query.q) {
+                    (SearchKind::SemanticOnly { .. }, _) => {
+                        ranking_rules::CanonicalizationKind::Vector
+                    }
+                    (_, Some(q)) if !q.is_empty() => ranking_rules::CanonicalizationKind::Keyword,
+                    _ => ranking_rules::CanonicalizationKind::Placeholder,
+                };
+
+                let sort = if let Some(sort) = &query.sort {
+                    let sorts: Vec<_> =
+                        match sort.iter().map(|s| milli::AscDesc::from_str(s)).collect() {
+                            Ok(sorts) => sorts,
+                            Err(asc_desc_error) => {
+                                return Err(milli::Error::from(milli::SortError::from(
+                                    asc_desc_error,
+                                ))
+                                .into())
+                            }
+                        };
+                    Some(sorts)
+                } else {
+                    None
+                };
+
+                let ranking_rules = ranking_rules::RankingRules::new(
+                    criteria.clone(),
+                    sort,
+                    query.matching_strategy.into(),
+                    canonicalization_kind,
+                );
+
+                if let Some((previous_ranking_rules, previous_query_index, previous_index_uid)) =
+                    previous_query_data.take()
+                {
+                    if let Err(error) = ranking_rules.is_compatible_with(&previous_ranking_rules) {
+                        return Err(error.to_response_error(
+                            &ranking_rules,
+                            &previous_ranking_rules,
+                            query_index,
+                            previous_query_index,
+                            &index_uid,
+                            &previous_index_uid,
+                        ));
+                    }
+                    previous_query_data = if previous_ranking_rules.constraint_count()
+                        > ranking_rules.constraint_count()
+                    {
+                        Some((previous_ranking_rules, previous_query_index, previous_index_uid))
+                    } else {
+                        Some((ranking_rules, query_index, index_uid.clone()))
+                    };
+                } else {
+                    previous_query_data = Some((ranking_rules, query_index, index_uid.clone()));
+                }
+
+                match search_kind {
+                    SearchKind::KeywordOnly => {}
+                    _ => semantic_hit_count = Some(0),
+                }
+
+                let retrieve_vectors = RetrieveVectors::new(query.retrieve_vectors, features)?;
+
+                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, features)?;
+
+                search.scoring_strategy(milli::score_details::ScoringStrategy::Detailed);
+                search.offset(0);
+                search.limit(required_hit_count);
+
+                let (result, _semantic_hit_count) = super::search_from_kind(search_kind, search)?;
+                let format = AttributesFormat {
+                    attributes_to_retrieve: query.attributes_to_retrieve,
+                    retrieve_vectors,
+                    attributes_to_highlight: query.attributes_to_highlight,
+                    attributes_to_crop: query.attributes_to_crop,
+                    crop_length: query.crop_length,
+                    crop_marker: query.crop_marker,
+                    highlight_pre_tag: query.highlight_pre_tag,
+                    highlight_post_tag: query.highlight_post_tag,
+                    show_matches_position: query.show_matches_position,
+                    sort: query.sort,
+                    show_ranking_score: query.show_ranking_score,
+                    show_ranking_score_details: query.show_ranking_score_details,
+                    locales: query.locales.map(|l| l.iter().copied().map(Into::into).collect()),
+                };
+
+                let milli::SearchResult {
+                    matching_words,
+                    candidates: query_candidates,
+                    documents_ids,
+                    document_scores,
+                    degraded: query_degraded,
+                    used_negative_operator: query_used_negative_operator,
+                } = result;
+
+                candidates |= query_candidates;
+                degraded |= query_degraded;
+                used_negative_operator |= query_used_negative_operator;
+
+                let tokenizer = HitMaker::tokenizer(dictionary.as_deref(), separators.as_deref());
+
+                let formatter_builder = HitMaker::formatter_builder(matching_words, tokenizer);
+
+                let hit_maker = HitMaker::new(&index, &rtxn, format, formatter_builder)?;
+
+                results_by_query.push(SearchResultByQuery {
+                    federation_options,
+                    hit_maker,
+                    query_index,
+                    documents_ids,
+                    document_scores,
+                });
+                Ok(())
+            })();
+
+            if let Err(mut error) = res {
+                error.message = format!("Inside `.queries[{query_index}]`: {}", error.message);
+                return Err(error);
+            }
+        }
+        // 2.2. merge inside index
+        let mut documents_seen = RoaringBitmap::new();
+        let merged_result: Result<Vec<_>, ResponseError> =
+            merge_index_local_results(results_by_query)
+                // skip documents we've already seen & mark that we saw the current document
+                .filter(|SearchResultByQueryIterItem { docid, .. }| documents_seen.insert(*docid))
+                .take(required_hit_count)
+                // 2.3 make hits
+                .map(
+                    |SearchResultByQueryIterItem {
+                         docid,
+                         score,
+                         federation_options,
+                         hit_maker,
+                         query_index,
+                     }| {
+                        let mut hit = hit_maker.make_hit(docid, &score)?;
+                        let weighted_score =
+                            ScoreDetails::global_score(score.iter()) * (*federation_options.weight);
+
+                        let _federation = serde_json::json!(
+                            {
+                                "indexUid": index_uid,
+                                "queriesPosition": query_index,
+                                "weightedRankingScore": weighted_score,
+                            }
+                        );
+                        hit.document.insert("_federation".to_string(), _federation);
+                        Ok(SearchHitByIndex { hit, score, federation_options, query_index })
+                    },
+                )
+                .collect();
+
+        let merged_result = merged_result?;
+
+        let estimated_total_hits = candidates.len() as usize;
+
+        let facets = facets_by_index
+            .map(|facets_by_index| {
+                compute_facet_distribution_stats(
+                    &facets_by_index,
+                    &index,
+                    &rtxn,
+                    candidates,
+                    super::Route::MultiSearch,
+                )
+            })
+            .transpose()
+            .map_err(|mut error| {
+                error.message = format!(
+                    "Inside `.federation.facetsByIndex.{index_uid}`: {}{}",
+                    error.message,
+                    if let Some(query_index) = first_query_index {
+                        format!("\n - Note: index `{index_uid}` used in `.queries[{query_index}]`")
+                    } else {
+                        Default::default()
+                    }
+                );
+                error
+            })?;
+
+        results_by_index.push(SearchResultByIndex {
+            index: index_uid,
+            hits: merged_result,
+            estimated_total_hits,
+            degraded,
+            used_negative_operator,
+            facets,
+        });
+    }
+
+    // bonus step, make sure to return an error if an index wants a non-faceted field, even if no query actually uses that index.
+    for (index_uid, facets) in federation.facets_by_index {
+        let index = match index_scheduler.index(&index_uid) {
+            Ok(index) => index,
+            Err(err) => {
+                let mut err = ResponseError::from(err);
+                // Patch the HTTP status code to 400 as it defaults to 404 for `index_not_found`, but
+                // here the resource not found is not part of the URL.
+                err.code = StatusCode::BAD_REQUEST;
+                err.message = format!(
+                    "Inside `.federation.facetsByIndex.{index_uid}`: {}\n - Note: index `{index_uid}` is not used in queries",
+                    err.message
+                );
+                return Err(err);
+            }
+        };
+
+        // Important: this is the only transaction we'll use for this index during this federated search
+        let rtxn = index.read_txn()?;
+
+        if let Err(mut error) =
+            check_facet_order(&mut facet_order, &index_uid, &facets, &index, &rtxn)
+        {
+            error.message = format!(
+                "Inside `.federation.facetsByIndex.{index_uid}`: {error}\n - Note: index `{index_uid}` is not used in queries",
+            );
+            return Err(error);
+        }
+
+        if let Some(facets) = facets {
+            if let Err(mut error) = compute_facet_distribution_stats(
+                &facets,
+                &index,
+                &rtxn,
+                Default::default(),
+                super::Route::MultiSearch,
+            ) {
+                error.message =
+                    format!("Inside `.federation.facetsByIndex.{index_uid}`: {}\n - Note: index `{index_uid}` is not used in queries", error.message);
+                return Err(error);
+            }
+        }
+    }
+
+    // 3. merge hits and metadata across indexes
+    // 3.1 merge metadata
+    let (estimated_total_hits, degraded, used_negative_operator, facets) = {
+        let mut estimated_total_hits = 0;
+        let mut degraded = false;
+        let mut used_negative_operator = false;
+
+        let mut facets: FederatedFacets = FederatedFacets::default();
+
+        for SearchResultByIndex {
+            index,
+            hits: _,
+            estimated_total_hits: estimated_total_hits_by_index,
+            facets: facets_by_index,
+            degraded: degraded_by_index,
+            used_negative_operator: used_negative_operator_by_index,
+        } in &mut results_by_index
+        {
+            estimated_total_hits += *estimated_total_hits_by_index;
+            degraded |= *degraded_by_index;
+            used_negative_operator |= *used_negative_operator_by_index;
+
+            let facets_by_index = std::mem::take(facets_by_index);
+            let index = std::mem::take(index);
+
+            facets.insert(index, facets_by_index);
+        }
+
+        (estimated_total_hits, degraded, used_negative_operator, facets)
+    };
+
+    // 3.2 merge hits
+    let merged_hits: Vec<_> = merge_index_global_results(results_by_index)
+        .skip(federation.offset)
+        .take(federation.limit)
+        .inspect(|hit| {
+            if let Some(semantic_hit_count) = &mut semantic_hit_count {
+                if hit.score.iter().any(|score| matches!(&score, ScoreDetails::Vector(_))) {
+                    *semantic_hit_count += 1;
+                }
+            }
+        })
+        .map(|hit| hit.hit)
+        .collect();
+
+    let (facet_distribution, facet_stats, facets_by_index) =
+        match federation.merge_facets.zip(facet_order) {
+            Some((merge_facets, facet_order)) => {
+                let facets = facets.merge(merge_facets, facet_order);
+
+                let (facet_distribution, facet_stats) = facets
+                    .map(|ComputedFacets { distribution, stats }| (distribution, stats))
+                    .unzip();
+
+                (facet_distribution, facet_stats, FederatedFacets::default())
+            }
+            None => (None, None, facets),
+        };
+
+    let search_result = FederatedSearchResult {
+        hits: merged_hits,
+        processing_time_ms: before_search.elapsed().as_millis(),
+        hits_info: HitsInfo::OffsetLimit {
+            limit: federation.limit,
+            offset: federation.offset,
+            estimated_total_hits,
+        },
+        semantic_hit_count,
+        degraded,
+        used_negative_operator,
+        facet_distribution,
+        facet_stats,
+        facets_by_index,
+    };
+
+    Ok(search_result)
+}
+
+fn check_facet_order(
+    facet_order: &mut Option<BTreeMap<String, (String, OrderBy)>>,
+    current_index: &str,
+    facets_by_index: &Option<Vec<String>>,
+    index: &milli::Index,
+    rtxn: &milli::heed::RoTxn<'_>,
+) -> Result<(), ResponseError> {
+    if let (Some(facet_order), Some(facets_by_index)) = (facet_order, facets_by_index) {
+        let index_facet_order = index.sort_facet_values_by(rtxn)?;
+        for facet in facets_by_index {
+            let index_facet_order = index_facet_order.get(facet);
+            let (previous_index, previous_facet_order) = facet_order
+                .entry(facet.to_owned())
+                .or_insert_with(|| (current_index.to_owned(), index_facet_order));
+            if previous_facet_order != &index_facet_order {
+                return Err(MeilisearchHttpError::InconsistentFacetOrder {
+                    facet: facet.clone(),
+                    previous_facet_order: *previous_facet_order,
+                    previous_uid: previous_index.clone(),
+                    current_uid: current_index.to_owned(),
+                    index_facet_order,
+                }
+                .into());
+            }
+        }
+    };
+    Ok(())
+}

meilisearch/src/search/mod.rs

@@ -1,12 +1,13 @@
 use core::fmt;
 use std::cmp::min;
-use std::collections::{BTreeMap, BTreeSet, HashMap, HashSet};
+use std::collections::{BTreeMap, BTreeSet, HashSet};
 use std::str::FromStr;
 use std::sync::Arc;
 use std::time::{Duration, Instant};
 
 use deserr::Deserr;
 use either::Either;
+use index_scheduler::RoFeatures;
 use indexmap::IndexMap;
 use meilisearch_auth::IndexSearchRules;
 use meilisearch_types::deserr::DeserrJsonError;
@@ -14,16 +15,17 @@ use meilisearch_types::error::deserr_codes::*;
 use meilisearch_types::error::{Code, ResponseError};
 use meilisearch_types::heed::RoTxn;
 use meilisearch_types::index_uid::IndexUid;
+use meilisearch_types::locales::Locale;
 use meilisearch_types::milli::score_details::{ScoreDetails, ScoringStrategy};
 use meilisearch_types::milli::vector::parsed_vectors::ExplicitVectors;
 use meilisearch_types::milli::vector::Embedder;
 use meilisearch_types::milli::{FacetValueHit, OrderBy, SearchForFacetValues, TimeBudget};
 use meilisearch_types::settings::DEFAULT_PAGINATION_MAX_TOTAL_HITS;
 use meilisearch_types::{milli, Document};
-use milli::tokenizer::TokenizerBuilder;
+use milli::tokenizer::{Language, TokenizerBuilder};
 use milli::{
-    AscDesc, FieldId, FieldsIdsMap, Filter, FormatOptions, Index, MatchBounds, MatcherBuilder,
-    SortError, TermsMatchingStrategy, DEFAULT_VALUES_PER_FACET,
+    AscDesc, FieldId, FieldsIdsMap, Filter, FormatOptions, Index, LocalizedAttributesRule,
+    MatchBounds, MatcherBuilder, SortError, TermsMatchingStrategy, DEFAULT_VALUES_PER_FACET,
 };
 use regex::Regex;
 use serde::Serialize;
@@ -31,6 +33,11 @@ use serde_json::{json, Value};
 
 use crate::error::MeilisearchHttpError;
 
+mod federated;
+pub use federated::{perform_federated_search, FederatedSearch, Federation, FederationOptions};
+
+mod ranking_rules;
+
 type MatchesPosition = BTreeMap<String, Vec<MatchBounds>>;
 
 pub const DEFAULT_SEARCH_OFFSET: fn() -> usize = || 0;
@@ -94,6 +101,8 @@ pub struct SearchQuery {
     pub attributes_to_search_on: Option<Vec<String>>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchRankingScoreThreshold>, default)]
     pub ranking_score_threshold: Option<RankingScoreThreshold>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchLocales>, default)]
+    pub locales: Option<Vec<Locale>>,
 }
 
 #[derive(Debug, Clone, Copy, PartialEq, Deserr)]
@@ -163,6 +172,7 @@ impl fmt::Debug for SearchQuery {
             matching_strategy,
             attributes_to_search_on,
             ranking_score_threshold,
+            locales,
         } = self;
 
         let mut debug = f.debug_struct("SearchQuery");
@@ -244,6 +254,10 @@ impl fmt::Debug for SearchQuery {
             debug.field("ranking_score_threshold", &ranking_score_threshold);
         }
 
