Start with a filterBoosting ranking rule when boostingFilter is present

4191: Remove banner r=Kerollmops a=curquiza



Co-authored-by: Clémentine U. - curqui <clementine@meilisearch.com>

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

@@ -50,7 +50,7 @@ jobs:
     needs: check-version
     steps:
       - name: Create PR to Homebrew
-        uses: mislav/bump-homebrew-formula-action@v2
+        uses: mislav/bump-homebrew-formula-action@v3
         with:
           formula-name: meilisearch
           formula-path: Formula/m/meilisearch.rb

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

@@ -63,7 +63,7 @@ jobs:
         uses: docker/setup-buildx-action@v3
 
       - name: Login to Docker Hub
-        uses: docker/login-action@v2
+        uses: docker/login-action@v3
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}

.github/workflows/sdks-tests.yml

@@ -160,7 +160,7 @@ jobs:
         with:
           repository: meilisearch/meilisearch-js
       - name: Setup node
-        uses: actions/setup-node@v3
+        uses: actions/setup-node@v4
         with:
           cache: 'yarn'
       - name: Install dependencies
@@ -318,7 +318,7 @@ jobs:
         with:
           repository: meilisearch/meilisearch-js-plugins
       - name: Setup node
-        uses: actions/setup-node@v3
+        uses: actions/setup-node@v4
         with:
           cache: yarn
       - name: Install dependencies

.github/workflows/test-suite.yml

@@ -43,7 +43,7 @@ jobs:
           toolchain: nightly
           override: true
       - name: Cache dependencies
-        uses: Swatinem/rust-cache@v2.6.2
+        uses: Swatinem/rust-cache@v2.7.1
       - name: Run cargo check without any default features
         uses: actions-rs/cargo@v1
         with:
@@ -65,7 +65,7 @@ jobs:
     steps:
       - uses: actions/checkout@v3
       - name: Cache dependencies
-        uses: Swatinem/rust-cache@v2.6.2
+        uses: Swatinem/rust-cache@v2.7.1
       - name: Run cargo check without any default features
         uses: actions-rs/cargo@v1
         with:
@@ -149,7 +149,7 @@ jobs:
           toolchain: stable
           override: true
       - name: Cache dependencies
-        uses: Swatinem/rust-cache@v2.6.2
+        uses: Swatinem/rust-cache@v2.7.1
       - name: Run tests in debug
         uses: actions-rs/cargo@v1
         with:
@@ -168,7 +168,7 @@ jobs:
           override: true
           components: clippy
       - name: Cache dependencies
-        uses: Swatinem/rust-cache@v2.6.2
+        uses: Swatinem/rust-cache@v2.7.1
       - name: Run cargo clippy
         uses: actions-rs/cargo@v1
         with:
@@ -187,7 +187,7 @@ jobs:
           override: true
           components: rustfmt
       - name: Cache dependencies
-        uses: Swatinem/rust-cache@v2.6.2
+        uses: Swatinem/rust-cache@v2.7.1
       - name: Run cargo fmt
         # Since we never ran the `build.rs` script in the benchmark directory we are missing one auto-generated import file.
         # Since we want to trigger (and fail) this action as fast as possible, instead of building the benchmark crate

Cargo.toml

@@ -2,7 +2,6 @@
 resolver = "2"
 members = [
     "meilisearch",
-    "meilitool",
     "meilisearch-types",
     "meilisearch-auth",
     "meili-snap",
@@ -19,7 +18,7 @@ members = [
 ]
 
 [workspace.package]
-version = "1.5.0"
+version = "1.4.1"
 authors = ["Quentin de Quelen <quentin@dequelen.me>", "Clément Renault <clement@meilisearch.com>"]
 description = "Meilisearch HTTP server"
 homepage = "https://meilisearch.com"

Dockerfile

@@ -3,7 +3,7 @@ FROM    rust:alpine3.16 AS compiler
 
 RUN     apk add -q --update-cache --no-cache build-base openssl-dev
 
-WORKDIR /
+WORKDIR /meilisearch
 
 ARG     COMMIT_SHA
 ARG     COMMIT_DATE
@@ -17,7 +17,7 @@ RUN     set -eux; \
         if [ "$apkArch" = "aarch64" ]; then \
             export JEMALLOC_SYS_WITH_LG_PAGE=16; \
         fi && \
-        cargo build --release -p meilisearch -p meilitool
+        cargo build --release
 
 # Run
 FROM    alpine:3.16
@@ -28,10 +28,9 @@ ENV     MEILI_SERVER_PROVIDER docker
 RUN     apk update --quiet \
         && apk add -q --no-cache libgcc tini curl
 
-# add meilisearch and meilitool to the `/bin` so you can run it from anywhere
-# and it's easy to find.
-COPY    --from=compiler /target/release/meilisearch /bin/meilisearch
-COPY    --from=compiler /target/release/meilitool /bin/meilitool
+# add meilisearch to the `/bin` so you can run it from anywhere and it's easy
+# to find.
+COPY    --from=compiler /meilisearch/target/release/meilisearch /bin/meilisearch
 # To stay compatible with the older version of the container (pre v0.27.0) we're
 # going to symlink the meilisearch binary in the path to `/meilisearch`
 RUN     ln -s /bin/meilisearch /meilisearch

README.md

@@ -25,12 +25,6 @@
 
 <p align="center">⚡ A lightning-fast search engine that fits effortlessly into your apps, websites, and workflow 🔍</p>
 
----
-
-### 🔥 On November 2nd, we are hosting our first-ever live demo and product updates for [Meilisearch Cloud](https://www.meilisearch.com/cloud?utm_campaign=oss&utm_source=github&utm_medium=meilisearch). Make sure to [register here](https://us06web.zoom.us/meeting/register/tZMlc-mqrjIsH912-HTRe-AaT-pp41bDe81a#/registration) and bring your questions for live Q&A!
-
----
-
 Meilisearch helps you shape a delightful search experience in a snap, offering features that work out-of-the-box to speed up your workflow.
 