+        if let Some(locales) = locales {
+            debug.field("locales", &locales);
+        }
+
         debug.finish()
     }
 }
@@ -253,58 +267,54 @@ impl fmt::Debug for SearchQuery {
 pub struct HybridQuery {
     #[deserr(default, error = DeserrJsonError<InvalidSearchSemanticRatio>, default)]
     pub semantic_ratio: SemanticRatio,
-    #[deserr(default, error = DeserrJsonError<InvalidEmbedder>, default)]
-    pub embedder: Option<String>,
+    #[deserr(error = DeserrJsonError<InvalidEmbedder>)]
+    pub embedder: String,
 }
 
 #[derive(Clone)]
 pub enum SearchKind {
     KeywordOnly,
-    SemanticOnly { embedder_name: String, embedder: Arc<Embedder> },
-    Hybrid { embedder_name: String, embedder: Arc<Embedder>, semantic_ratio: f32 },
+    SemanticOnly { embedder_name: String, embedder: Arc<Embedder>, quantized: bool },
+    Hybrid { embedder_name: String, embedder: Arc<Embedder>, quantized: bool, semantic_ratio: f32 },
 }
 
 impl SearchKind {
     pub(crate) fn semantic(
         index_scheduler: &index_scheduler::IndexScheduler,
         index: &Index,
-        embedder_name: Option<&str>,
+        embedder_name: &str,
         vector_len: Option<usize>,
     ) -> Result<Self, ResponseError> {
-        let (embedder_name, embedder) =
+        let (embedder_name, embedder, quantized) =
             Self::embedder(index_scheduler, index, embedder_name, vector_len)?;
-        Ok(Self::SemanticOnly { embedder_name, embedder })
+        Ok(Self::SemanticOnly { embedder_name, embedder, quantized })
     }
 
     pub(crate) fn hybrid(
         index_scheduler: &index_scheduler::IndexScheduler,
         index: &Index,
-        embedder_name: Option<&str>,
+        embedder_name: &str,
         semantic_ratio: f32,
         vector_len: Option<usize>,
     ) -> Result<Self, ResponseError> {
-        let (embedder_name, embedder) =
+        let (embedder_name, embedder, quantized) =
             Self::embedder(index_scheduler, index, embedder_name, vector_len)?;
-        Ok(Self::Hybrid { embedder_name, embedder, semantic_ratio })
+        Ok(Self::Hybrid { embedder_name, embedder, quantized, semantic_ratio })
     }
 
     pub(crate) fn embedder(
         index_scheduler: &index_scheduler::IndexScheduler,
         index: &Index,
-        embedder_name: Option<&str>,
+        embedder_name: &str,
         vector_len: Option<usize>,
-    ) -> Result<(String, Arc<Embedder>), ResponseError> {
+    ) -> Result<(String, Arc<Embedder>, bool), ResponseError> {
         let embedder_configs = index.embedding_configs(&index.read_txn()?)?;
         let embedders = index_scheduler.embedders(embedder_configs)?;
 
-        let embedder_name = embedder_name.unwrap_or_else(|| embedders.get_default_embedder_name());
-
-        let embedder = embedders.get(embedder_name);
-
-        let embedder = embedder
+        let (embedder, _, quantized) = embedders
+            .get(embedder_name)
             .ok_or(milli::UserError::InvalidEmbedder(embedder_name.to_owned()))
-            .map_err(milli::Error::from)?
-            .0;
+            .map_err(milli::Error::from)?;
 
         if let Some(vector_len) = vector_len {
             if vector_len != embedder.dimensions() {
@@ -318,7 +328,7 @@ impl SearchKind {
             }
         }
 
-        Ok((embedder_name.to_owned(), embedder))
+        Ok((embedder_name.to_owned(), embedder, quantized))
     }
 }
 
@@ -360,7 +370,7 @@ impl SearchQuery {
     }
 }
 
-/// A `SearchQuery` + an index UID.
+/// A `SearchQuery` + an index UID and optional FederationOptions.
 // This struct contains the fields of `SearchQuery` inline.
 // This is because neither deserr nor serde support `flatten` when using `deny_unknown_fields.
 // The `From<SearchQueryWithIndex>` implementation ensures both structs remain up to date.
@@ -375,10 +385,10 @@ pub struct SearchQueryWithIndex {
     pub vector: Option<Vec<f32>>,
     #[deserr(default, error = DeserrJsonError<InvalidHybridQuery>)]
     pub hybrid: Option<HybridQuery>,
-    #[deserr(default = DEFAULT_SEARCH_OFFSET(), error = DeserrJsonError<InvalidSearchOffset>)]
-    pub offset: usize,
-    #[deserr(default = DEFAULT_SEARCH_LIMIT(), error = DeserrJsonError<InvalidSearchLimit>)]
-    pub limit: usize,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchOffset>)]
+    pub offset: Option<usize>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchLimit>)]
+    pub limit: Option<usize>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchPage>)]
     pub page: Option<usize>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchHitsPerPage>)]
@@ -419,12 +429,36 @@ pub struct SearchQueryWithIndex {
     pub attributes_to_search_on: Option<Vec<String>>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchRankingScoreThreshold>, default)]
     pub ranking_score_threshold: Option<RankingScoreThreshold>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchLocales>, default)]
+    pub locales: Option<Vec<Locale>>,
+
+    #[deserr(default)]
+    pub federation_options: Option<FederationOptions>,
 }
 
 impl SearchQueryWithIndex {
-    pub fn into_index_query(self) -> (IndexUid, SearchQuery) {
+    pub fn has_pagination(&self) -> Option<&'static str> {
+        if self.offset.is_some() {
+            Some("offset")
+        } else if self.limit.is_some() {
+            Some("limit")
+        } else if self.page.is_some() {
+            Some("page")
+        } else if self.hits_per_page.is_some() {
+            Some("hitsPerPage")
+        } else {
+            None
+        }
+    }
+
+    pub fn has_facets(&self) -> Option<&[String]> {
+        self.facets.as_deref().filter(|v| !v.is_empty())
+    }
+
+    pub fn into_index_query_federation(self) -> (IndexUid, SearchQuery, Option<FederationOptions>) {
         let SearchQueryWithIndex {
             index_uid,
+            federation_options,
             q,
             vector,
             offset,
@@ -450,14 +484,15 @@ impl SearchQueryWithIndex {
             attributes_to_search_on,
             hybrid,
             ranking_score_threshold,
+            locales,
         } = self;
         (
             index_uid,
             SearchQuery {
                 q,
                 vector,
-                offset,
-                limit,
+                offset: offset.unwrap_or(DEFAULT_SEARCH_OFFSET()),
+                limit: limit.unwrap_or(DEFAULT_SEARCH_LIMIT()),
                 page,
                 hits_per_page,
                 attributes_to_retrieve,
@@ -479,9 +514,11 @@ impl SearchQueryWithIndex {
                 attributes_to_search_on,
                 hybrid,
                 ranking_score_threshold,
+                locales,
                 // do not use ..Default::default() here,
                 // rather add any missing field from `SearchQuery` to `SearchQueryWithIndex`
             },
+            federation_options,
         )
     }
 }
@@ -497,8 +534,8 @@ pub struct SimilarQuery {
     pub limit: usize,
     #[deserr(default, error = DeserrJsonError<InvalidSimilarFilter>)]
     pub filter: Option<Value>,
-    #[deserr(default, error = DeserrJsonError<InvalidEmbedder>, default)]
-    pub embedder: Option<String>,
+    #[deserr(error = DeserrJsonError<InvalidEmbedder>)]
+    pub embedder: String,
     #[deserr(default, error = DeserrJsonError<InvalidSimilarAttributesToRetrieve>)]
     pub attributes_to_retrieve: Option<BTreeSet<String>>,
     #[deserr(default, error = DeserrJsonError<InvalidSimilarRetrieveVectors>)]
@@ -734,7 +771,8 @@ fn prepare_search<'t>(
     query: &'t SearchQuery,
     search_kind: &SearchKind,
     time_budget: TimeBudget,
-) -> Result<(milli::Search<'t>, bool, usize, usize), MeilisearchHttpError> {
+    features: RoFeatures,
+) -> Result<(milli::Search<'t>, bool, usize, usize), ResponseError> {
     let mut search = index.search(rtxn);
     search.time_budget(time_budget);
     if let Some(ranking_score_threshold) = query.ranking_score_threshold {
@@ -751,7 +789,7 @@ fn prepare_search<'t>(
                 search.query(q);
             }
         }
-        SearchKind::SemanticOnly { embedder_name, embedder } => {
+        SearchKind::SemanticOnly { embedder_name, embedder, quantized } => {
             let vector = match query.vector.clone() {
                 Some(vector) => vector,
                 None => {
@@ -765,14 +803,19 @@ fn prepare_search<'t>(
                 }
             };
 
-            search.semantic(embedder_name.clone(), embedder.clone(), Some(vector));
+            search.semantic(embedder_name.clone(), embedder.clone(), *quantized, Some(vector));
         }
-        SearchKind::Hybrid { embedder_name, embedder, semantic_ratio: _ } => {
+        SearchKind::Hybrid { embedder_name, embedder, quantized, semantic_ratio: _ } => {
             if let Some(q) = &query.q {
                 search.query(q);
             }
             // will be embedded in hybrid search if necessary
-            search.semantic(embedder_name.clone(), embedder.clone(), query.vector.clone());
+            search.semantic(
+                embedder_name.clone(),
+                embedder.clone(),
+                *quantized,
+                query.vector.clone(),
+            );
         }
     }
 
@@ -821,7 +864,7 @@ fn prepare_search<'t>(
     search.limit(limit);
 
     if let Some(ref filter) = query.filter {
-        if let Some(facets) = parse_filter(filter)? {
+        if let Some(facets) = parse_filter(filter, Code::InvalidSearchFilter, features)? {
             search.filter(facets);
         }
     }
@@ -837,6 +880,10 @@ fn prepare_search<'t>(
         search.sort_criteria(sort);
     }
 
+    if let Some(ref locales) = query.locales {
+        search.locales(locales.iter().copied().map(Into::into).collect());
+    }
+
     Ok((search, is_finite_pagination, max_total_hits, offset))
 }
 
@@ -845,7 +892,8 @@ pub fn perform_search(
     query: SearchQuery,
     search_kind: SearchKind,
     retrieve_vectors: RetrieveVectors,
-) -> Result<SearchResult, MeilisearchHttpError> {
+    features: RoFeatures,
+) -> Result<SearchResult, ResponseError> {
     let before_search = Instant::now();
     let rtxn = index.read_txn()?;
     let time_budget = match index.search_cutoff(&rtxn)? {
@@ -854,7 +902,7 @@ pub fn perform_search(
     };
 
     let (search, is_finite_pagination, max_total_hits, offset) =
-        prepare_search(index, &rtxn, &query, &search_kind, time_budget)?;
+        prepare_search(index, &rtxn, &query, &search_kind, time_budget, features)?;
 
     let (
         milli::SearchResult {
@@ -887,6 +935,7 @@ pub fn perform_search(
         highlight_pre_tag,
         highlight_post_tag,
         crop_marker,
+        locales,
         // already used in prepare_search
         vector: _,
         hybrid: _,
@@ -911,6 +960,7 @@ pub fn perform_search(
         sort,
         show_ranking_score,
         show_ranking_score_details,
+        locales: locales.map(|l| l.iter().copied().map(Into::into).collect()),
     };
 
     let documents = make_hits(
@@ -939,39 +989,13 @@ pub fn perform_search(
         HitsInfo::OffsetLimit { limit, offset, estimated_total_hits: number_of_hits }
     };
 
-    let (facet_distribution, facet_stats) = match facets {
-        Some(ref fields) => {
-            let mut facet_distribution = index.facets_distribution(&rtxn);
-
-            let max_values_by_facet = index
-                .max_values_per_facet(&rtxn)
-                .map_err(milli::Error::from)?
-                .map(|x| x as usize)
-                .unwrap_or(DEFAULT_VALUES_PER_FACET);
-            facet_distribution.max_values_per_facet(max_values_by_facet);
-
-            let sort_facet_values_by =
-                index.sort_facet_values_by(&rtxn).map_err(milli::Error::from)?;
-
-            if fields.iter().all(|f| f != "*") {
-                let fields: Vec<_> =
-                    fields.iter().map(|n| (n, sort_facet_values_by.get(n))).collect();
-                facet_distribution.facets(fields);
-            }
-
-            let distribution = facet_distribution
-                .candidates(candidates)
-                .default_order_by(sort_facet_values_by.get("*"))
-                .execute()?;
-            let stats = facet_distribution.compute_stats()?;
-            (Some(distribution), Some(stats))
-        }
-        None => (None, None),
-    };
-
-    let facet_stats = facet_stats.map(|stats| {
-        stats.into_iter().map(|(k, (min, max))| (k, FacetStats { min, max })).collect()
-    });
+    let (facet_distribution, facet_stats) = facets
+        .map(move |facets| {
+            compute_facet_distribution_stats(&facets, index, &rtxn, candidates, Route::Search)
+        })
+        .transpose()?
+        .map(|ComputedFacets { distribution, stats }| (distribution, stats))
+        .unzip();
 
     let result = SearchResult {
         hits: documents,
@@ -987,6 +1011,61 @@ pub fn perform_search(
     Ok(result)
 }
 
+#[derive(Debug, Clone, Default, Serialize)]
+pub struct ComputedFacets {
+    pub distribution: BTreeMap<String, IndexMap<String, u64>>,
+    pub stats: BTreeMap<String, FacetStats>,
+}
+
+enum Route {
+    Search,
+    MultiSearch,
+}
+
+fn compute_facet_distribution_stats<S: AsRef<str>>(
+    facets: &[S],
+    index: &Index,
+    rtxn: &RoTxn,
+    candidates: roaring::RoaringBitmap,
+    route: Route,
+) -> Result<ComputedFacets, ResponseError> {
+    let mut facet_distribution = index.facets_distribution(rtxn);
+
+    let max_values_by_facet = index
+        .max_values_per_facet(rtxn)
+        .map_err(milli::Error::from)?
+        .map(|x| x as usize)
+        .unwrap_or(DEFAULT_VALUES_PER_FACET);
+
+    facet_distribution.max_values_per_facet(max_values_by_facet);
+
+    let sort_facet_values_by = index.sort_facet_values_by(rtxn).map_err(milli::Error::from)?;
+
+    // add specific facet if there is no placeholder
+    if facets.iter().all(|f| f.as_ref() != "*") {
+        let fields: Vec<_> =
+            facets.iter().map(|n| (n, sort_facet_values_by.get(n.as_ref()))).collect();
+        facet_distribution.facets(fields);
+    }
+
+    let distribution = facet_distribution
+        .candidates(candidates)
+        .default_order_by(sort_facet_values_by.get("*"))
+        .execute()
+        .map_err(|error| match (error, route) {
+            (
+                error @ milli::Error::UserError(milli::UserError::InvalidFacetsDistribution {
+                    ..
+                }),
+                Route::MultiSearch,
+            ) => ResponseError::from_msg(error.to_string(), Code::InvalidMultiSearchFacets),
+            (error, _) => error.into(),
+        })?;
+    let stats = facet_distribution.compute_stats()?;
+    let stats = stats.into_iter().map(|(k, (min, max))| (k, FacetStats { min, max })).collect();
+    Ok(ComputedFacets { distribution, stats })
+}
+
 pub fn search_from_kind(
     search_kind: SearchKind,
     search: milli::Search<'_>,
@@ -1016,6 +1095,7 @@ struct AttributesFormat {
     sort: Option<Vec<String>>,
     show_ranking_score: bool,
     show_ranking_score_details: bool,
+    locales: Option<Vec<Language>>,
 }
 
 #[derive(Debug, Clone, Copy, PartialEq, Eq)]
@@ -1063,19 +1143,16 @@ struct HitMaker<'a> {
     show_ranking_score_details: bool,
     sort: Option<Vec<String>>,
     show_matches_position: bool,
+    locales: Option<Vec<Language>>,
 }
 
 impl<'a> HitMaker<'a> {
     pub fn tokenizer<'b>(
-        script_lang_map: &'b HashMap<milli::tokenizer::Script, Vec<milli::tokenizer::Language>>,
         dictionary: Option<&'b [&'b str]>,
         separators: Option<&'b [&'b str]>,
     ) -> milli::tokenizer::Tokenizer<'b> {
         let mut tokenizer_builder = TokenizerBuilder::default();
         tokenizer_builder.create_char_map(true);
-        if !script_lang_map.is_empty() {
-            tokenizer_builder.allow_list(script_lang_map);
-        }
 
         if let Some(separators) = separators {
             tokenizer_builder.separators(separators);
@@ -1188,6 +1265,7 @@ impl<'a> HitMaker<'a> {
             show_ranking_score_details: format.show_ranking_score_details,
             show_matches_position: format.show_matches_position,
             sort: format.sort,
+            locales: format.locales,
         })
     }
 
@@ -1243,6 +1321,9 @@ impl<'a> HitMaker<'a> {
             document.insert("_vectors".into(), vectors.into());
         }
 
+        let localized_attributes =
+            self.index.localized_attributes_rules(self.rtxn)?.unwrap_or_default();
+
         let (matches_position, formatted) = format_fields(
             &displayed_document,
             &self.fields_ids_map,
@@ -1250,6 +1331,8 @@ impl<'a> HitMaker<'a> {
             &self.formatted_options,
             self.show_matches_position,
             &self.displayed_ids,
+            self.locales.as_deref(),
+            &localized_attributes,
         )?;
 
         if let Some(sort) = self.sort.as_ref() {
@@ -1282,8 +1365,6 @@ fn make_hits<'a>(
 ) -> Result<Vec<SearchHit>, MeilisearchHttpError> {
     let mut documents = Vec::new();
 
-    let script_lang_map = index.script_language(rtxn)?;
-
     let dictionary = index.dictionary(rtxn)?;
     let dictionary: Option<Vec<_>> =
         dictionary.as_ref().map(|x| x.iter().map(String::as_str).collect());
@@ -1291,8 +1372,7 @@ fn make_hits<'a>(
     let separators: Option<Vec<_>> =
         separators.as_ref().map(|x| x.iter().map(String::as_str).collect());
 
-    let tokenizer =
-        HitMaker::tokenizer(&script_lang_map, dictionary.as_deref(), separators.as_deref());
+    let tokenizer = HitMaker::tokenizer(dictionary.as_deref(), separators.as_deref());
 
     let formatter_builder = HitMaker::formatter_builder(matching_words, tokenizer);
 
@@ -1310,7 +1390,9 @@ pub fn perform_facet_search(
     facet_query: Option<String>,
     facet_name: String,
     search_kind: SearchKind,
-) -> Result<FacetSearchResult, MeilisearchHttpError> {
+    features: RoFeatures,
+    locales: Option<Vec<Language>>,
+) -> Result<FacetSearchResult, ResponseError> {
     let before_search = Instant::now();
     let rtxn = index.read_txn()?;
     let time_budget = match index.search_cutoff(&rtxn)? {
@@ -1318,7 +1400,22 @@ pub fn perform_facet_search(
         None => TimeBudget::default(),
     };
 
-    let (search, _, _, _) = prepare_search(index, &rtxn, &search_query, &search_kind, time_budget)?;
+    // In the faceted search context, we want to use the intersection between the locales provided by the user
+    // and the locales of the facet string.
+    // If the facet string is not localized, we **ignore** the locales provided by the user because the facet data has no locale.
+    // If the user does not provide locales, we use the locales of the facet string.
+    let localized_attributes = index.localized_attributes_rules(&rtxn)?.unwrap_or_default();
+    let localized_attributes_locales =
+        localized_attributes.into_iter().find(|attr| attr.match_str(&facet_name));
+    let locales = localized_attributes_locales.map(|attr| {
+        attr.locales
+            .into_iter()
+            .filter(|locale| locales.as_ref().map_or(true, |locales| locales.contains(locale)))
+            .collect()
+    });
+
+    let (search, _, _, _) =
+        prepare_search(index, &rtxn, &search_query, &search_kind, time_budget, features)?;
     let mut facet_search = SearchForFacetValues::new(
         facet_name,
         search,
@@ -1331,6 +1428,10 @@ pub fn perform_facet_search(
         facet_search.max_values(max_facets as usize);
     }
 
+    if let Some(locales) = locales {
+        facet_search.locales(locales);
+    }
+
     Ok(FacetSearchResult {
         facet_hits: facet_search.execute()?,
         facet_query,
@@ -1343,7 +1444,9 @@ pub fn perform_similar(
     query: SimilarQuery,
     embedder_name: String,
     embedder: Arc<Embedder>,
+    quantized: bool,
     retrieve_vectors: RetrieveVectors,
+    features: RoFeatures,
 ) -> Result<SimilarResult, ResponseError> {
     let before_search = Instant::now();
     let rtxn = index.read_txn()?;
@@ -1370,14 +1473,19 @@ pub fn perform_similar(
         ));
     };
 
-    let mut similar =
-        milli::Similar::new(internal_id, offset, limit, index, &rtxn, embedder_name, embedder);
+    let mut similar = milli::Similar::new(
+        internal_id,
+        offset,
+        limit,
+        index,
+        &rtxn,
+        embedder_name,
+        embedder,
+        quantized,
+    );
 
     if let Some(ref filter) = query.filter {
-        if let Some(facets) = parse_filter(filter)
-            // inject InvalidSimilarFilter code
-            .map_err(|e| ResponseError::from_msg(e.to_string(), Code::InvalidSimilarFilter))?
-        {
+        if let Some(facets) = parse_filter(filter, Code::InvalidSimilarFilter, features)? {
             similar.filter(facets);
         }
     }
@@ -1413,6 +1521,7 @@ pub fn perform_similar(
         sort: None,
         show_ranking_score,
         show_ranking_score_details,
+        locales: None,
     };
 
     let hits = make_hits(
@@ -1594,6 +1703,7 @@ fn make_document(
     Ok(document)
 }
 
+#[allow(clippy::too_many_arguments)]
 fn format_fields(
     document: &Document,
     field_ids_map: &FieldsIdsMap,
@@ -1601,6 +1711,8 @@ fn format_fields(
     formatted_options: &BTreeMap<FieldId, FormatOptions>,
     compute_matches: bool,
     displayable_ids: &BTreeSet<FieldId>,
+    locales: Option<&[Language]>,
+    localized_attributes: &[LocalizedAttributesRule],
 ) -> Result<(Option<MatchesPosition>, Document), MeilisearchHttpError> {
     let mut matches_position = compute_matches.then(BTreeMap::new);
     let mut document = document.clone();
@@ -1633,7 +1745,22 @@ fn format_fields(
             .reduce(|acc, option| acc.merge(option));
         let mut infos = Vec::new();
 
-        *value = format_value(std::mem::take(value), builder, format, &mut infos, compute_matches);
+        // if no locales has been provided, we try to find the locales in the localized_attributes.
+        let locales = locales.or_else(|| {
+            localized_attributes
+                .iter()
+                .find(|rule| rule.match_str(key))
+                .map(LocalizedAttributesRule::locales)
+        });
+
+        *value = format_value(
+            std::mem::take(value),
+            builder,
+            format,
+            &mut infos,
+            compute_matches,
+            locales,
+        );
 
         if let Some(matches) = matches_position.as_mut() {
             if !infos.is_empty() {
@@ -1658,10 +1785,11 @@ fn format_value(
     format_options: Option<FormatOptions>,
     infos: &mut Vec<MatchBounds>,
     compute_matches: bool,
+    locales: Option<&[Language]>,
 ) -> Value {
     match value {
         Value::String(old_string) => {
-            let mut matcher = builder.build(&old_string);
+            let mut matcher = builder.build(&old_string, locales);
             if compute_matches {
                 let matches = matcher.matches();
                 infos.extend_from_slice(&matches[..]);
@@ -1688,6 +1816,7 @@ fn format_value(
                         }),
                         infos,
                         compute_matches,
+                        locales,
                     )
                 })
                 .collect(),
@@ -1707,6 +1836,7 @@ fn format_value(
                             }),
                             infos,
                             compute_matches,
+                            locales,
                         ),
                     )
                 })
@@ -1715,7 +1845,7 @@ fn format_value(
         Value::Number(number) => {
             let s = number.to_string();
 
-            let mut matcher = builder.build(&s);
+            let mut matcher = builder.build(&s, locales);
             if compute_matches {
                 let matches = matcher.matches();
                 infos.extend_from_slice(&matches[..]);
@@ -1733,15 +1863,33 @@ fn format_value(
     }
 }
 
-pub(crate) fn parse_filter(facets: &Value) -> Result<Option<Filter>, MeilisearchHttpError> {
-    match facets {
-        Value::String(expr) => {
-            let condition = Filter::from_str(expr)?;
-            Ok(condition)
+pub(crate) fn parse_filter(
+    facets: &Value,
+    filter_parsing_error_code: Code,
+    features: RoFeatures,
+) -> Result<Option<Filter>, ResponseError> {
+    let filter = match facets {
+        Value::String(expr) => Filter::from_str(expr).map_err(|e| e.into()),
+        Value::Array(arr) => parse_filter_array(arr).map_err(|e| e.into()),
+        v => Err(MeilisearchHttpError::InvalidExpression(&["String", "Array"], v.clone()).into()),
+    };
+    let filter = filter.map_err(|err: ResponseError| {
+        ResponseError::from_msg(err.to_string(), filter_parsing_error_code)
+    })?;
+
+    if let Some(ref filter) = filter {
+        // If the contains operator is used while the contains filter features is not enabled, errors out
+        if let Some((token, error)) =
+            filter.use_contains_operator().zip(features.check_contains_filter().err())
+        {
+            return Err(ResponseError::from_msg(
+                token.as_external_error(error).to_string(),
+                Code::FeatureNotEnabled,
+            ));
         }
-        Value::Array(arr) => parse_filter_array(arr),
-        v => Err(MeilisearchHttpError::InvalidExpression(&["String", "Array"], v.clone())),
     }
+
+    Ok(filter)
 }
 
 fn parse_filter_array(arr: &[Value]) -> Result<Option<Filter>, MeilisearchHttpError> {