 <p align="center" name="demo">

benchmarks/benches/utils.rs

@@ -12,7 +12,7 @@ use milli::heed::EnvOpenOptions;
 use milli::update::{
     IndexDocuments, IndexDocumentsConfig, IndexDocumentsMethod, IndexerConfig, Settings,
 };
-use milli::{Criterion, Filter, Index, Object, TermsMatchingStrategy};
+use milli::{Filter, Index, Object, RankingRule, TermsMatchingStrategy};
 use serde_json::Value;
 
 pub struct Conf<'a> {
@@ -78,11 +78,11 @@ pub fn base_setup(conf: &Conf) -> Index {
 
     if let Some(criterion) = conf.criterion {
         builder.reset_filterable_fields();
-        builder.reset_criteria();
+        builder.reset_ranking_rules();
         builder.reset_stop_words();
 
-        let criterion = criterion.iter().map(|s| Criterion::from_str(s).unwrap()).collect();
-        builder.set_criteria(criterion);
+        let criterion = criterion.iter().map(|s| RankingRule::from_str(s).unwrap()).collect();
+        builder.set_ranking_rules(criterion);
     }
 
     (conf.configure)(&mut builder);

dump/src/reader/mod.rs

@@ -526,12 +526,12 @@ pub(crate) mod test {
         assert!(indexes.is_empty());
 
         // products
-        insta::assert_json_snapshot!(products.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(products.metadata(), @r###"
         {
           "uid": "products",
           "primaryKey": "sku",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2022-10-09T20:27:22.688964637Z",
+          "updatedAt": "2022-10-09T20:27:23.951017769Z"
         }
         "###);
 
@@ -541,12 +541,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"548284a84de510f71e88e6cdea495cf5");
 
         // movies
-        insta::assert_json_snapshot!(movies.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(movies.metadata(), @r###"
         {
           "uid": "movies",
           "primaryKey": "id",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2022-10-09T20:27:22.197788495Z",
+          "updatedAt": "2022-10-09T20:28:01.93111053Z"
         }
         "###);
 
@@ -571,12 +571,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"d751713988987e9331980363e24189ce");
 
         // spells
-        insta::assert_json_snapshot!(spells.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(spells.metadata(), @r###"
         {
           "uid": "dnd_spells",
           "primaryKey": "index",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2022-10-09T20:27:24.242683494Z",
+          "updatedAt": "2022-10-09T20:27:24.312809641Z"
         }
         "###);
 
@@ -617,12 +617,12 @@ pub(crate) mod test {
         assert!(indexes.is_empty());
 
         // products
-        insta::assert_json_snapshot!(products.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(products.metadata(), @r###"
         {
           "uid": "products",
           "primaryKey": "sku",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2023-01-30T16:25:56.595257Z",
+          "updatedAt": "2023-01-30T16:25:58.70348Z"
         }
         "###);
 
@@ -632,12 +632,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"548284a84de510f71e88e6cdea495cf5");
 
         // movies
-        insta::assert_json_snapshot!(movies.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(movies.metadata(), @r###"
         {
           "uid": "movies",
           "primaryKey": "id",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2023-01-30T16:25:56.192178Z",
+          "updatedAt": "2023-01-30T16:25:56.455714Z"
         }
         "###);
 
@@ -647,12 +647,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"0227598af846e574139ee0b80e03a720");
 
         // spells
-        insta::assert_json_snapshot!(spells.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(spells.metadata(), @r###"
         {
           "uid": "dnd_spells",
           "primaryKey": "index",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2023-01-30T16:25:58.876405Z",
+          "updatedAt": "2023-01-30T16:25:59.079906Z"
         }
         "###);
 

dump/src/reader/v2/mod.rs

@@ -46,6 +46,7 @@ pub type Checked = settings::Checked;
 pub type Unchecked = settings::Unchecked;
 
 pub type Task = updates::UpdateEntry;
+pub type Kind = updates::UpdateMeta;
 
 // everything related to the errors
 pub type ResponseError = errors::ResponseError;
@@ -107,8 +108,11 @@ impl V2Reader {
     pub fn indexes(&self) -> Result<impl Iterator<Item = Result<V2IndexReader>> + '_> {
         Ok(self.index_uuid.iter().map(|index| -> Result<_> {
             V2IndexReader::new(
-                index.uid.clone(),
                 &self.dump.path().join("indexes").join(format!("index-{}", index.uuid)),
+                index,
+                BufReader::new(
+                    File::open(self.dump.path().join("updates").join("data.jsonl")).unwrap(),
+                ),
             )
         }))
     }
@@ -143,16 +147,41 @@ pub struct V2IndexReader {
 }
 
 impl V2IndexReader {
-    pub fn new(name: String, path: &Path) -> Result<Self> {
+    pub fn new(path: &Path, index_uuid: &IndexUuid, tasks: BufReader<File>) -> Result<Self> {
         let meta = File::open(path.join("meta.json"))?;
         let meta: DumpMeta = serde_json::from_reader(meta)?;
 
+        let mut created_at = None;
+        let mut updated_at = None;
+
+        for line in tasks.lines() {
+            let task: Task = serde_json::from_str(&line?)?;
+            if !(task.uuid == index_uuid.uuid && task.is_finished()) {
+                continue;
+            }
+
+            let new_created_at = match task.update.meta() {
+                Kind::DocumentsAddition { .. } | Kind::Settings(_) => task.update.finished_at(),
+                _ => None,
+            };
+            let new_updated_at = task.update.finished_at();
+
+            if created_at.is_none() || created_at > new_created_at {
+                created_at = new_created_at;
+            }
+
+            if updated_at.is_none() || updated_at < new_updated_at {
+                updated_at = new_updated_at;
+            }
+        }
+
+        let current_time = OffsetDateTime::now_utc();
+
         let metadata = IndexMetadata {
-            uid: name,
+            uid: index_uuid.uid.clone(),
             primary_key: meta.primary_key,
-            // FIXME: Iterate over the whole task queue to find the creation and last update date.
-            created_at: OffsetDateTime::now_utc(),
-            updated_at: OffsetDateTime::now_utc(),
+            created_at: created_at.unwrap_or(current_time),
+            updated_at: updated_at.unwrap_or(current_time),
         };
 
         let ret = V2IndexReader {
@@ -248,12 +277,12 @@ pub(crate) mod test {
         assert!(indexes.is_empty());
 
         // products
-        insta::assert_json_snapshot!(products.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(products.metadata(), @r###"
         {
           "uid": "products",
           "primaryKey": "sku",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2022-10-09T20:27:22.688964637Z",
+          "updatedAt": "2022-10-09T20:27:23.951017769Z"
         }
         "###);
 
@@ -263,12 +292,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"548284a84de510f71e88e6cdea495cf5");
 
         // movies
-        insta::assert_json_snapshot!(movies.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(movies.metadata(), @r###"
         {
           "uid": "movies",
           "primaryKey": "id",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2022-10-09T20:27:22.197788495Z",
+          "updatedAt": "2022-10-09T20:28:01.93111053Z"
         }
         "###);
 
@@ -293,12 +322,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"d751713988987e9331980363e24189ce");
 
         // spells
-        insta::assert_json_snapshot!(spells.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(spells.metadata(), @r###"
         {
           "uid": "dnd_spells",
           "primaryKey": "index",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2022-10-09T20:27:24.242683494Z",
+          "updatedAt": "2022-10-09T20:27:24.312809641Z"
         }
         "###);
 
@@ -340,12 +369,12 @@ pub(crate) mod test {
         assert!(indexes.is_empty());
 
         // products
-        insta::assert_json_snapshot!(products.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(products.metadata(), @r###"
         {
           "uid": "products",
           "primaryKey": "sku",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2023-01-30T16:25:56.595257Z",
+          "updatedAt": "2023-01-30T16:25:58.70348Z"
         }
         "###);
 
@@ -355,12 +384,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"548284a84de510f71e88e6cdea495cf5");
 
         // movies
-        insta::assert_json_snapshot!(movies.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(movies.metadata(), @r###"
         {
           "uid": "movies",
           "primaryKey": "id",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2023-01-30T16:25:56.192178Z",
+          "updatedAt": "2023-01-30T16:25:56.455714Z"
         }
         "###);
 
@@ -370,12 +399,12 @@ pub(crate) mod test {
         meili_snap::snapshot_hash!(format!("{:#?}", documents), @"0227598af846e574139ee0b80e03a720");
 
         // spells
-        insta::assert_json_snapshot!(spells.metadata(), { ".createdAt" => "[now]", ".updatedAt" => "[now]" }, @r###"
+        insta::assert_json_snapshot!(spells.metadata(), @r###"
         {
           "uid": "dnd_spells",
           "primaryKey": "index",
-          "createdAt": "[now]",
-          "updatedAt": "[now]"
+          "createdAt": "2023-01-30T16:25:58.876405Z",
+          "updatedAt": "2023-01-30T16:25:59.079906Z"
         }
         "###);
 

dump/src/reader/v2/updates.rs

@@ -227,4 +227,14 @@ impl UpdateStatus {
             _ => None,
         }
     }
+
+    pub fn finished_at(&self) -> Option<OffsetDateTime> {
+        match self {
+            UpdateStatus::Processing(_) => None,
+            UpdateStatus::Enqueued(_) => None,
+            UpdateStatus::Processed(u) => Some(u.processed_at),
+            UpdateStatus::Aborted(_) => None,
+            UpdateStatus::Failed(u) => Some(u.failed_at),
+        }
+    }
 }

index-scheduler/src/batch.rs

@@ -923,10 +923,6 @@ impl IndexScheduler {
                     self.index_mapper.index(&rtxn, &index_uid)?
                 };
 
-                // the index operation can take a long time, so save this handle to make it available to the search for the duration of the tick
-                *self.currently_updating_index.write().unwrap() =
-                    Some((index_uid.clone(), index.clone()));
-
                 let mut index_wtxn = index.write_txn()?;
                 let tasks = self.apply_index_operation(&mut index_wtxn, &index, op)?;
                 index_wtxn.commit()?;

index-scheduler/src/insta_snapshot.rs

@@ -39,7 +39,6 @@ pub fn snapshot_index_scheduler(scheduler: &IndexScheduler) -> String {
         test_breakpoint_sdr: _,
         planned_failures: _,
         run_loop_iteration: _,
-        currently_updating_index: _,
     } = scheduler;
 
     let rtxn = env.read_txn().unwrap();

index-scheduler/src/lib.rs

@@ -27,7 +27,7 @@ mod index_mapper;
 mod insta_snapshot;
 mod lru;
 mod utils;
-pub mod uuid_codec;
+mod uuid_codec;
 
 pub type Result<T> = std::result::Result<T, Error>;
 pub type TaskId = u32;
@@ -331,10 +331,6 @@ pub struct IndexScheduler {
     /// The path to the version file of Meilisearch.
     pub(crate) version_file_path: PathBuf,
 
-    /// A few types of long running batches of tasks that act on a single index set this field
-    /// so that a handle to the index is available from other threads (search) in an optimized manner.
-    currently_updating_index: Arc<RwLock<Option<(String, Index)>>>,
-
     // ================= test
     // The next entry is dedicated to the tests.
     /// Provide a way to set a breakpoint in multiple part of the scheduler.
@@ -378,7 +374,6 @@ impl IndexScheduler {
             dumps_path: self.dumps_path.clone(),
             auth_path: self.auth_path.clone(),
             version_file_path: self.version_file_path.clone(),
-            currently_updating_index: self.currently_updating_index.clone(),
             #[cfg(test)]
             test_breakpoint_sdr: self.test_breakpoint_sdr.clone(),
             #[cfg(test)]
@@ -475,7 +470,6 @@ impl IndexScheduler {
             snapshots_path: options.snapshots_path,
             auth_path: options.auth_path,
             version_file_path: options.version_file_path,
-            currently_updating_index: Arc::new(RwLock::new(None)),
 
             #[cfg(test)]
             test_breakpoint_sdr,
@@ -658,13 +652,6 @@ impl IndexScheduler {
     /// If you need to fetch information from or perform an action on all indexes,
     /// see the `try_for_each_index` function.
     pub fn index(&self, name: &str) -> Result<Index> {
-        if let Some((current_name, current_index)) =
-            self.currently_updating_index.read().unwrap().as_ref()
-        {
-            if current_name == name {
-                return Ok(current_index.clone());
-            }
-        }
         let rtxn = self.env.read_txn()?;
         self.index_mapper.index(&rtxn, name)
     }
@@ -1146,9 +1133,6 @@ impl IndexScheduler {
             handle.join().unwrap_or(Err(Error::ProcessBatchPanicked))
         };
 