meilisearch/src/search/ranking_rules.rs

@@ -0,0 +1,823 @@
+use std::collections::HashMap;
+use std::fmt::Write;
+
+use itertools::Itertools as _;
+use meilisearch_types::error::{Code, ResponseError};
+use meilisearch_types::milli::{AscDesc, Criterion, Member, TermsMatchingStrategy};
+
+pub struct RankingRules {
+    canonical_criteria: Vec<Criterion>,
+    canonical_sort: Option<Vec<AscDesc>>,
+    canonicalization_actions: Vec<CanonicalizationAction>,
+    source_criteria: Vec<Criterion>,
+    source_sort: Option<Vec<AscDesc>>,
+}
+
+pub enum CanonicalizationAction {
+    PrependedWords {
+        prepended_index: RankingRuleSource,
+    },
+    RemovedDuplicate {
+        earlier_occurrence: RankingRuleSource,
+        removed_occurrence: RankingRuleSource,
+    },
+    RemovedWords {
+        reason: RemoveWords,
+        removed_occurrence: RankingRuleSource,
+    },
+    RemovedPlaceholder {
+        removed_occurrence: RankingRuleSource,
+    },
+    TruncatedVector {
+        vector_rule: RankingRuleSource,
+        truncated_from: RankingRuleSource,
+    },
+    RemovedVector {
+        vector_rule: RankingRuleSource,
+        removed_occurrence: RankingRuleSource,
+    },
+    RemovedSort {
+        removed_occurrence: RankingRuleSource,
+    },
+}
+
+pub enum RemoveWords {
+    WasPrepended,
+    MatchingStrategyAll,
+}
+
+impl std::fmt::Display for RemoveWords {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        let reason = match self {
+            RemoveWords::WasPrepended => "it was previously prepended",
+            RemoveWords::MatchingStrategyAll => "`query.matchingWords` is set to `all`",
+        };
+        f.write_str(reason)
+    }
+}
+
+pub enum CanonicalizationKind {
+    Placeholder,
+    Keyword,
+    Vector,
+}
+
+pub struct CompatibilityError {
+    previous: RankingRule,
+    current: RankingRule,
+}
+impl CompatibilityError {
+    pub(crate) fn to_response_error(
+        &self,
+        ranking_rules: &RankingRules,
+        previous_ranking_rules: &RankingRules,
+        query_index: usize,
+        previous_query_index: usize,
+        index_uid: &str,
+        previous_index_uid: &str,
+    ) -> meilisearch_types::error::ResponseError {
+        let rule = self.current.as_string(
+            &ranking_rules.canonical_criteria,
+            &ranking_rules.canonical_sort,
+            query_index,
+            index_uid,
+        );
+        let previous_rule = self.previous.as_string(
+            &previous_ranking_rules.canonical_criteria,
+            &previous_ranking_rules.canonical_sort,
+            previous_query_index,
+            previous_index_uid,
+        );
+
+        let canonicalization_actions = ranking_rules.canonicalization_notes();
+        let previous_canonicalization_actions = previous_ranking_rules.canonicalization_notes();
+
+        let mut msg = String::new();
+        let reason = self.reason();
+        let _ = writeln!(
+            &mut msg,
+            "The results of queries #{previous_query_index} and #{query_index} are incompatible: "
+        );
+        let _ = writeln!(&mut msg, "  1. {previous_rule}");
+        let _ = writeln!(&mut msg, "  2. {rule}");
+        let _ = writeln!(&mut msg, "  - {reason}");
+
+        if !previous_canonicalization_actions.is_empty() {
+            let _ = write!(&mut msg, "  - note: The ranking rules of query #{previous_query_index} were modified during canonicalization:\n{previous_canonicalization_actions}");
+        }
+
+        if !canonicalization_actions.is_empty() {
+            let _ = write!(&mut msg, "  - note: The ranking rules of query #{query_index} were modified during canonicalization:\n{canonicalization_actions}");
+        }
+
+        ResponseError::from_msg(msg, Code::InvalidMultiSearchQueryRankingRules)
+    }
+    pub fn reason(&self) -> &'static str {
+        match (self.previous.kind, self.current.kind) {
+            (RankingRuleKind::Relevancy, RankingRuleKind::AscendingSort)
+            | (RankingRuleKind::Relevancy, RankingRuleKind::DescendingSort)
+            | (RankingRuleKind::AscendingSort, RankingRuleKind::Relevancy)
+            | (RankingRuleKind::DescendingSort, RankingRuleKind::Relevancy) => {
+                "cannot compare a relevancy rule with a sort rule"
+            }
+
+            (RankingRuleKind::Relevancy, RankingRuleKind::AscendingGeoSort)
+            | (RankingRuleKind::Relevancy, RankingRuleKind::DescendingGeoSort)
+            | (RankingRuleKind::AscendingGeoSort, RankingRuleKind::Relevancy)
+            | (RankingRuleKind::DescendingGeoSort, RankingRuleKind::Relevancy) => {
+                "cannot compare a relevancy rule with a geosort rule"
+            }
+
+            (RankingRuleKind::AscendingSort, RankingRuleKind::DescendingSort)
+            | (RankingRuleKind::DescendingSort, RankingRuleKind::AscendingSort) => {
+                "cannot compare two sort rules in opposite directions"
+            }
+
+            (RankingRuleKind::AscendingSort, RankingRuleKind::AscendingGeoSort)
+            | (RankingRuleKind::AscendingSort, RankingRuleKind::DescendingGeoSort)
+            | (RankingRuleKind::DescendingSort, RankingRuleKind::AscendingGeoSort)
+            | (RankingRuleKind::DescendingSort, RankingRuleKind::DescendingGeoSort)
+            | (RankingRuleKind::AscendingGeoSort, RankingRuleKind::AscendingSort)
+            | (RankingRuleKind::AscendingGeoSort, RankingRuleKind::DescendingSort)
+            | (RankingRuleKind::DescendingGeoSort, RankingRuleKind::AscendingSort)
+            | (RankingRuleKind::DescendingGeoSort, RankingRuleKind::DescendingSort) => {
+                "cannot compare a sort rule with a geosort rule"
+            }
+
+            (RankingRuleKind::AscendingGeoSort, RankingRuleKind::DescendingGeoSort)
+            | (RankingRuleKind::DescendingGeoSort, RankingRuleKind::AscendingGeoSort) => {
+                "cannot compare two geosort rules in opposite directions"
+            }
+            (RankingRuleKind::Relevancy, RankingRuleKind::Relevancy)
+            | (RankingRuleKind::AscendingSort, RankingRuleKind::AscendingSort)
+            | (RankingRuleKind::DescendingSort, RankingRuleKind::DescendingSort)
+            | (RankingRuleKind::AscendingGeoSort, RankingRuleKind::AscendingGeoSort)
+            | (RankingRuleKind::DescendingGeoSort, RankingRuleKind::DescendingGeoSort) => {
+                "internal error, comparison should be possible"
+            }
+        }
+    }
+}
+
+impl RankingRules {
+    pub fn new(
+        criteria: Vec<Criterion>,
+        sort: Option<Vec<AscDesc>>,
+        terms_matching_strategy: TermsMatchingStrategy,
+        canonicalization_kind: CanonicalizationKind,
+    ) -> Self {
+        let (canonical_criteria, canonical_sort, canonicalization_actions) =
+            Self::canonicalize(&criteria, &sort, terms_matching_strategy, canonicalization_kind);
+        Self {
+            canonical_criteria,
+            canonical_sort,
+            canonicalization_actions,
+            source_criteria: criteria,
+            source_sort: sort,
+        }
+    }
+
+    fn canonicalize(
+        criteria: &[Criterion],
+        sort: &Option<Vec<AscDesc>>,
+        terms_matching_strategy: TermsMatchingStrategy,
+        canonicalization_kind: CanonicalizationKind,
+    ) -> (Vec<Criterion>, Option<Vec<AscDesc>>, Vec<CanonicalizationAction>) {
+        match canonicalization_kind {
+            CanonicalizationKind::Placeholder => Self::canonicalize_placeholder(criteria, sort),
+            CanonicalizationKind::Keyword => {
+                Self::canonicalize_keyword(criteria, sort, terms_matching_strategy)
+            }
+            CanonicalizationKind::Vector => Self::canonicalize_vector(criteria, sort),
+        }
+    }
+
+    fn canonicalize_placeholder(
+        criteria: &[Criterion],
+        sort_query: &Option<Vec<AscDesc>>,
+    ) -> (Vec<Criterion>, Option<Vec<AscDesc>>, Vec<CanonicalizationAction>) {
+        let mut sort = None;
+
+        let mut sorted_fields = HashMap::new();
+        let mut canonicalization_actions = Vec::new();
+        let mut canonical_criteria = Vec::new();
+        let mut canonical_sort = None;
+
+        for (criterion_index, criterion) in criteria.iter().enumerate() {
+            match criterion.clone() {
+                Criterion::Words
+                | Criterion::Typo
+                | Criterion::Proximity
+                | Criterion::Attribute
+                | Criterion::Exactness => {
+                    canonicalization_actions.push(CanonicalizationAction::RemovedPlaceholder {
+                        removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                    })
+                }
+
+                Criterion::Sort => {
+                    if let Some(previous_index) = sort {
+                        canonicalization_actions.push(CanonicalizationAction::RemovedDuplicate {
+                            earlier_occurrence: RankingRuleSource::Criterion(previous_index),
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        });
+                    } else if let Some(sort_query) = sort_query {
+                        sort = Some(criterion_index);
+                        canonical_criteria.push(criterion.clone());
+                        canonical_sort = Some(canonicalize_sort(
+                            &mut sorted_fields,
+                            sort_query.as_slice(),
+                            criterion_index,
+                            &mut canonicalization_actions,
+                        ));
+                    } else {
+                        canonicalization_actions.push(CanonicalizationAction::RemovedSort {
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        })
+                    }
+                }
+                Criterion::Asc(s) | Criterion::Desc(s) => match sorted_fields.entry(s) {
+                    std::collections::hash_map::Entry::Occupied(entry) => canonicalization_actions
+                        .push(CanonicalizationAction::RemovedDuplicate {
+                            earlier_occurrence: *entry.get(),
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        }),
+                    std::collections::hash_map::Entry::Vacant(entry) => {
+                        entry.insert(RankingRuleSource::Criterion(criterion_index));
+                        canonical_criteria.push(criterion.clone())
+                    }
+                },
+            }
+        }
+
+        (canonical_criteria, canonical_sort, canonicalization_actions)
+    }
+
+    fn canonicalize_vector(
+        criteria: &[Criterion],
+        sort_query: &Option<Vec<AscDesc>>,
+    ) -> (Vec<Criterion>, Option<Vec<AscDesc>>, Vec<CanonicalizationAction>) {
+        let mut sort = None;
+
+        let mut sorted_fields = HashMap::new();
+        let mut canonicalization_actions = Vec::new();
+        let mut canonical_criteria = Vec::new();
+        let mut canonical_sort = None;
+
+        let mut vector = None;
+
+        'criteria: for (criterion_index, criterion) in criteria.iter().enumerate() {
+            match criterion.clone() {
+                Criterion::Words
+                | Criterion::Typo
+                | Criterion::Proximity
+                | Criterion::Attribute
+                | Criterion::Exactness => match vector {
+                    Some(previous_occurrence) => {
+                        if sorted_fields.is_empty() {
+                            canonicalization_actions.push(CanonicalizationAction::RemovedVector {
+                                vector_rule: RankingRuleSource::Criterion(previous_occurrence),
+                                removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                            });
+                        } else {
+                            canonicalization_actions.push(
+                                CanonicalizationAction::TruncatedVector {
+                                    vector_rule: RankingRuleSource::Criterion(previous_occurrence),
+                                    truncated_from: RankingRuleSource::Criterion(criterion_index),
+                                },
+                            );
+                            break 'criteria;
+                        }
+                    }
+                    None => {
+                        canonical_criteria.push(criterion.clone());
+                        vector = Some(criterion_index);
+                    }
+                },
+
+                Criterion::Sort => {
+                    if let Some(previous_index) = sort {
+                        canonicalization_actions.push(CanonicalizationAction::RemovedDuplicate {
+                            earlier_occurrence: RankingRuleSource::Criterion(previous_index),
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        });
+                    } else if let Some(sort_query) = sort_query {
+                        sort = Some(criterion_index);
+                        canonical_criteria.push(criterion.clone());
+                        canonical_sort = Some(canonicalize_sort(
+                            &mut sorted_fields,
+                            sort_query.as_slice(),
+                            criterion_index,
+                            &mut canonicalization_actions,
+                        ));
+                    } else {
+                        canonicalization_actions.push(CanonicalizationAction::RemovedSort {
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        })
+                    }
+                }
+                Criterion::Asc(s) | Criterion::Desc(s) => match sorted_fields.entry(s) {
+                    std::collections::hash_map::Entry::Occupied(entry) => canonicalization_actions
+                        .push(CanonicalizationAction::RemovedDuplicate {
+                            earlier_occurrence: *entry.get(),
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        }),
+                    std::collections::hash_map::Entry::Vacant(entry) => {
+                        entry.insert(RankingRuleSource::Criterion(criterion_index));
+                        canonical_criteria.push(criterion.clone())
+                    }
+                },
+            }
+        }
+
+        (canonical_criteria, canonical_sort, canonicalization_actions)
+    }
+
+    fn canonicalize_keyword(
+        criteria: &[Criterion],
+        sort_query: &Option<Vec<AscDesc>>,
+        terms_matching_strategy: TermsMatchingStrategy,
+    ) -> (Vec<Criterion>, Option<Vec<AscDesc>>, Vec<CanonicalizationAction>) {
+        let mut words = None;
+        let mut typo = None;
+        let mut proximity = None;
+        let mut sort = None;
+        let mut attribute = None;
+        let mut exactness = None;
+        let mut sorted_fields = HashMap::new();
+
+        let mut canonical_criteria = Vec::new();
+        let mut canonical_sort = None;
+
+        let mut canonicalization_actions = Vec::new();
+
+        for (criterion_index, criterion) in criteria.iter().enumerate() {
+            let criterion = criterion.clone();
+            match criterion.clone() {
+                Criterion::Words => {
+                    if let TermsMatchingStrategy::All = terms_matching_strategy {
+                        canonicalization_actions.push(CanonicalizationAction::RemovedWords {
+                            reason: RemoveWords::MatchingStrategyAll,
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        });
+                        continue;
+                    }
+                    if let Some(maybe_previous_index) = words {
+                        if let Some(previous_index) = maybe_previous_index {
+                            canonicalization_actions.push(
+                                CanonicalizationAction::RemovedDuplicate {
+                                    earlier_occurrence: RankingRuleSource::Criterion(
+                                        previous_index,
+                                    ),
+                                    removed_occurrence: RankingRuleSource::Criterion(
+                                        criterion_index,
+                                    ),
+                                },
+                            );
+                            continue;
+                        }
+                        canonicalization_actions.push(CanonicalizationAction::RemovedWords {
+                            reason: RemoveWords::WasPrepended,
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        })
+                    }
+                    words = Some(Some(criterion_index));
+                    canonical_criteria.push(criterion);
+                }
+                Criterion::Typo => {
+                    canonicalize_criterion(
+                        criterion,
+                        criterion_index,
+                        terms_matching_strategy,
+                        &mut words,
+                        &mut canonicalization_actions,
+                        &mut canonical_criteria,
+                        &mut typo,
+                    );
+                }
+                Criterion::Proximity => {
+                    canonicalize_criterion(
+                        criterion,
+                        criterion_index,
+                        terms_matching_strategy,
+                        &mut words,
+                        &mut canonicalization_actions,
+                        &mut canonical_criteria,
+                        &mut proximity,
+                    );
+                }
+                Criterion::Attribute => {
+                    canonicalize_criterion(
+                        criterion,
+                        criterion_index,
+                        terms_matching_strategy,
+                        &mut words,
+                        &mut canonicalization_actions,
+                        &mut canonical_criteria,
+                        &mut attribute,
+                    );
+                }
+                Criterion::Exactness => {
+                    canonicalize_criterion(
+                        criterion,
+                        criterion_index,
+                        terms_matching_strategy,
+                        &mut words,
+                        &mut canonicalization_actions,
+                        &mut canonical_criteria,
+                        &mut exactness,
+                    );
+                }
+
+                Criterion::Sort => {
+                    if let Some(previous_index) = sort {
+                        canonicalization_actions.push(CanonicalizationAction::RemovedDuplicate {
+                            earlier_occurrence: RankingRuleSource::Criterion(previous_index),
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        });
+                    } else if let Some(sort_query) = sort_query {
+                        sort = Some(criterion_index);
+                        canonical_criteria.push(criterion);
+                        canonical_sort = Some(canonicalize_sort(
+                            &mut sorted_fields,
+                            sort_query.as_slice(),
+                            criterion_index,
+                            &mut canonicalization_actions,
+                        ));
+                    } else {
+                        canonicalization_actions.push(CanonicalizationAction::RemovedSort {
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        })
+                    }
+                }
+                Criterion::Asc(s) | Criterion::Desc(s) => match sorted_fields.entry(s) {
+                    std::collections::hash_map::Entry::Occupied(entry) => canonicalization_actions
+                        .push(CanonicalizationAction::RemovedDuplicate {
+                            earlier_occurrence: *entry.get(),
+                            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+                        }),
+                    std::collections::hash_map::Entry::Vacant(entry) => {
+                        entry.insert(RankingRuleSource::Criterion(criterion_index));
+                        canonical_criteria.push(criterion)
+                    }
+                },
+            }
+        }
+
+        (canonical_criteria, canonical_sort, canonicalization_actions)
+    }
+
+    pub fn is_compatible_with(&self, previous: &Self) -> Result<(), CompatibilityError> {
+        for (current, previous) in self.coalesce_iterator().zip(previous.coalesce_iterator()) {
+            if current.kind != previous.kind {
+                return Err(CompatibilityError { current, previous });
+            }
+        }
+        Ok(())
+    }
+
+    pub fn constraint_count(&self) -> usize {
+        self.coalesce_iterator().count()
+    }
+
+    fn coalesce_iterator(&self) -> impl Iterator<Item = RankingRule> + '_ {
+        self.canonical_criteria
+            .iter()
+            .enumerate()
+            .flat_map(|(criterion_index, criterion)| {
+                RankingRule::from_criterion(criterion_index, criterion, &self.canonical_sort)
+            })
+            .coalesce(
+                |previous @ RankingRule { source: previous_source, kind: previous_kind },
+                 current @ RankingRule { source, kind }| {
+                    match (previous_kind, kind) {
+                        (RankingRuleKind::Relevancy, RankingRuleKind::Relevancy) => {
+                            let merged_source = match (previous_source, source) {
+                                (
+                                    RankingRuleSource::Criterion(previous),
+                                    RankingRuleSource::Criterion(current),
+                                ) => RankingRuleSource::CoalescedCriteria(previous, current),
+                                (
+                                    RankingRuleSource::CoalescedCriteria(begin, _end),
+                                    RankingRuleSource::Criterion(current),
+                                ) => RankingRuleSource::CoalescedCriteria(begin, current),
+                                (_previous, current) => current,
+                            };
+                            Ok(RankingRule { source: merged_source, kind })
+                        }
+                        _ => Err((previous, current)),
+                    }
+                },
+            )
+    }
+
+    fn canonicalization_notes(&self) -> String {
+        use CanonicalizationAction::*;
+        let mut notes = String::new();
+        for (index, action) in self.canonicalization_actions.iter().enumerate() {
+            let index = index + 1;
+            let _ = match action {
+                PrependedWords { prepended_index } => writeln!(
+                    &mut notes,
+                    "    {index}. Prepended rule `words` before first relevancy rule `{}` at position {}",
+                    prepended_index.rule_name(&self.source_criteria, &self.source_sort),
+                    prepended_index.rule_position()
+                ),
+                RemovedDuplicate { earlier_occurrence, removed_occurrence } => writeln!(
+                    &mut notes,
+                    "    {index}. Removed duplicate rule `{}` at position {} as it already appears at position {}",
+                    earlier_occurrence.rule_name(&self.source_criteria, &self.source_sort),
+                    removed_occurrence.rule_position(),
+                    earlier_occurrence.rule_position(),
+                ),
+                RemovedWords { reason, removed_occurrence } => writeln!(
+                    &mut notes,
+                    "    {index}. Removed rule `words` at position {} because {reason}",
+                    removed_occurrence.rule_position()
+                ),
+                RemovedPlaceholder { removed_occurrence } => writeln!(
+                    &mut notes,
+                    "    {index}. Removed relevancy rule `{}` at position {} because the query is a placeholder search (`q`: \"\")",
+                    removed_occurrence.rule_name(&self.source_criteria, &self.source_sort),
+                    removed_occurrence.rule_position()
+                ),
+                TruncatedVector { vector_rule, truncated_from } => writeln!(
+                    &mut notes,
+                    "    {index}. Truncated relevancy rule `{}` at position {} and later rules because the query is a vector search and `vector` was inserted at position {}",
+                    truncated_from.rule_name(&self.source_criteria, &self.source_sort),
+                    truncated_from.rule_position(),
+                    vector_rule.rule_position(),
+                ),
+                RemovedVector { vector_rule, removed_occurrence } => writeln!(
+                    &mut notes,
+                    "    {index}. Removed relevancy rule `{}` at position {} because the query is a vector search and `vector` was already inserted at position {}",
+                    removed_occurrence.rule_name(&self.source_criteria, &self.source_sort),
+                    removed_occurrence.rule_position(),
+                    vector_rule.rule_position(),
+                ),
+                RemovedSort { removed_occurrence } => writeln!(
+                    &mut notes,
+                    "   {index}. Removed rule `sort` at position {} because `query.sort` is empty",
+removed_occurrence.rule_position()
+                ),
+            };
+        }
+        notes
+    }
+}
+
+fn canonicalize_sort(
+    sorted_fields: &mut HashMap<String, RankingRuleSource>,
+    sort_query: &[AscDesc],
+    criterion_index: usize,
+    canonicalization_actions: &mut Vec<CanonicalizationAction>,
+) -> Vec<AscDesc> {
+    let mut geo_sorted = None;
+    let mut canonical_sort = Vec::new();
+    for (sort_index, asc_desc) in sort_query.iter().enumerate() {
+        let source = RankingRuleSource::Sort { criterion_index, sort_index };
+        let asc_desc = asc_desc.clone();
+        match asc_desc.clone() {
+            AscDesc::Asc(Member::Field(s)) | AscDesc::Desc(Member::Field(s)) => {
+                match sorted_fields.entry(s) {
+                    std::collections::hash_map::Entry::Occupied(entry) => canonicalization_actions
+                        .push(CanonicalizationAction::RemovedDuplicate {
+                            earlier_occurrence: *entry.get(),
+                            removed_occurrence: source,
+                        }),
+                    std::collections::hash_map::Entry::Vacant(entry) => {
+                        entry.insert(source);
+                        canonical_sort.push(asc_desc);
+                    }
+                }
+            }
+            AscDesc::Asc(Member::Geo(_)) | AscDesc::Desc(Member::Geo(_)) => match geo_sorted {
+                Some(earlier_sort_index) => {
+                    canonicalization_actions.push(CanonicalizationAction::RemovedDuplicate {
+                        earlier_occurrence: RankingRuleSource::Sort {
+                            criterion_index,
+                            sort_index: earlier_sort_index,
+                        },
+                        removed_occurrence: source,
+                    })
+                }
+                None => {
+                    geo_sorted = Some(sort_index);
+                    canonical_sort.push(asc_desc);
+                }
+            },
+        }
+    }
+    canonical_sort
+}
+
+fn canonicalize_criterion(
+    criterion: Criterion,
+    criterion_index: usize,
+    terms_matching_strategy: TermsMatchingStrategy,
+    words: &mut Option<Option<usize>>,
+    canonicalization_actions: &mut Vec<CanonicalizationAction>,
+    canonical_criteria: &mut Vec<Criterion>,
+    rule: &mut Option<usize>,
+) {
+    *words = match (terms_matching_strategy, words.take()) {
+        (TermsMatchingStrategy::All, words) => words,
+        (_, None) => {
+            // inject words
+            canonicalization_actions.push(CanonicalizationAction::PrependedWords {
+                prepended_index: RankingRuleSource::Criterion(criterion_index),
+            });
+            canonical_criteria.push(Criterion::Words);
+            Some(None)
+        }
+        (_, words) => words,
+    };
+    if let Some(previous_index) = *rule {
+        canonicalization_actions.push(CanonicalizationAction::RemovedDuplicate {
+            earlier_occurrence: RankingRuleSource::Criterion(previous_index),
+            removed_occurrence: RankingRuleSource::Criterion(criterion_index),
+        });
+    } else {
+        *rule = Some(criterion_index);
+        canonical_criteria.push(criterion)
+    }
+}
+
+#[derive(Debug, Clone, Copy, PartialEq, Eq)]
+enum RankingRuleKind {
+    Relevancy,
+    AscendingSort,
+    DescendingSort,
+    AscendingGeoSort,
+    DescendingGeoSort,
+}
+
+#[derive(Debug, Clone, Copy)]
+pub struct RankingRule {
+    source: RankingRuleSource,
+    kind: RankingRuleKind,
+}
+
+#[derive(Debug, Clone, Copy)]
+pub enum RankingRuleSource {
+    Criterion(usize),
+    CoalescedCriteria(usize, usize),
+    Sort { criterion_index: usize, sort_index: usize },
+}
+
+impl RankingRuleSource {
+    fn rule_name(&self, criteria: &[Criterion], sort: &Option<Vec<AscDesc>>) -> String {
+        match self {
+            RankingRuleSource::Criterion(criterion_index) => criteria
+                .get(*criterion_index)
+                .map(|c| c.to_string())
+                .unwrap_or_else(|| "unknown".into()),
+            RankingRuleSource::CoalescedCriteria(begin, end) => {
+                let rules: Vec<_> = criteria
+                    .get(*begin..=*end)
+                    .iter()
+                    .flat_map(|c| c.iter())
+                    .map(|c| c.to_string())
+                    .collect();
+                rules.join(", ")
+            }
+            RankingRuleSource::Sort { criterion_index: _, sort_index } => {
+                match sort.as_deref().and_then(|sort| sort.get(*sort_index)) {
+                    Some(sort) => match sort {
+                        AscDesc::Asc(Member::Field(field_name)) => format!("{field_name}:asc"),
+                        AscDesc::Desc(Member::Field(field_name)) => {
+                            format!("{field_name}:desc")
+                        }
+                        AscDesc::Asc(Member::Geo(_)) => "_geo(..):asc".to_string(),
+                        AscDesc::Desc(Member::Geo(_)) => "_geo(..):desc".to_string(),
+                    },
+                    None => "unknown".into(),
+                }
+            }
+        }
+    }
+
+    fn rule_position(&self) -> String {
+        match self {
+            RankingRuleSource::Criterion(criterion_index) => {
+                format!("#{criterion_index} in ranking rules")
+            }
+            RankingRuleSource::CoalescedCriteria(begin, end) => {
+                format!("#{begin} to #{end} in ranking rules")
+            }
+            RankingRuleSource::Sort { criterion_index, sort_index } => format!(
+                "#{sort_index} in `query.sort` (as `sort` is #{criterion_index} in ranking rules)"
+            ),
+        }
+    }
+}
+
+impl RankingRule {
+    fn from_criterion<'a>(
+        criterion_index: usize,
+        criterion: &'a Criterion,
+        sort: &'a Option<Vec<AscDesc>>,
+    ) -> impl Iterator<Item = Self> + 'a {
+        let kind = match criterion {
+            Criterion::Words
+            | Criterion::Typo
+            | Criterion::Proximity
+            | Criterion::Attribute
+            | Criterion::Exactness => RankingRuleKind::Relevancy,
+            Criterion::Asc(s) if s == "_geo" => RankingRuleKind::AscendingGeoSort,
+
+            Criterion::Asc(_) => RankingRuleKind::AscendingSort,
+            Criterion::Desc(s) if s == "_geo" => RankingRuleKind::DescendingGeoSort,
+
+            Criterion::Desc(_) => RankingRuleKind::DescendingSort,
+            Criterion::Sort => {
+                return either::Right(sort.iter().flatten().enumerate().map(
+                    move |(rule_index, asc_desc)| {
+                        Self::from_asc_desc(asc_desc, criterion_index, rule_index)
+                    },
+                ))
+            }
+        };
+
+        either::Left(std::iter::once(Self {
+            source: RankingRuleSource::Criterion(criterion_index),
+            kind,
+        }))
+    }
+
+    fn from_asc_desc(asc_desc: &AscDesc, sort_index: usize, rule_index_in_sort: usize) -> Self {
+        let kind = match asc_desc {
+            AscDesc::Asc(Member::Field(_)) => RankingRuleKind::AscendingSort,
+            AscDesc::Desc(Member::Field(_)) => RankingRuleKind::DescendingSort,
+            AscDesc::Asc(Member::Geo(_)) => RankingRuleKind::AscendingGeoSort,
+            AscDesc::Desc(Member::Geo(_)) => RankingRuleKind::DescendingGeoSort,
+        };
+        Self {
+            source: RankingRuleSource::Sort {
+                criterion_index: sort_index,
+                sort_index: rule_index_in_sort,
+            },
+            kind,
+        }
+    }
+
+    fn as_string(
+        &self,
+        canonical_criteria: &[Criterion],
+        canonical_sort: &Option<Vec<AscDesc>>,
+        query_index: usize,
+        index_uid: &str,
+    ) -> String {
+        let kind = match self.kind {
+            RankingRuleKind::Relevancy => "relevancy",
+            RankingRuleKind::AscendingSort => "ascending sort",
+            RankingRuleKind::DescendingSort => "descending sort",
+            RankingRuleKind::AscendingGeoSort => "ascending geo sort",
+            RankingRuleKind::DescendingGeoSort => "descending geo sort",
+        };
+        let rules = self.fetch_from_source(canonical_criteria, canonical_sort);
+
+        let source = match self.source {
+            RankingRuleSource::Criterion(criterion_index) => format!("`queries[{query_index}]`, `{index_uid}.rankingRules[{criterion_index}]`"),
+            RankingRuleSource::CoalescedCriteria(begin, end) => format!("`queries[{query_index}]`, `{index_uid}.rankingRules[{begin}..={end}]`"),
+            RankingRuleSource::Sort { criterion_index, sort_index } => format!("`queries[{query_index}].sort[{sort_index}]`, `{index_uid}.rankingRules[{criterion_index}]`"),
+        };
+
+        format!("{source}: {kind} {rules}")
+    }
+
+    fn fetch_from_source(
+        &self,
+        canonical_criteria: &[Criterion],
+        canonical_sort: &Option<Vec<AscDesc>>,
+    ) -> String {
+        let rule_name = match self.source {
+            RankingRuleSource::Criterion(index) => {
+                canonical_criteria.get(index).map(|criterion| criterion.to_string())
+            }
+            RankingRuleSource::CoalescedCriteria(begin, end) => {
+                let rules: Vec<String> = canonical_criteria
+                    .get(begin..=end)
+                    .into_iter()
+                    .flat_map(|criteria| criteria.iter())
+                    .map(|criterion| criterion.to_string())
+                    .collect();
+
+                (!rules.is_empty()).then_some(rules.join(", "))
+            }
+            RankingRuleSource::Sort { criterion_index: _, sort_index } => canonical_sort
+                .as_deref()
+                .and_then(|canonical_sort| canonical_sort.get(sort_index))
+                .and_then(|asc_desc: &AscDesc| match asc_desc {
+                    AscDesc::Asc(Member::Field(s)) | AscDesc::Desc(Member::Field(s)) => {
+                        Some(format!("on field `{s}`"))
+                    }
+                    _ => None,
+                }),
+        };
+
+        let rule_name = rule_name.unwrap_or_else(|| "default".into());
+
+        format!("rule(s) {rule_name}")
+    }
+}

meilisearch/src/search_queue.rs

@@ -18,6 +18,7 @@
 //!                         And should drop the Permit only once you have freed all the RAM consumed by the method.
 
 use std::num::NonZeroUsize;
+use std::time::Duration;
 
 use rand::rngs::StdRng;
 use rand::{Rng, SeedableRng};
@@ -29,16 +30,31 @@ use crate::error::MeilisearchHttpError;
 pub struct SearchQueue {
     sender: mpsc::Sender<oneshot::Sender<Permit>>,
     capacity: usize,
+    /// If we have waited longer than this to get a permit, we should abort the search request entirely.
+    /// The client probably already closed the connection, but we have no way to find out.
+    time_to_abort: Duration,
 }
 
 /// You should only run search requests while holding this permit.
 /// Once it's dropped, a new search request will be able to process.
+/// You should always try to drop the permit yourself calling the `drop` async method on it.
 #[derive(Debug)]
 pub struct Permit {
     sender: mpsc::Sender<()>,
 }
 
+impl Permit {
+    /// Drop the permit giving back on permit to the search queue.
+    pub async fn drop(self) {
+        // if the channel is closed then the whole instance is down
+        let _ = self.sender.send(()).await;
+    }
+}
+
 impl Drop for Permit {
+    /// The implicit drop implementation can still be called in multiple cases:
+    /// - We forgot to call the explicit one somewhere => this should be fixed on our side asap
+    /// - The future is cancelled while running and the permit dropped with it
     fn drop(&mut self) {
         let sender = self.sender.clone();
         // if the channel is closed then the whole instance is down
@@ -53,7 +69,11 @@ impl SearchQueue {
         let (sender, receiver) = mpsc::channel(1);
 
         tokio::task::spawn(Self::run(capacity, paralellism, receiver));
-        Self { sender, capacity }
+        Self { sender, capacity, time_to_abort: Duration::from_secs(60) }
+    }
+
+    pub fn with_time_to_abort(self, time_to_abort: Duration) -> Self {
+        Self { time_to_abort, ..self }
     }
 