-        // Reset the currently updating index to relinquish the index handle
-        *self.currently_updating_index.write().unwrap() = None;
-
         #[cfg(test)]
         self.maybe_fail(tests::FailureLocation::AcquiringWtxn)?;
 

meilisearch-types/Cargo.toml

@@ -50,7 +50,6 @@ hebrew = ["milli/hebrew"]
 japanese = ["milli/japanese"]
 # thai specialized tokenization
 thai = ["milli/thai"]
+
 # allow greek specialized tokenization
 greek = ["milli/greek"]
-# allow khmer specialized tokenization
-khmer = ["milli/khmer"]

meilisearch-types/src/error.rs

@@ -235,6 +235,7 @@ InvalidSearchCropMarker               , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchFacets                   , InvalidRequest       , BAD_REQUEST ;
 InvalidFacetSearchFacetName           , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchFilter                   , InvalidRequest       , BAD_REQUEST ;
+InvalidSearchBoostingFilter           , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchHighlightPostTag         , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchHighlightPreTag          , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchHitsPerPage              , InvalidRequest       , BAD_REQUEST ;
@@ -331,6 +332,7 @@ impl ErrorCode for milli::Error {
                     UserError::MaxDatabaseSizeReached => Code::DatabaseSizeLimitReached,
                     UserError::AttributeLimitReached => Code::MaxFieldsLimitExceeded,
                     UserError::InvalidFilter(_) => Code::InvalidSearchFilter,
+                    UserError::InvalidBoostingFilter(_) => Code::InvalidSearchBoostingFilter,
                     UserError::InvalidFilterExpression(..) => Code::InvalidSearchFilter,
                     UserError::MissingDocumentId { .. } => Code::MissingDocumentId,
                     UserError::InvalidDocumentId { .. } | UserError::TooManyDocumentIds { .. } => {

meilisearch-types/src/settings.rs

@@ -9,7 +9,7 @@ use std::str::FromStr;
 use deserr::{DeserializeError, Deserr, ErrorKind, MergeWithError, ValuePointerRef};
 use fst::IntoStreamer;
 use milli::update::Setting;
-use milli::{Criterion, CriterionError, Index, DEFAULT_VALUES_PER_FACET};
+use milli::{Index, RankingRule, RankingRuleError, DEFAULT_VALUES_PER_FACET};
 use serde::{Deserialize, Serialize, Serializer};
 
 use crate::deserr::DeserrJsonError;
@@ -117,10 +117,10 @@ pub struct PaginationSettings {
     pub max_total_hits: Setting<usize>,
 }
 
-impl MergeWithError<milli::CriterionError> for DeserrJsonError<InvalidSettingsRankingRules> {
+impl MergeWithError<milli::RankingRuleError> for DeserrJsonError<InvalidSettingsRankingRules> {
     fn merge(
         _self_: Option<Self>,
-        other: milli::CriterionError,
+        other: milli::RankingRuleError,
         merge_location: ValuePointerRef,
     ) -> ControlFlow<Self, Self> {
         Self::error::<Infallible>(
@@ -344,9 +344,9 @@ pub fn apply_settings_to_builder(
 
     match settings.ranking_rules {
         Setting::Set(ref criteria) => {
-            builder.set_criteria(criteria.iter().map(|c| c.clone().into()).collect())
+            builder.set_ranking_rules(criteria.iter().map(|c| c.clone().into()).collect())
         }
-        Setting::Reset => builder.reset_criteria(),
+        Setting::Reset => builder.reset_ranking_rules(),
         Setting::NotSet => (),
     }
 
@@ -578,11 +578,13 @@ pub fn settings(
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, Deserr)]
-#[deserr(try_from(&String) = FromStr::from_str -> CriterionError)]
+#[deserr(try_from(&String) = FromStr::from_str -> RankingRuleError)]
 pub enum RankingRuleView {
     /// Sorted by decreasing number of matched query terms.
     /// Query words at the front of an attribute is considered better than if it was at the back.
     Words,
+    /// Sorted by documents matching the given filter and then documents not matching it.
+    Boost(String),
     /// Sorted by increasing number of typos.
     Typo,
     /// Sorted by increasing distance between matched query terms.
@@ -605,7 +607,7 @@ impl Serialize for RankingRuleView {
     where
         S: Serializer,
     {
-        serializer.serialize_str(&format!("{}", Criterion::from(self.clone())))
+        serializer.serialize_str(&format!("{}", RankingRule::from(self.clone())))
     }
 }
 impl<'de> Deserialize<'de> for RankingRuleView {
@@ -623,7 +625,7 @@ impl<'de> Deserialize<'de> for RankingRuleView {
             where
                 E: serde::de::Error,
             {
-                let criterion = Criterion::from_str(v).map_err(|_| {
+                let criterion = RankingRule::from_str(v).map_err(|_| {
                     E::invalid_value(serde::de::Unexpected::Str(v), &"a valid ranking rule")
                 })?;
                 Ok(RankingRuleView::from(criterion))
@@ -633,42 +635,44 @@ impl<'de> Deserialize<'de> for RankingRuleView {
     }
 }
 impl FromStr for RankingRuleView {
-    type Err = <Criterion as FromStr>::Err;
+    type Err = <RankingRule as FromStr>::Err;
 
     fn from_str(s: &str) -> Result<Self, Self::Err> {
-        Ok(RankingRuleView::from(Criterion::from_str(s)?))
+        Ok(RankingRuleView::from(RankingRule::from_str(s)?))
     }
 }
 impl fmt::Display for RankingRuleView {
     fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
-        fmt::Display::fmt(&Criterion::from(self.clone()), f)
+        fmt::Display::fmt(&RankingRule::from(self.clone()), f)
     }
 }
-impl From<Criterion> for RankingRuleView {
-    fn from(value: Criterion) -> Self {
+impl From<RankingRule> for RankingRuleView {
+    fn from(value: RankingRule) -> Self {
         match value {
-            Criterion::Words => RankingRuleView::Words,
-            Criterion::Typo => RankingRuleView::Typo,
-            Criterion::Proximity => RankingRuleView::Proximity,
-            Criterion::Attribute => RankingRuleView::Attribute,
-            Criterion::Sort => RankingRuleView::Sort,
-            Criterion::Exactness => RankingRuleView::Exactness,
-            Criterion::Asc(x) => RankingRuleView::Asc(x),
-            Criterion::Desc(x) => RankingRuleView::Desc(x),
+            RankingRule::Words => RankingRuleView::Words,
+            RankingRule::FilterBoosting(filter) => RankingRuleView::Boost(filter),
+            RankingRule::Typo => RankingRuleView::Typo,
+            RankingRule::Proximity => RankingRuleView::Proximity,
+            RankingRule::Attribute => RankingRuleView::Attribute,
+            RankingRule::Sort => RankingRuleView::Sort,
+            RankingRule::Exactness => RankingRuleView::Exactness,
+            RankingRule::Asc(x) => RankingRuleView::Asc(x),
+            RankingRule::Desc(x) => RankingRuleView::Desc(x),
         }
     }
 }
-impl From<RankingRuleView> for Criterion {
+impl From<RankingRuleView> for RankingRule {
     fn from(value: RankingRuleView) -> Self {
         match value {
-            RankingRuleView::Words => Criterion::Words,
-            RankingRuleView::Typo => Criterion::Typo,
-            RankingRuleView::Proximity => Criterion::Proximity,
-            RankingRuleView::Attribute => Criterion::Attribute,
-            RankingRuleView::Sort => Criterion::Sort,
-            RankingRuleView::Exactness => Criterion::Exactness,
-            RankingRuleView::Asc(x) => Criterion::Asc(x),
-            RankingRuleView::Desc(x) => Criterion::Desc(x),
+            RankingRuleView::Words => RankingRule::Words,
+            RankingRuleView::Boost(filter) => RankingRule::FilterBoosting(filter),
+            RankingRuleView::Typo => RankingRule::Typo,
+            RankingRuleView::Proximity => RankingRule::Proximity,
+            RankingRuleView::Attribute => RankingRule::Attribute,
+            RankingRuleView::Sort => RankingRule::Sort,
+            RankingRuleView::Exactness => RankingRule::Exactness,
+            RankingRuleView::Asc(x) => RankingRule::Asc(x),
+            RankingRuleView::Desc(x) => RankingRule::Desc(x),
         }
     }
 }