     /// This function is the main loop, it's in charge on scheduling which search request should execute first and
@@ -119,9 +139,23 @@ impl SearchQueue {
     /// Returns a search `Permit`.
     /// It should be dropped as soon as you've freed all the RAM associated with the search request being processed.
     pub async fn try_get_search_permit(&self) -> Result<Permit, MeilisearchHttpError> {
+        let now = std::time::Instant::now();
         let (sender, receiver) = oneshot::channel();
         self.sender.send(sender).await.map_err(|_| MeilisearchHttpError::SearchLimiterIsDown)?;
-        receiver.await.map_err(|_| MeilisearchHttpError::TooManySearchRequests(self.capacity))
+        let permit = receiver
+            .await
+            .map_err(|_| MeilisearchHttpError::TooManySearchRequests(self.capacity))?;
+
+        // If we've been for more than one minute to get a search permit, it's better to simply
+        // abort the search request than spending time processing something were the client
+        // most certainly exited or got a timeout a long time ago.
+        // We may find a better solution in https://github.com/actix/actix-web/issues/3462.
+        if now.elapsed() > self.time_to_abort {
+            permit.drop().await;
+            Err(MeilisearchHttpError::TooManySearchRequests(self.capacity))
+        } else {
+            Ok(permit)
+        }
     }
 
     /// Returns `Ok(())` if everything seems normal.

meilisearch/tests/auth/mod.rs

@@ -5,63 +5,3 @@ mod payload;
 mod tenant_token;
 
 mod tenant_token_multi_search;
-
-use actix_web::http::StatusCode;
-
-use crate::common::{Server, Value};
-use crate::json;
-
-impl Server {
-    pub fn use_api_key(&mut self, api_key: impl AsRef<str>) {
-        self.service.api_key = Some(api_key.as_ref().to_string());
-    }
-
-    /// Fetch and use the default admin key for nexts http requests.
-    pub async fn use_admin_key(&mut self, master_key: impl AsRef<str>) {
-        self.use_api_key(master_key);
-        let (response, code) = self.list_api_keys("").await;
-        assert_eq!(200, code, "{:?}", response);
-        let admin_key = &response["results"][1]["key"];
-        self.use_api_key(admin_key.as_str().unwrap());
-    }
-
-    pub async fn add_api_key(&self, content: Value) -> (Value, StatusCode) {
-        let url = "/keys";
-        self.service.post(url, content).await
-    }
-
-    pub async fn get_api_key(&self, key: impl AsRef<str>) -> (Value, StatusCode) {
-        let url = format!("/keys/{}", key.as_ref());
-        self.service.get(url).await
-    }
-
-    pub async fn patch_api_key(&self, key: impl AsRef<str>, content: Value) -> (Value, StatusCode) {
-        let url = format!("/keys/{}", key.as_ref());
-        self.service.patch(url, content).await
-    }
-
-    pub async fn list_api_keys(&self, params: &str) -> (Value, StatusCode) {
-        let url = format!("/keys{params}");
-        self.service.get(url).await
-    }
-
-    pub async fn delete_api_key(&self, key: impl AsRef<str>) -> (Value, StatusCode) {
-        let url = format!("/keys/{}", key.as_ref());
-        self.service.delete(url).await
-    }
-
-    pub async fn dummy_request(
-        &self,
-        method: impl AsRef<str>,
-        url: impl AsRef<str>,
-    ) -> (Value, StatusCode) {
-        match method.as_ref() {
-            "POST" => self.service.post(url, json!({})).await,
-            "PUT" => self.service.put(url, json!({})).await,
-            "PATCH" => self.service.patch(url, json!({})).await,
-            "GET" => self.service.get(url).await,
-            "DELETE" => self.service.delete(url).await,
-            _ => unreachable!(),
-        }
-    }
-}

meilisearch/tests/auth/tenant_token.rs

@@ -6,7 +6,7 @@ use once_cell::sync::Lazy;
 use time::{Duration, OffsetDateTime};
 
 use super::authorization::{ALL_ACTIONS, AUTHORIZATIONS};
-use crate::common::{Server, Value};
+use crate::common::{Server, Value, DOCUMENTS};
 use crate::json;
 
 fn generate_tenant_token(
@@ -22,36 +22,6 @@ fn generate_tenant_token(
         .unwrap()
 }
 
-static DOCUMENTS: Lazy<Value> = Lazy::new(|| {
-    json!([
-        {
-            "title": "Shazam!",
-            "id": "287947",
-            "color": ["green", "blue"]
-        },
-        {
-            "title": "Captain Marvel",
-            "id": "299537",
-            "color": ["yellow", "blue"]
-        },
-        {
-            "title": "Escape Room",
-            "id": "522681",
-            "color": ["yellow", "red"]
-        },
-        {
-            "title": "How to Train Your Dragon: The Hidden World",
-            "id": "166428",
-            "color": ["green", "red"]
-        },
-        {
-            "title": "Glass",
-            "id": "450465",
-            "color": ["blue", "red"]
-        }
-    ])
-});
-
 static INVALID_RESPONSE: Lazy<Value> = Lazy::new(|| {
     json!({
         "message": null,

meilisearch/tests/auth/tenant_token_multi_search.rs

@@ -310,6 +310,23 @@ macro_rules! compute_authorized_single_search {
                     tenant_token,
                     key_content
                 );
+
+                // federated
+                let (response, code) = server.multi_search(json!({"federation": {}, "queries" : [{"indexUid": "sales", "filter": $filter}]})).await;
+                assert_eq!(
+                    200, code,
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response, tenant_token, key_content
+                );
+                assert_eq!(
+                    // same count as the search is federated over a single query
+                    $expected_count,
+                    response["hits"].as_array().unwrap().len(),
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response,
+                    tenant_token,
+                    key_content
+                );
             }
         }
     };
@@ -375,6 +392,25 @@ macro_rules! compute_authorized_multiple_search {
                     tenant_token,
                     key_content
                 );
+
+                let (response, code) = server.multi_search(json!({"federation": {}, "queries" : [
+                    {"indexUid": "sales", "filter": $filter1},
+                    {"indexUid": "products", "filter": $filter2},
+                ]})).await;
+                assert_eq!(
+                    code, 200,
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response, tenant_token, key_content
+                );
+                assert_eq!(
+                    response["hits"].as_array().unwrap().len(),
+                    // sum of counts as the search is federated across to queries in different indexes
+                    $expected_count1 + $expected_count2,
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response,
+                    tenant_token,
+                    key_content
+                );
             }
         }
     };
@@ -433,6 +469,24 @@ macro_rules! compute_forbidden_single_search {
                     "{} using tenant_token: {:?} generated with parent_key: {:?}",
                     response, tenant_token, key_content
                 );
+
+                let (mut response, code) = server.multi_search(json!({"federation": {}, "queries" : [{"indexUid": "sales"}]})).await;
+                if failed_query_index.is_none() && !response["message"].is_null() {
+                    response["message"] = serde_json::json!(null);
+                }
+                assert_eq!(
+                    response,
+                    invalid_response(failed_query_index),
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response,
+                    tenant_token,
+                    key_content
+                );
+                assert_eq!(
+                    code, 403,
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response, tenant_token, key_content
+                );
             }
         }
     };
@@ -494,6 +548,27 @@ macro_rules! compute_forbidden_multiple_search {
                     "{} using tenant_token: {:?} generated with parent_key: {:?}",
                     response, tenant_token, key_content
                 );
+
+                let (mut response, code) = server.multi_search(json!({"federation": {}, "queries" : [
+                    {"indexUid": "sales"},
+                    {"indexUid": "products"},
+                ]})).await;
+                if failed_query_index.is_none() && !response["message"].is_null() {
+                    response["message"] = serde_json::json!(null);
+                }
+                assert_eq!(
+                    response,
+                    invalid_response(failed_query_index),
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response,
+                    tenant_token,
+                    key_content
+                );
+                assert_eq!(
+                    code, 403,
+                    "{} using tenant_token: {:?} generated with parent_key: {:?}",
+                    response, tenant_token, key_content
+                );
             }
         }
     };

meilisearch/tests/common/index.rs

@@ -1,4 +1,5 @@
 use std::fmt::Write;
+use std::marker::PhantomData;
 use std::panic::{catch_unwind, resume_unwind, UnwindSafe};
 use std::time::Duration;
 
@@ -9,19 +10,24 @@ use urlencoding::encode as urlencode;
 use super::encoder::Encoder;
 use super::service::Service;
 use super::Value;
+use super::{Owned, Shared};
 use crate::json;
 
-pub struct Index<'a> {
+pub struct Index<'a, State = Owned> {
     pub uid: String,
     pub service: &'a Service,
-    pub encoder: Encoder,
+    pub(super) encoder: Encoder,
+    pub(super) marker: PhantomData<State>,
 }
 
-#[allow(dead_code)]
-impl Index<'_> {
-    pub async fn get(&self) -> (Value, StatusCode) {
-        let url = format!("/indexes/{}", urlencode(self.uid.as_ref()));
-        self.service.get(url).await
+impl<'a> Index<'a, Owned> {
+    pub fn to_shared(&self) -> Index<'a, Shared> {
+        Index {
+            uid: self.uid.clone(),
+            service: self.service,
+            encoder: self.encoder,
+            marker: PhantomData,
+        }
     }
 
     pub async fn load_test_set(&self) -> u64 {
@@ -57,11 +63,7 @@ impl Index<'_> {
     }
 
     pub async fn create(&self, primary_key: Option<&str>) -> (Value, StatusCode) {
-        let body = json!({
-            "uid": self.uid,
-            "primaryKey": primary_key,
-        });
-        self.service.post_encoded("/indexes", body, self.encoder).await
+        self._create(primary_key).await
     }
 
     pub async fn update_raw(&self, body: Value) -> (Value, StatusCode) {
@@ -88,13 +90,7 @@ impl Index<'_> {
         documents: Value,
         primary_key: Option<&str>,
     ) -> (Value, StatusCode) {
-        let url = match primary_key {
-            Some(key) => {
-                format!("/indexes/{}/documents?primaryKey={}", urlencode(self.uid.as_ref()), key)
-            }
-            None => format!("/indexes/{}/documents", urlencode(self.uid.as_ref())),
-        };
-        self.service.post_encoded(url, documents, self.encoder).await
+        self._add_documents(documents, primary_key).await
     }
 
     pub async fn raw_add_documents(
@@ -136,80 +132,11 @@ impl Index<'_> {
         }
     }
 
-    pub async fn wait_task(&self, update_id: u64) -> Value {
-        // try several times to get status, or panic to not wait forever
-        let url = format!("/tasks/{}", update_id);
-        for _ in 0..100 {
-            let (response, status_code) = self.service.get(&url).await;
-            assert_eq!(200, status_code, "response: {}", response);
-
-            if response["status"] == "succeeded" || response["status"] == "failed" {
-                return response;
-            }
-
-            // wait 0.5 second.
-            sleep(Duration::from_millis(500)).await;
-        }
-        panic!("Timeout waiting for update id");
-    }
-
-    pub async fn get_task(&self, update_id: u64) -> (Value, StatusCode) {
-        let url = format!("/tasks/{}", update_id);
-        self.service.get(url).await
-    }
-
     pub async fn list_tasks(&self) -> (Value, StatusCode) {
         let url = format!("/tasks?indexUids={}", self.uid);
         self.service.get(url).await
     }
 
-    pub async fn filtered_tasks(
-        &self,
-        types: &[&str],
-        statuses: &[&str],
-        canceled_by: &[&str],
-    ) -> (Value, StatusCode) {
-        let mut url = format!("/tasks?indexUids={}", self.uid);
-        if !types.is_empty() {
-            let _ = write!(url, "&types={}", types.join(","));
-        }
-        if !statuses.is_empty() {
-            let _ = write!(url, "&statuses={}", statuses.join(","));
-        }
-        if !canceled_by.is_empty() {
-            let _ = write!(url, "&canceledBy={}", canceled_by.join(","));
-        }
-        self.service.get(url).await
-    }
-
-    pub async fn get_document(&self, id: u64, options: Option<Value>) -> (Value, StatusCode) {
-        let mut url = format!("/indexes/{}/documents/{}", urlencode(self.uid.as_ref()), id);
-        if let Some(options) = options {
-            write!(url, "{}", yaup::to_string(&options).unwrap()).unwrap();
-        }
-        self.service.get(url).await
-    }
-
-    pub async fn get_document_by_filter(&self, payload: Value) -> (Value, StatusCode) {
-        let url = format!("/indexes/{}/documents/fetch", urlencode(self.uid.as_ref()));
-        self.service.post(url, payload).await
-    }
-
-    pub async fn get_all_documents_raw(&self, options: &str) -> (Value, StatusCode) {
-        let url = format!("/indexes/{}/documents{}", urlencode(self.uid.as_ref()), options);
-        self.service.get(url).await
-    }
-
-    pub async fn get_all_documents(&self, options: GetAllDocumentsOptions) -> (Value, StatusCode) {
-        let url = format!(
-            "/indexes/{}/documents{}",
-            urlencode(self.uid.as_ref()),
-            yaup::to_string(&options).unwrap()
-        );
-
-        self.service.get(url).await
-    }
-
     pub async fn delete_document(&self, id: u64) -> (Value, StatusCode) {
         let url = format!("/indexes/{}/documents/{}", urlencode(self.uid.as_ref()), id);
         self.service.delete(url).await
@@ -237,14 +164,8 @@ impl Index<'_> {
         self.service.post_encoded(url, body, self.encoder).await
     }
 
-    pub async fn settings(&self) -> (Value, StatusCode) {
-        let url = format!("/indexes/{}/settings", urlencode(self.uid.as_ref()));
-        self.service.get(url).await
-    }
-
     pub async fn update_settings(&self, settings: Value) -> (Value, StatusCode) {
-        let url = format!("/indexes/{}/settings", urlencode(self.uid.as_ref()));
-        self.service.patch_encoded(url, settings, self.encoder).await
+        self._update_settings(settings).await
     }
 
     pub async fn update_settings_displayed_attributes(
@@ -327,6 +248,146 @@ impl Index<'_> {
         self.service.delete(url).await
     }
 