meilisearch/Cargo.toml

@@ -150,7 +150,6 @@ hebrew = ["meilisearch-types/hebrew"]
 japanese = ["meilisearch-types/japanese"]
 thai = ["meilisearch-types/thai"]
 greek = ["meilisearch-types/greek"]
-khmer = ["meilisearch-types/khmer"]
 
 [package.metadata.mini-dashboard]
 assets-url = "https://github.com/meilisearch/mini-dashboard/releases/download/v0.2.11/build.zip"

meilisearch/src/analytics/segment_analytics.rs

@@ -629,6 +629,7 @@ impl SearchAggregator {
             show_ranking_score,
             show_ranking_score_details,
             filter,
+            boosting_filter,
             sort,
             facets: _,
             highlight_pre_tag,
@@ -1002,6 +1003,7 @@ impl MultiSearchAggregator {
                     show_ranking_score_details: _,
                     show_matches_position: _,
                     filter: _,
+                    boosting_filter: _,
                     sort: _,
                     facets: _,
                     highlight_pre_tag: _,

meilisearch/src/routes/indexes/facet_search.rs

@@ -111,6 +111,7 @@ impl From<FacetSearchQuery> for SearchQuery {
             show_ranking_score: false,
             show_ranking_score_details: false,
             filter,
+            boosting_filter: None,
             sort: None,
             facets: None,
             highlight_pre_tag: DEFAULT_HIGHLIGHT_PRE_TAG(),

meilisearch/src/routes/indexes/search.rs

@@ -54,6 +54,8 @@ pub struct SearchQueryGet {
     attributes_to_highlight: Option<CS<String>>,
     #[deserr(default, error = DeserrQueryParamError<InvalidSearchFilter>)]
     filter: Option<String>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidSearchBoostingFilter>)]
+    boosting_filter: Option<String>,
     #[deserr(default, error = DeserrQueryParamError<InvalidSearchSort>)]
     sort: Option<String>,
     #[deserr(default, error = DeserrQueryParamError<InvalidSearchShowMatchesPosition>)]
@@ -86,6 +88,14 @@ impl From<SearchQueryGet> for SearchQuery {
             None => None,
         };
 
+        let boosting_filter = match other.boosting_filter {
+            Some(f) => match serde_json::from_str(&f) {
+                Ok(v) => Some(v),
+                _ => Some(Value::String(f)),
+            },
+            None => None,
+        };
+
         Self {
             q: other.q,
             vector: other.vector.map(CS::into_inner),
@@ -98,6 +108,7 @@ impl From<SearchQueryGet> for SearchQuery {
             crop_length: other.crop_length.0,
             attributes_to_highlight: other.attributes_to_highlight.map(|o| o.into_iter().collect()),
             filter,
+            boosting_filter,
             sort: other.sort.map(|attr| fix_sort_query_parameters(&attr)),
             show_matches_position: other.show_matches_position.0,
             show_ranking_score: other.show_ranking_score.0,

meilisearch/src/search.rs

@@ -71,6 +71,8 @@ pub struct SearchQuery {
     pub show_ranking_score_details: bool,
     #[deserr(default, error = DeserrJsonError<InvalidSearchFilter>)]
     pub filter: Option<Value>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchBoostingFilter>)]
+    pub boosting_filter: Option<Value>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchSort>)]
     pub sort: Option<Vec<String>>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchFacets>)]
@@ -130,6 +132,8 @@ pub struct SearchQueryWithIndex {
     pub show_matches_position: bool,
     #[deserr(default, error = DeserrJsonError<InvalidSearchFilter>)]
     pub filter: Option<Value>,
+    #[deserr(default, error = DeserrJsonError<InvalidSearchBoostingFilter>)]
+    pub boosting_filter: Option<Value>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchSort>)]
     pub sort: Option<Vec<String>>,
     #[deserr(default, error = DeserrJsonError<InvalidSearchFacets>)]
@@ -164,6 +168,7 @@ impl SearchQueryWithIndex {
             show_ranking_score_details,
             show_matches_position,
             filter,
+            boosting_filter,
             sort,
             facets,
             highlight_pre_tag,
@@ -189,6 +194,7 @@ impl SearchQueryWithIndex {
                 show_ranking_score_details,
                 show_matches_position,
                 filter,
+                boosting_filter,
                 sort,
                 facets,
                 highlight_pre_tag,
@@ -397,8 +403,14 @@ fn prepare_search<'t>(
     search.limit(limit);
 
     if let Some(ref filter) = query.filter {
-        if let Some(facets) = parse_filter(filter)? {
-            search.filter(facets);
+        if let Some(filter) = parse_filter(filter)? {
+            search.filter(filter);
+        }
+    }
+
+    if let Some(ref boosting_filter) = query.boosting_filter {
+        if let Some(boosting_filter) = parse_filter(boosting_filter)? {
+            search.boosting_filter(boosting_filter);
         }
     }
 

meilisearch/tests/common/mod.rs

@@ -5,11 +5,9 @@ pub mod service;
 
 use std::fmt::{self, Display};
 
-#[allow(unused)]
 pub use index::{GetAllDocumentsOptions, GetDocumentOptions};
 use meili_snap::json_string;
 use serde::{Deserialize, Serialize};
-#[allow(unused)]
 pub use server::{default_settings, Server};
 
 #[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq)]

meilisearch/tests/search/distinct.rs

@@ -6,109 +6,21 @@ use crate::json;
 
 pub(self) static DOCUMENTS: Lazy<Value> = Lazy::new(|| {
     json!([
-      {
-        "id": 1,
-        "description": "Leather Jacket",
-        "brand": "Lee Jeans",
-        "product_id": "123456",
-        "color": "Brown"
-      },
-      {
-        "id": 2,
-        "description": "Leather Jacket",
-        "brand": "Lee Jeans",
-        "product_id": "123456",
-        "color": "Black"
-      },
-      {
-        "id": 3,
-        "description": "Leather Jacket",
-        "brand": "Lee Jeans",
-        "product_id": "123456",
-        "color": "Blue"
-      },
-      {
-        "id": 4,
-        "description": "T-Shirt",
-        "brand": "Nike",
-        "product_id": "789012",
-        "color": "Red"
-      },
-      {
-        "id": 5,
-        "description": "T-Shirt",
-        "brand": "Nike",
-        "product_id": "789012",
-        "color": "Blue"
-      },
-      {
-        "id": 6,
-        "description": "Running Shoes",
-        "brand": "Adidas",
-        "product_id": "456789",
-        "color": "Black"
-      },
-      {
-        "id": 7,
-        "description": "Running Shoes",
-        "brand": "Adidas",
-        "product_id": "456789",
-        "color": "White"
-      },
-      {
-        "id": 8,
-        "description": "Hoodie",
-        "brand": "Puma",
-        "product_id": "987654",
-        "color": "Gray"
-      },
-      {
-        "id": 9,
-        "description": "Sweater",
-        "brand": "Gap",
-        "product_id": "234567",
-        "color": "Green"
-      },
-      {
-        "id": 10,
-        "description": "Sweater",
-        "brand": "Gap",
-        "product_id": "234567",
-        "color": "Red"
-      },
-      {
-        "id": 11,
-        "description": "Sweater",
-        "brand": "Gap",
-        "product_id": "234567",
-        "color": "Blue"
-      },
-      {
-        "id": 12,
-        "description": "Jeans",
-        "brand": "Levi's",
-        "product_id": "345678",
-        "color": "Indigo"
-      },
-      {
-        "id": 13,
-        "description": "Jeans",
-        "brand": "Levi's",
-        "product_id": "345678",
-        "color": "Black"
-      },
-      {
-        "id": 14,
-        "description": "Jeans",
-        "brand": "Levi's",
-        "product_id": "345678",
-        "color": "Stone Wash"
-      }
+        {"productId": 1, "shopId": 1},
+        {"productId": 2, "shopId": 1},
+        {"productId": 3, "shopId": 2},
+        {"productId": 4, "shopId": 2},
+        {"productId": 5, "shopId": 3},
+        {"productId": 6, "shopId": 3},
+        {"productId": 7, "shopId": 4},
+        {"productId": 8, "shopId": 4},
+        {"productId": 9, "shopId": 5},
+        {"productId": 10, "shopId": 5}
     ])
 });
 
-pub(self) static DOCUMENT_PRIMARY_KEY: &str = "id";
-pub(self) static DOCUMENT_DISTINCT_KEY: &str = "product_id";
+pub(self) static DOCUMENT_PRIMARY_KEY: &str = "productId";
+pub(self) static DOCUMENT_DISTINCT_KEY: &str = "shopId";
 
 /// testing: https://github.com/meilisearch/meilisearch/issues/4078
 #[actix_rt::test]
@@ -121,121 +33,31 @@ async fn distinct_search_with_offset_no_ranking() {
     index.update_distinct_attribute(json!(DOCUMENT_DISTINCT_KEY)).await;
     index.wait_task(1).await;
 
-    fn get_hits(response: &Value) -> Vec<&str> {
+    fn get_hits(Value(response): Value) -> Vec<i64> {
         let hits_array = response["hits"].as_array().unwrap();
-        hits_array.iter().map(|h| h[DOCUMENT_DISTINCT_KEY].as_str().unwrap()).collect::<Vec<_>>()
+        hits_array.iter().map(|h| h[DOCUMENT_DISTINCT_KEY].as_i64().unwrap()).collect::<Vec<_>>()
     }
 
-    let (response, code) = index.search_post(json!({"offset": 0, "limit": 2})).await;
-    let hits = get_hits(&response);
+    let (response, code) = index.search_post(json!({"limit": 2, "offset": 0})).await;
+    let hits = get_hits(response);
     snapshot!(code, @"200 OK");
     snapshot!(hits.len(), @"2");
-    snapshot!(format!("{:?}", hits), @r#"["123456", "789012"]"#);
-    snapshot!(response["estimatedTotalHits"] , @"11");
+    snapshot!(format!("{:?}", hits), @"[1, 2]");
 
-    let (response, code) = index.search_post(json!({"offset": 2, "limit": 2})).await;
-    let hits = get_hits(&response);
+    let (response, code) = index.search_post(json!({"limit": 2, "offset": 2})).await;
+    let hits = get_hits(response);
     snapshot!(code, @"200 OK");
     snapshot!(hits.len(), @"2");
-    snapshot!(format!("{:?}", hits), @r#"["456789", "987654"]"#);
-    snapshot!(response["estimatedTotalHits"], @"10");
+    snapshot!(format!("{:?}", hits), @"[3, 4]");
 
-    let (response, code) = index.search_post(json!({"offset": 4, "limit": 2})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"2");
-    snapshot!(format!("{:?}", hits), @r#"["234567", "345678"]"#);
-    snapshot!(response["estimatedTotalHits"], @"6");
-
-    let (response, code) = index.search_post(json!({"offset": 5, "limit": 2})).await;
-    let hits = get_hits(&response);
+    let (response, code) = index.search_post(json!({"limit": 10, "offset": 4})).await;
+    let hits = get_hits(response);
     snapshot!(code, @"200 OK");
     snapshot!(hits.len(), @"1");
-    snapshot!(format!("{:?}", hits), @r#"["345678"]"#);
-    snapshot!(response["estimatedTotalHits"], @"6");
+    snapshot!(format!("{:?}", hits), @"[5]");
 