+    pub async fn update_distinct_attribute(&self, value: Value) -> (Value, StatusCode) {
+        let url =
+            format!("/indexes/{}/settings/{}", urlencode(self.uid.as_ref()), "distinct-attribute");
+        self.service.put_encoded(url, value, self.encoder).await
+    }
+}
+
+impl<'a> Index<'a, Shared> {
+    /// You cannot modify the content of a shared index, thus the delete_document_by_filter call
+    /// must fail. If the task successfully enqueue itself, we'll wait for the task to finishes,
+    /// and if it succeed the function will panic.
+    pub async fn delete_document_by_filter_fail(&self, body: Value) -> (Value, StatusCode) {
+        let (mut task, code) = self._delete_document_by_filter(body).await;
+        if code.is_success() {
+            task = self.wait_task(task.uid()).await;
+            if task.is_success() {
+                panic!(
+                    "`delete_document_by_filter_fail` succeeded: {}",
+                    serde_json::to_string_pretty(&task).unwrap()
+                );
+            }
+        }
+        (task, code)
+    }
+}
+
+#[allow(dead_code)]
+impl<State> Index<'_, State> {
+    pub async fn get(&self) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}", urlencode(self.uid.as_ref()));
+        self.service.get(url).await
+    }
+
+    /// add_documents is not allowed on shared index but we need to use it to initialize
+    /// a bunch of very common indexes in `common/mod.rs`.
+    pub(super) async fn _add_documents(
+        &self,
+        documents: Value,
+        primary_key: Option<&str>,
+    ) -> (Value, StatusCode) {
+        let url = match primary_key {
+            Some(key) => {
+                format!("/indexes/{}/documents?primaryKey={}", urlencode(self.uid.as_ref()), key)
+            }
+            None => format!("/indexes/{}/documents", urlencode(self.uid.as_ref())),
+        };
+        self.service.post_encoded(url, documents, self.encoder).await
+    }
+
+    pub(super) async fn _update_settings(&self, settings: Value) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/settings", urlencode(self.uid.as_ref()));
+        self.service.patch_encoded(url, settings, self.encoder).await
+    }
+
+    pub(super) async fn _delete_document_by_filter(&self, body: Value) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/documents/delete", urlencode(self.uid.as_ref()));
+        self.service.post_encoded(url, body, self.encoder).await
+    }
+
+    pub(super) async fn _create(&self, primary_key: Option<&str>) -> (Value, StatusCode) {
+        let body = json!({
+            "uid": self.uid,
+            "primaryKey": primary_key,
+        });
+        self.service.post_encoded("/indexes", body, self.encoder).await
+    }
+    pub async fn wait_task(&self, update_id: u64) -> Value {
+        // try several times to get status, or panic to not wait forever
+        let url = format!("/tasks/{}", update_id);
+        for _ in 0..100 {
+            let (response, status_code) = self.service.get(&url).await;
+            assert_eq!(200, status_code, "response: {}", response);
+
+            if response["status"] == "succeeded" || response["status"] == "failed" {
+                return response;
+            }
+
+            // wait 0.5 second.
+            sleep(Duration::from_millis(500)).await;
+        }
+        panic!("Timeout waiting for update id");
+    }
+
+    pub async fn get_task(&self, update_id: u64) -> (Value, StatusCode) {
+        let url = format!("/tasks/{}", update_id);
+        self.service.get(url).await
+    }
+
+    pub async fn filtered_tasks(
+        &self,
+        types: &[&str],
+        statuses: &[&str],
+        canceled_by: &[&str],
+    ) -> (Value, StatusCode) {
+        let mut url = format!("/tasks?indexUids={}", self.uid);
+        if !types.is_empty() {
+            let _ = write!(url, "&types={}", types.join(","));
+        }
+        if !statuses.is_empty() {
+            let _ = write!(url, "&statuses={}", statuses.join(","));
+        }
+        if !canceled_by.is_empty() {
+            let _ = write!(url, "&canceledBy={}", canceled_by.join(","));
+        }
+        self.service.get(url).await
+    }
+
+    pub async fn get_document(&self, id: u64, options: Option<Value>) -> (Value, StatusCode) {
+        let mut url = format!("/indexes/{}/documents/{}", urlencode(self.uid.as_ref()), id);
+        if let Some(options) = options {
+            write!(url, "{}", yaup::to_string(&options).unwrap()).unwrap();
+        }
+        self.service.get(url).await
+    }
+
+    pub async fn get_document_by_filter(&self, payload: Value) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/documents/fetch", urlencode(self.uid.as_ref()));
+        self.service.post(url, payload).await
+    }
+
+    pub async fn get_all_documents_raw(&self, options: &str) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/documents{}", urlencode(self.uid.as_ref()), options);
+        self.service.get(url).await
+    }
+
+    pub async fn get_all_documents(&self, options: GetAllDocumentsOptions) -> (Value, StatusCode) {
+        let url = format!(
+            "/indexes/{}/documents{}",
+            urlencode(self.uid.as_ref()),
+            yaup::to_string(&options).unwrap()
+        );
+
+        self.service.get(url).await
+    }
+
+    pub async fn settings(&self) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/settings", urlencode(self.uid.as_ref()));
+        self.service.get(url).await
+    }
+
     pub async fn stats(&self) -> (Value, StatusCode) {
         let url = format!("/indexes/{}/stats", urlencode(self.uid.as_ref()));
         self.service.get(url).await
@@ -411,12 +472,6 @@ impl Index<'_> {
         self.service.post_encoded(url, query, self.encoder).await
     }
 
-    pub async fn update_distinct_attribute(&self, value: Value) -> (Value, StatusCode) {
-        let url =
-            format!("/indexes/{}/settings/{}", urlencode(self.uid.as_ref()), "distinct-attribute");
-        self.service.put_encoded(url, value, self.encoder).await
-    }
-
     pub async fn get_distinct_attribute(&self) -> (Value, StatusCode) {
         let url =
             format!("/indexes/{}/settings/{}", urlencode(self.uid.as_ref()), "distinct-attribute");

meilisearch/tests/common/mod.rs

@@ -8,9 +8,16 @@ use std::fmt::{self, Display};
 #[allow(unused)]
 pub use index::GetAllDocumentsOptions;
 use meili_snap::json_string;
+use once_cell::sync::Lazy;
 use serde::{Deserialize, Serialize};
 #[allow(unused)]
 pub use server::{default_settings, Server};
+use tokio::sync::OnceCell;
+
+use crate::common::index::Index;
+
+pub enum Shared {}
+pub enum Owned {}
 
 #[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq)]
 pub struct Value(pub serde_json::Value);
@@ -26,6 +33,25 @@ impl Value {
             panic!("Didn't find any task id in: {self}");
         }
     }
+
+    /// Return `true` if the `status` field is set to `succeeded`.
+    /// Panic if the `status` field doesn't exists.
+    #[track_caller]
+    pub fn is_success(&self) -> bool {
+        if !self["status"].is_string() {
+            panic!("Called `is_success` on {}", serde_json::to_string_pretty(&self.0).unwrap());
+        }
+        self["status"] == serde_json::Value::String(String::from("succeeded"))
+    }
+
+    // Panic if the json doesn't contain the `status` field set to "succeeded"
+    #[track_caller]
+    pub fn succeeded(&self) -> &Self {
+        if !self.is_success() {
+            panic!("Called succeeded on {}", serde_json::to_string_pretty(&self.0).unwrap());
+        }
+        self
+    }
 }
 
 impl From<serde_json::Value> for Value {
@@ -71,7 +97,15 @@ impl Display for Value {
         write!(
             f,
             "{}",
-            json_string!(self, { ".enqueuedAt" => "[date]", ".startedAt" => "[date]", ".finishedAt" => "[date]", ".duration" => "[duration]", ".processingTimeMs" => "[duration]" })
+            json_string!(self, {
+                ".uid" => "[uid]",
+                ".enqueuedAt" => "[date]",
+                ".startedAt" => "[date]",
+                ".finishedAt" => "[date]",
+                ".duration" => "[duration]",
+                ".processingTimeMs" => "[duration]",
+                ".details.embedders.*.url" => "[url]"
+            })
         )
     }
 }
@@ -105,3 +139,253 @@ macro_rules! test_post_get_search {
             .map_err(|e| panic!("panic in post route: {:?}", e.downcast_ref::<&str>().unwrap()));
     };
 }
+
+pub async fn shared_does_not_exists_index() -> &'static Index<'static, Shared> {
+    static INDEX: Lazy<Index<'static, Shared>> = Lazy::new(|| {
+        let server = Server::new_shared();
+        server._index("DOES_NOT_EXISTS").to_shared()
+    });
+    &INDEX
+}
+
+pub async fn shared_empty_index() -> &'static Index<'static, Shared> {
+    static INDEX: OnceCell<Index<'static, Shared>> = OnceCell::const_new();
+
+    INDEX
+        .get_or_init(|| async {
+            let server = Server::new_shared();
+            let index = server._index("EMPTY_INDEX").to_shared();
+            let (response, _code) = index._create(None).await;
+            index.wait_task(response.uid()).await.succeeded();
+            index
+        })
+        .await
+}
+
+pub static DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+        {
+            "title": "Shazam!",
+            "id": "287947",
+            "color": ["green", "blue"],
+            "_vectors": { "manual": [1, 2, 3]},
+        },
+        {
+            "title": "Captain Marvel",
+            "id": "299537",
+            "color": ["yellow", "blue"],
+            "_vectors": { "manual": [1, 2, 54] },
+        },
+        {
+            "title": "Escape Room",
+            "id": "522681",
+            "color": ["yellow", "red"],
+            "_vectors": { "manual": [10, -23, 32] },
+        },
+        {
+            "title": "How to Train Your Dragon: The Hidden World",
+            "id": "166428",
+            "color": ["green", "red"],
+            "_vectors": { "manual": [-100, 231, 32] },
+        },
+        {
+            "title": "Gläss",
+            "id": "450465",
+            "color": ["blue", "red"],
+            "_vectors": { "manual": [-100, 340, 90] },
+        }
+    ])
+});
+
+pub async fn shared_index_with_documents() -> &'static Index<'static, Shared> {
+    static INDEX: OnceCell<Index<'static, Shared>> = OnceCell::const_new();
+    INDEX.get_or_init(|| async {
+        let server = Server::new_shared();
+        let index = server._index("SHARED_DOCUMENTS").to_shared();
+        let documents = DOCUMENTS.clone();
+        let (response, _code) = index._add_documents(documents, None).await;
+        index.wait_task(response.uid()).await.succeeded();
+        let (response, _code) = index
+            ._update_settings(
+                json!({"filterableAttributes": ["id", "title"], "sortableAttributes": ["id", "title"]}),
+            )
+            .await;
+        index.wait_task(response.uid()).await.succeeded();
+        index
+    }).await
+}
+
+pub static SCORE_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+        {
+            "title": "Batman the dark knight returns: Part 1",
+            "id": "A",
+        },
+        {
+            "title": "Batman the dark knight returns: Part 2",
+            "id": "B",
+        },
+        {
+            "title": "Batman Returns",
+            "id": "C",
+        },
+        {
+            "title": "Batman",
+            "id": "D",
+        },
+        {
+            "title": "Badman",
+            "id": "E",
+        }
+    ])
+});
+
+pub static NESTED_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+        {
+            "id": 852,
+            "father": "jean",
+            "mother": "michelle",
+            "doggos": [
+                {
+                    "name": "bobby",
+                    "age": 2,
+                },
+                {
+                    "name": "buddy",
+                    "age": 4,
+                },
+            ],
+            "cattos": "pésti",
+            "_vectors": { "manual": [1, 2, 3]},
+        },
+        {
+            "id": 654,
+            "father": "pierre",
+            "mother": "sabine",
+            "doggos": [
+                {
+                    "name": "gros bill",
+                    "age": 8,
+                },
+            ],
+            "cattos": ["simba", "pestiféré"],
+            "_vectors": { "manual": [1, 2, 54] },
+        },
+        {
+            "id": 750,
+            "father": "romain",
+            "mother": "michelle",
+            "cattos": ["enigma"],
+            "_vectors": { "manual": [10, 23, 32] },
+        },
+        {
+            "id": 951,
+            "father": "jean-baptiste",
+            "mother": "sophie",
+            "doggos": [
+                {
+                    "name": "turbo",
+                    "age": 5,
+                },
+                {
+                    "name": "fast",
+                    "age": 6,
+                },
+            ],
+            "cattos": ["moumoute", "gomez"],
+            "_vectors": { "manual": [10, 23, 32] },
+        },
+    ])
+});
+
+pub async fn shared_index_with_nested_documents() -> &'static Index<'static, Shared> {
+    static INDEX: OnceCell<Index<'static, Shared>> = OnceCell::const_new();
+    INDEX.get_or_init(|| async {
+        let server = Server::new_shared();
+        let index = server._index("SHARED_NESTED_DOCUMENTS").to_shared();
+        let documents = NESTED_DOCUMENTS.clone();
+        let (response, _code) = index._add_documents(documents, None).await;
+        index.wait_task(response.uid()).await.succeeded();
+        let (response, _code) = index
+            ._update_settings(
+                json!({"filterableAttributes": ["father", "doggos"], "sortableAttributes": ["doggos"]}),
+            )
+            .await;
+        index.wait_task(response.uid()).await.succeeded();
+        index
+    }).await
+}
+
+pub static FRUITS_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+        {
+            "name": "Exclusive sale: green apple",
+            "id": "green-apple-boosted",
+            "BOOST": true
+        },
+        {
+            "name": "Pear",
+            "id": "pear",
+        },
+        {
+            "name": "Red apple gala",
+            "id": "red-apple-gala",
+        },
+        {
+            "name": "Exclusive sale: Red Tomato",
+            "id": "red-tomatoes-boosted",
+            "BOOST": true
+        },
+        {
+            "name": "Exclusive sale: Red delicious apple",
+            "id": "red-delicious-boosted",
+            "BOOST": true,
+        }
+    ])
+});
+
+pub static VECTOR_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+      {
+        "id": "A",
+        "description": "the dog barks at the cat",
+        "_vectors": {
+          // dimensions [canine, feline, young]
+          "animal": [0.9, 0.8, 0.05],
+          // dimensions [negative/positive, energy]
+          "sentiment": [-0.1, 0.55]
+        }
+      },
+      {
+        "id": "B",
+        "description": "the kitten scratched the beagle",
+        "_vectors": {
+          // dimensions [canine, feline, young]
+          "animal": [0.8, 0.9, 0.5],
+          // dimensions [negative/positive, energy]
+          "sentiment": [-0.2, 0.65]
+        }
+      },
+      {
+        "id": "C",
+        "description": "the dog had to stay alone today",
+        "_vectors": {
+          // dimensions [canine, feline, young]
+          "animal": [0.85, 0.02, 0.1],
+          // dimensions [negative/positive, energy]
+          "sentiment": [-1.0, 0.1]
+        }
+      },
+      {
+        "id": "D",
+        "description": "the little boy pets the puppy",
+        "_vectors": {
+          // dimensions [canine, feline, young]
+          "animal": [0.8, 0.09, 0.8],
+          // dimensions [negative/positive, energy]
+          "sentiment": [0.8, 0.3]
+        }
+      },
+    ])
+});

meilisearch/tests/common/server.rs

@@ -1,6 +1,8 @@
 #![allow(dead_code)]
 
+use std::marker::PhantomData;
 use std::path::Path;
+use std::str::FromStr;
 use std::time::Duration;
 
 use actix_http::body::MessageBody;
@@ -8,29 +10,35 @@ use actix_web::dev::ServiceResponse;
 use actix_web::http::StatusCode;
 use byte_unit::{Byte, Unit};
 use clap::Parser;
-use meilisearch::option::{IndexerOpts, MaxMemory, Opt};
-use meilisearch::{analytics, create_app, setup_meilisearch, SubscriberForSecondLayer};
+use meilisearch::option::{IndexerOpts, MaxMemory, MaxThreads, Opt};
+use meilisearch::setup_meilisearch;
 use once_cell::sync::Lazy;
 use tempfile::TempDir;
+use tokio::sync::OnceCell;
 use tokio::time::sleep;
-use tracing::level_filters::LevelFilter;
-use tracing_subscriber::Layer;
+use uuid::Uuid;
 
 use super::index::Index;
 use super::service::Service;
+use super::{Owned, Shared};
 use crate::common::encoder::Encoder;
 use crate::common::Value;
 use crate::json;
 