-    let (response, code) = index.search_post(json!({"offset": 6, "limit": 2})).await;
-    let hits = get_hits(&response);
+    let (response, code) = index.search_post(json!({"limit": 10, "offset": 5})).await;
+    let hits = get_hits(response);
     snapshot!(code, @"200 OK");
     snapshot!(hits.len(), @"0");
-    snapshot!(format!("{:?}", hits), @r#"[]"#);
-    snapshot!(response["estimatedTotalHits"], @"6");
-
-    let (response, code) = index.search_post(json!({"offset": 7, "limit": 2})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"0");
-    snapshot!(format!("{:?}", hits), @r#"[]"#);
-    snapshot!(response["estimatedTotalHits"], @"6");
-}
-
-/// testing: https://github.com/meilisearch/meilisearch/issues/4130
-#[actix_rt::test]
-async fn distinct_search_with_pagination_no_ranking() {
-    let server = Server::new().await;
-    let index = server.index("test");
-
-    let documents = DOCUMENTS.clone();
-    index.add_documents(documents, Some(DOCUMENT_PRIMARY_KEY)).await;
-    index.update_distinct_attribute(json!(DOCUMENT_DISTINCT_KEY)).await;
-    index.wait_task(1).await;
-
-    fn get_hits(response: &Value) -> Vec<&str> {
-        let hits_array = response["hits"].as_array().unwrap();
-        hits_array.iter().map(|h| h[DOCUMENT_DISTINCT_KEY].as_str().unwrap()).collect::<Vec<_>>()
-    }
-
-    let (response, code) = index.search_post(json!({"page": 0, "hitsPerPage": 2})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"0");
-    snapshot!(format!("{:?}", hits), @r#"[]"#);
-    snapshot!(response["page"], @"0");
-    snapshot!(response["totalPages"], @"3");
-    snapshot!(response["totalHits"], @"6");
-
-    let (response, code) = index.search_post(json!({"page": 1, "hitsPerPage": 2})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"2");
-    snapshot!(format!("{:?}", hits), @r#"["123456", "789012"]"#);
-    snapshot!(response["page"], @"1");
-    snapshot!(response["totalPages"], @"3");
-    snapshot!(response["totalHits"], @"6");
-
-    let (response, code) = index.search_post(json!({"page": 2, "hitsPerPage": 2})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"2");
-    snapshot!(format!("{:?}", hits), @r#"["456789", "987654"]"#);
-    snapshot!(response["page"], @"2");
-    snapshot!(response["totalPages"], @"3");
-    snapshot!(response["totalHits"], @"6");
-
-    let (response, code) = index.search_post(json!({"page": 3, "hitsPerPage": 2})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"2");
-    snapshot!(format!("{:?}", hits), @r#"["234567", "345678"]"#);
-    snapshot!(response["page"], @"3");
-    snapshot!(response["totalPages"], @"3");
-    snapshot!(response["totalHits"], @"6");
-
-    let (response, code) = index.search_post(json!({"page": 4, "hitsPerPage": 2})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"0");
-    snapshot!(format!("{:?}", hits), @r#"[]"#);
-    snapshot!(response["page"], @"4");
-    snapshot!(response["totalPages"], @"3");
-    snapshot!(response["totalHits"], @"6");
-
-    let (response, code) = index.search_post(json!({"page": 2, "hitsPerPage": 3})).await;
-    let hits = get_hits(&response);
-    snapshot!(code, @"200 OK");
-    snapshot!(hits.len(), @"3");
-    snapshot!(format!("{:?}", hits), @r#"["987654", "234567", "345678"]"#);
-    snapshot!(response["page"], @"2");
-    snapshot!(response["totalPages"], @"2");
-    snapshot!(response["totalHits"], @"6");
 }

meilitool/Cargo.toml

@@ -1,19 +0,0 @@
-[package]
-name = "meilitool"
-description = "A CLI to edit a Meilisearch database from the command line"
-version.workspace = true
-authors.workspace = true
-homepage.workspace = true
-readme.workspace = true
-edition.workspace = true
-license.workspace = true
-
-[dependencies]
-anyhow = "1.0.75"
-clap = { version = "4.2.1", features = ["derive"] }
-dump = { path = "../dump" }
-file-store = { path = "../file-store" }
-meilisearch-auth = { path = "../meilisearch-auth" }
-meilisearch-types = { path = "../meilisearch-types" }
-time = { version = "0.3.30", features = ["formatting"] }
-uuid = { version = "1.5.0", features = ["v4"], default-features = false }