-pub struct Server {
+pub struct Server<State = Owned> {
     pub service: Service,
     // hold ownership to the tempdir while we use the server instance.
     _dir: Option<TempDir>,
+    _marker: PhantomData<State>,
 }
 
 pub static TEST_TEMP_DIR: Lazy<TempDir> = Lazy::new(|| TempDir::new().unwrap());
 
-impl Server {
+impl Server<Owned> {
+    fn into_shared(self) -> Server<Shared> {
+        Server { service: self.service, _dir: self._dir, _marker: PhantomData }
+    }
+
     pub async fn new() -> Self {
         let dir = TempDir::new().unwrap();
 
@@ -45,7 +53,7 @@ impl Server {
         let (index_scheduler, auth) = setup_meilisearch(&options).unwrap();
         let service = Service { index_scheduler, auth, options, api_key: None };
 
-        Server { service, _dir: Some(dir) }
+        Server { service, _dir: Some(dir), _marker: PhantomData }
     }
 
     pub async fn new_auth_with_options(mut options: Opt, dir: TempDir) -> Self {
@@ -60,7 +68,7 @@ impl Server {
         let (index_scheduler, auth) = setup_meilisearch(&options).unwrap();
         let service = Service { index_scheduler, auth, options, api_key: None };
 
-        Server { service, _dir: Some(dir) }
+        Server { service, _dir: Some(dir), _marker: PhantomData }
     }
 
     pub async fn new_auth() -> Self {
@@ -73,38 +81,35 @@ impl Server {
         let (index_scheduler, auth) = setup_meilisearch(&options)?;
         let service = Service { index_scheduler, auth, options, api_key: None };
 
-        Ok(Server { service, _dir: None })
+        Ok(Server { service, _dir: None, _marker: PhantomData })
     }
 
-    pub async fn init_web_app(
-        &self,
-    ) -> impl actix_web::dev::Service<
-        actix_http::Request,
-        Response = ServiceResponse<impl MessageBody>,
-        Error = actix_web::Error,
-    > {
-        let (_route_layer, route_layer_handle) =
-            tracing_subscriber::reload::Layer::new(None.with_filter(
-                tracing_subscriber::filter::Targets::new().with_target("", LevelFilter::OFF),
-            ));
-        let (_stderr_layer, stderr_layer_handle) = tracing_subscriber::reload::Layer::new(
-            (Box::new(
-                tracing_subscriber::fmt::layer()
-                    .with_span_events(tracing_subscriber::fmt::format::FmtSpan::CLOSE),
-            )
-                as Box<dyn tracing_subscriber::Layer<SubscriberForSecondLayer> + Send + Sync>)
-                .with_filter(tracing_subscriber::filter::Targets::new()),
-        );
+    pub fn use_api_key(&mut self, api_key: impl AsRef<str>) {
+        self.service.api_key = Some(api_key.as_ref().to_string());
+    }
 
-        actix_web::test::init_service(create_app(
-            self.service.index_scheduler.clone().into(),
-            self.service.auth.clone().into(),
-            self.service.options.clone(),
-            (route_layer_handle, stderr_layer_handle),
-            analytics::MockAnalytics::new(&self.service.options),
-            true,
-        ))
-        .await
+    /// Fetch and use the default admin key for nexts http requests.
+    pub async fn use_admin_key(&mut self, master_key: impl AsRef<str>) {
+        self.use_api_key(master_key);
+        let (response, code) = self.list_api_keys("").await;
+        assert_eq!(200, code, "{:?}", response);
+        let admin_key = &response["results"][1]["key"];
+        self.use_api_key(admin_key.as_str().unwrap());
+    }
+
+    pub async fn add_api_key(&self, content: Value) -> (Value, StatusCode) {
+        let url = "/keys";
+        self.service.post(url, content).await
+    }
+
+    pub async fn patch_api_key(&self, key: impl AsRef<str>, content: Value) -> (Value, StatusCode) {
+        let url = format!("/keys/{}", key.as_ref());
+        self.service.patch(url, content).await
+    }
+
+    pub async fn delete_api_key(&self, key: impl AsRef<str>) -> (Value, StatusCode) {
+        let url = format!("/keys/{}", key.as_ref());
+        self.service.delete(url).await
     }
 
     /// Returns a view to an index. There is no guarantee that the index exists.
@@ -117,15 +122,12 @@ impl Server {
     }
 
     pub fn index_with_encoder(&self, uid: impl AsRef<str>, encoder: Encoder) -> Index<'_> {
-        Index { uid: uid.as_ref().to_string(), service: &self.service, encoder }
-    }
-
-    pub async fn multi_search(&self, queries: Value) -> (Value, StatusCode) {
-        self.service.post("/multi-search", queries).await
-    }
-
-    pub async fn list_indexes_raw(&self, parameters: &str) -> (Value, StatusCode) {
-        self.service.get(format!("/indexes{parameters}")).await
+        Index {
+            uid: uid.as_ref().to_string(),
+            service: &self.service,
+            encoder,
+            marker: PhantomData,
+        }
     }
 
     pub async fn list_indexes(
@@ -149,10 +151,6 @@ impl Server {
         }
     }
 
-    pub async fn version(&self) -> (Value, StatusCode) {
-        self.service.get("/version").await
-    }
-
     pub async fn stats(&self) -> (Value, StatusCode) {
         self.service.get("/stats").await
     }
@@ -161,12 +159,174 @@ impl Server {
         self.service.get("/tasks").await
     }
 
+    pub async fn set_features(&self, value: Value) -> (Value, StatusCode) {
+        self.service.patch("/experimental-features", value).await
+    }
+
+    pub async fn get_metrics(&self) -> (Value, StatusCode) {
+        self.service.get("/metrics").await
+    }
+}
+
+impl Server<Shared> {
+    fn init_new_shared_instance() -> Server<Shared> {
+        let dir = TempDir::new().unwrap();
+
+        if cfg!(windows) {
+            std::env::set_var("TMP", TEST_TEMP_DIR.path());
+        } else {
+            std::env::set_var("TMPDIR", TEST_TEMP_DIR.path());
+        }
+
+        let options = default_settings(dir.path());
+
+        let (index_scheduler, auth) = setup_meilisearch(&options).unwrap();
+        let service = Service { index_scheduler, auth, api_key: None, options };
+
+        Server { service, _dir: Some(dir), _marker: PhantomData }
+    }
+
+    pub fn new_shared() -> &'static Server<Shared> {
+        static SERVER: Lazy<Server<Shared>> = Lazy::new(Server::init_new_shared_instance);
+        &SERVER
+    }
+
+    pub async fn new_shared_with_admin_key() -> &'static Server<Shared> {
+        static SERVER: OnceCell<Server<Shared>> = OnceCell::const_new();
+        SERVER
+            .get_or_init(|| async {
+                let mut server = Server::new_auth().await;
+                server.use_admin_key("MASTER_KEY").await;
+                server.into_shared()
+            })
+            .await
+    }
+
+    /// You shouldn't access random indexes on a shared instance thus this method
+    /// must fail.
+    pub async fn get_index_fail(&self, uid: impl AsRef<str>) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}", urlencoding::encode(uid.as_ref()));
+        let (value, code) = self.service.get(url).await;
+        if code.is_success() {
+            panic!("`get_index_fail` succeeded with uid: {}", uid.as_ref());
+        }
+        (value, code)
+    }
+
+    pub async fn delete_index_fail(&self, uid: impl AsRef<str>) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}", urlencoding::encode(uid.as_ref()));
+        let (value, code) = self.service.delete(url).await;
+        if code.is_success() {
+            panic!("`delete_index_fail` succeeded with uid: {}", uid.as_ref());
+        }
+        (value, code)
+    }
+
+    pub async fn update_raw_index_fail(
+        &self,
+        uid: impl AsRef<str>,
+        body: Value,
+    ) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}", urlencoding::encode(uid.as_ref()));
+        let (value, code) = self.service.patch_encoded(url, body, Encoder::Plain).await;
+        if code.is_success() {
+            panic!("`update_raw_index_fail` succeeded with uid: {}", uid.as_ref());
+        }
+        (value, code)
+    }
+
+    /// Since this call updates the state of the instance, it must fail.
+    /// If it doesn't fail, the test will panic to help you debug what
+    /// is going on.
+    pub async fn create_index_fail(&self, body: Value) -> (Value, StatusCode) {
+        let (mut task, code) = self._create_index(body).await;
+        if code.is_success() {
+            task = self.wait_task(task.uid()).await;
+            if task.is_success() {
+                panic!(
+                    "`create_index_fail` succeeded: {}",
+                    serde_json::to_string_pretty(&task).unwrap()
+                );
+            }
+        }
+        (task, code)
+    }
+}
+
+impl<State> Server<State> {
+    pub async fn init_web_app(
+        &self,
+    ) -> impl actix_web::dev::Service<
+        actix_http::Request,
+        Response = ServiceResponse<impl MessageBody>,
+        Error = actix_web::Error,
+    > {
+        self.service.init_web_app().await
+    }
+
+    pub async fn list_api_keys(&self, params: &str) -> (Value, StatusCode) {
+        let url = format!("/keys{params}");
+        self.service.get(url).await
+    }
+
+    pub async fn dummy_request(
+        &self,
+        method: impl AsRef<str>,
+        url: impl AsRef<str>,
+    ) -> (Value, StatusCode) {
+        match method.as_ref() {
+            "POST" => self.service.post(url, json!({})).await,
+            "PUT" => self.service.put(url, json!({})).await,
+            "PATCH" => self.service.patch(url, json!({})).await,
+            "GET" => self.service.get(url).await,
+            "DELETE" => self.service.delete(url).await,
+            _ => unreachable!(),
+        }
+    }
+
+    pub async fn get_api_key(&self, key: impl AsRef<str>) -> (Value, StatusCode) {
+        let url = format!("/keys/{}", key.as_ref());
+        self.service.get(url).await
+    }
+
+    pub(super) fn _index(&self, uid: impl AsRef<str>) -> Index<'_> {
+        Index {
+            uid: uid.as_ref().to_string(),
+            service: &self.service,
+            encoder: Encoder::Plain,
+            marker: PhantomData,
+        }
+    }
+
+    /// Returns a view to an index. There is no guarantee that the index exists.
+    pub fn unique_index(&self) -> Index<'_> {
+        let uuid = Uuid::new_v4();
+        Index {
+            uid: uuid.to_string(),
+            service: &self.service,
+            encoder: Encoder::Plain,
+            marker: PhantomData,
+        }
+    }
+
+    pub(super) async fn _create_index(&self, body: Value) -> (Value, StatusCode) {
+        self.service.post("/indexes", body).await
+    }
+
+    pub async fn multi_search(&self, queries: Value) -> (Value, StatusCode) {
+        self.service.post("/multi-search", queries).await
+    }
+
+    pub async fn list_indexes_raw(&self, parameters: &str) -> (Value, StatusCode) {
+        self.service.get(format!("/indexes{parameters}")).await
+    }
+
     pub async fn tasks_filter(&self, filter: &str) -> (Value, StatusCode) {
         self.service.get(format!("/tasks?{}", filter)).await
     }
 
-    pub async fn get_dump_status(&self, uid: &str) -> (Value, StatusCode) {
-        self.service.get(format!("/dumps/{}/status", uid)).await
+    pub async fn version(&self) -> (Value, StatusCode) {
+        self.service.get("/version").await
     }
 
     pub async fn create_dump(&self) -> (Value, StatusCode) {
@@ -214,14 +374,6 @@ impl Server {
     pub async fn get_features(&self) -> (Value, StatusCode) {
         self.service.get("/experimental-features").await
     }
-
-    pub async fn set_features(&self, value: Value) -> (Value, StatusCode) {
-        self.service.patch("/experimental-features", value).await
-    }
-
-    pub async fn get_metrics(&self) -> (Value, StatusCode) {
-        self.service.get("/metrics").await
-    }
 }
 
 pub fn default_settings(dir: impl AsRef<Path>) -> Opt {
@@ -239,7 +391,8 @@ pub fn default_settings(dir: impl AsRef<Path>) -> Opt {
             // memory has to be unlimited because several meilisearch are running in test context.
             max_indexing_memory: MaxMemory::unlimited(),
             skip_index_budget: true,
-            ..Parser::parse_from(None as Option<&str>)
+            // Having 2 threads makes the tests way faster
+            max_indexing_threads: MaxThreads::from_str("2").unwrap(),
         },
         experimental_enable_metrics: false,
         ..Parser::parse_from(None as Option<&str>)

meilisearch/tests/common/service.rs

@@ -1,10 +1,15 @@
+use std::num::NonZeroUsize;
 use std::sync::Arc;
 
+use actix_web::body::MessageBody;
+use actix_web::dev::ServiceResponse;
 use actix_web::http::header::ContentType;
 use actix_web::http::StatusCode;
 use actix_web::test;
 use actix_web::test::TestRequest;
+use actix_web::web::Data;
 use index_scheduler::IndexScheduler;
+use meilisearch::search_queue::SearchQueue;
 use meilisearch::{analytics, create_app, Opt, SubscriberForSecondLayer};
 use meilisearch_auth::AuthController;
 use tracing::level_filters::LevelFilter;
@@ -106,7 +111,13 @@ impl Service {
         self.request(req).await
     }
 
-    pub async fn request(&self, mut req: test::TestRequest) -> (Value, StatusCode) {
+    pub async fn init_web_app(
+        &self,
+    ) -> impl actix_web::dev::Service<
+        actix_http::Request,
+        Response = ServiceResponse<impl MessageBody>,
+        Error = actix_web::Error,
+    > {
         let (_route_layer, route_layer_handle) =
             tracing_subscriber::reload::Layer::new(None.with_filter(
                 tracing_subscriber::filter::Targets::new().with_target("", LevelFilter::OFF),
@@ -119,16 +130,25 @@ impl Service {
                 as Box<dyn tracing_subscriber::Layer<SubscriberForSecondLayer> + Send + Sync>)
                 .with_filter(tracing_subscriber::filter::Targets::new()),
         );
+        let search_queue = SearchQueue::new(
+            self.options.experimental_search_queue_size,
+            NonZeroUsize::new(1).unwrap(),
+        );
 
-        let app = test::init_service(create_app(
+        actix_web::test::init_service(create_app(
             self.index_scheduler.clone().into(),
             self.auth.clone().into(),
+            Data::new(search_queue),
             self.options.clone(),
             (route_layer_handle, stderr_layer_handle),
             analytics::MockAnalytics::new(&self.options),
             true,
         ))
-        .await;
+        .await
+    }
+
+    pub async fn request(&self, mut req: test::TestRequest) -> (Value, StatusCode) {
+        let app = self.init_web_app().await;
 
         if let Some(api_key) = &self.api_key {
             req = req.insert_header(("Authorization", ["Bearer ", api_key].concat()));