meilitool/src/main.rs

@@ -1,312 +0,0 @@
-use std::fs::{read_dir, read_to_string, remove_file, File};
-use std::io::BufWriter;
-use std::path::PathBuf;
-
-use anyhow::Context;
-use clap::{Parser, Subcommand};
-use dump::{DumpWriter, IndexMetadata};
-use file_store::FileStore;
-use meilisearch_auth::AuthController;
-use meilisearch_types::heed::types::{OwnedType, SerdeJson, Str};
-use meilisearch_types::heed::{Database, Env, EnvOpenOptions, PolyDatabase, RoTxn, RwTxn};
-use meilisearch_types::milli::documents::{obkv_to_object, DocumentsBatchReader};
-use meilisearch_types::milli::{obkv_to_json, BEU32};
-use meilisearch_types::tasks::{Status, Task};
-use meilisearch_types::versioning::check_version_file;
-use meilisearch_types::Index;
-use time::macros::format_description;
-use time::OffsetDateTime;
-use uuid_codec::UuidCodec;
-
-mod uuid_codec;
-
-#[derive(Parser)]
-#[command(author, version, about, long_about = None)]
-struct Cli {
-    /// The database path where the Meilisearch is running.
-    #[arg(long, default_value = "data.ms/")]
-    db_path: PathBuf,
-
-    #[command(subcommand)]
-    command: Command,
-}
-
-#[derive(Subcommand)]
-enum Command {
-    /// Clears the task queue and make it empty.
-    ///
-    /// This command can be safely executed even if Meilisearch is running and processing tasks.
-    /// Once the task queue is empty you can restart Meilisearch and no more tasks must be visible,
-    /// even the ones that were processing. However, it's highly possible that you see the processing
-    /// tasks in the queue again with an associated internal error message.
-    ClearTaskQueue,
-
-    /// Exports a dump from the Meilisearch database.
-    ///
-    /// Make sure to run this command when Meilisearch is not running or running but not processing tasks.
-    /// If tasks are being processed while a dump is being exported there are chances for the dump to be
-    /// malformed with missing tasks.
-    ///
-    /// TODO Verify this claim or make sure it cannot happen and we can export dumps
-    ///      without caring about killing Meilisearch first!
-    ExportADump {
-        /// The directory in which the dump will be created.
-        #[arg(long, default_value = "dumps/")]
-        dump_dir: PathBuf,
-
-        /// Skip dumping the enqueued or processing tasks.
-        ///
-        /// Can be useful when there are a lot of them and it is not particularly useful
-        /// to keep them. Note that only the enqueued tasks takes up space so skipping
-        /// the processed ones is not particularly interesting.
-        #[arg(long)]
-        skip_enqueued_tasks: bool,
-    },
-}
-
-fn main() -> anyhow::Result<()> {
-    let Cli { db_path, command } = Cli::parse();
-
-    check_version_file(&db_path).context("While checking the version file")?;
-
-    match command {
-        Command::ClearTaskQueue => clear_task_queue(db_path),
-        Command::ExportADump { dump_dir, skip_enqueued_tasks } => {
-            export_a_dump(db_path, dump_dir, skip_enqueued_tasks)
-        }
-    }
-}
-
-/// Clears the task queue located at `db_path`.
-fn clear_task_queue(db_path: PathBuf) -> anyhow::Result<()> {
-    let path = db_path.join("tasks");
-    let env = EnvOpenOptions::new()
-        .max_dbs(100)
-        .open(&path)
-        .with_context(|| format!("While trying to open {:?}", path.display()))?;
-
-    eprintln!("Deleting tasks from the database...");
-
-    let mut wtxn = env.write_txn()?;
-    let all_tasks = try_opening_poly_database(&env, &wtxn, "all-tasks")?;
-    let total = all_tasks.len(&wtxn)?;
-    let status = try_opening_poly_database(&env, &wtxn, "status")?;
-    let kind = try_opening_poly_database(&env, &wtxn, "kind")?;
-    let index_tasks = try_opening_poly_database(&env, &wtxn, "index-tasks")?;
-    let canceled_by = try_opening_poly_database(&env, &wtxn, "canceled_by")?;
-    let enqueued_at = try_opening_poly_database(&env, &wtxn, "enqueued-at")?;
-    let started_at = try_opening_poly_database(&env, &wtxn, "started-at")?;
-    let finished_at = try_opening_poly_database(&env, &wtxn, "finished-at")?;
-
-    try_clearing_poly_database(&mut wtxn, all_tasks, "all-tasks")?;
-    try_clearing_poly_database(&mut wtxn, status, "status")?;
-    try_clearing_poly_database(&mut wtxn, kind, "kind")?;
-    try_clearing_poly_database(&mut wtxn, index_tasks, "index-tasks")?;
-    try_clearing_poly_database(&mut wtxn, canceled_by, "canceled_by")?;
-    try_clearing_poly_database(&mut wtxn, enqueued_at, "enqueued-at")?;
-    try_clearing_poly_database(&mut wtxn, started_at, "started-at")?;
-    try_clearing_poly_database(&mut wtxn, finished_at, "finished-at")?;
-
-    wtxn.commit().context("While committing the transaction")?;
-
-    eprintln!("Successfully deleted {total} tasks from the tasks database!");
-    eprintln!("Deleting the content files from disk...");
-
-    let mut count = 0usize;
-    let update_files = db_path.join("update_files");
-    let entries = read_dir(&update_files).with_context(|| {
-        format!("While trying to read the content of {:?}", update_files.display())
-    })?;
-    for result in entries {
-        match result {
-            Ok(ent) => match remove_file(ent.path()) {
-                Ok(_) => count += 1,
-                Err(e) => eprintln!("Error while deleting {:?}: {}", ent.path().display(), e),
-            },
-            Err(e) => {
-                eprintln!("Error while reading a file in {:?}: {}", update_files.display(), e)
-            }
-        }
-    }
-
-    eprintln!("Sucessfully deleted {count} content files from disk!");
-
-    Ok(())
-}
-
-fn try_opening_database<KC: 'static, DC: 'static>(
-    env: &Env,
-    rtxn: &RoTxn,
-    db_name: &str,
-) -> anyhow::Result<Database<KC, DC>> {
-    env.open_database(rtxn, Some(db_name))
-        .with_context(|| format!("While opening the {db_name:?} database"))?
-        .with_context(|| format!("Missing the {db_name:?} database"))
-}
-
-fn try_opening_poly_database(
-    env: &Env,
-    rtxn: &RoTxn,
-    db_name: &str,
-) -> anyhow::Result<PolyDatabase> {
-    env.open_poly_database(rtxn, Some(db_name))
-        .with_context(|| format!("While opening the {db_name:?} poly database"))?
-        .with_context(|| format!("Missing the {db_name:?} poly database"))
-}
-
-fn try_clearing_poly_database(
-    wtxn: &mut RwTxn,
-    database: PolyDatabase,
-    db_name: &str,
-) -> anyhow::Result<()> {
-    database.clear(wtxn).with_context(|| format!("While clearing the {db_name:?} database"))
-}
-
-/// Exports a dump into the dump directory.
-fn export_a_dump(
-    db_path: PathBuf,
-    dump_dir: PathBuf,
-    skip_enqueued_tasks: bool,
-) -> Result<(), anyhow::Error> {
-    let started_at = OffsetDateTime::now_utc();
-
-    // 1. Extracts the instance UID from disk
-    let instance_uid_path = db_path.join("instance-uid");
-    let instance_uid = match read_to_string(&instance_uid_path) {
-        Ok(content) => match content.trim().parse() {
-            Ok(uuid) => Some(uuid),
-            Err(e) => {
-                eprintln!("Impossible to parse instance-uid: {e}");
-                None
-            }
-        },
-        Err(e) => {
-            eprintln!("Impossible to read {}: {}", instance_uid_path.display(), e);
-            None
-        }
-    };
-
-    let dump = DumpWriter::new(instance_uid).context("While creating a new dump")?;
-    let file_store =
-        FileStore::new(db_path.join("update_files")).context("While opening the FileStore")?;
-
-    let index_scheduler_path = db_path.join("tasks");
-    let env = EnvOpenOptions::new()
-        .max_dbs(100)
-        .open(&index_scheduler_path)
-        .with_context(|| format!("While trying to open {:?}", index_scheduler_path.display()))?;
-
-    eprintln!("Dumping the keys...");
-
-    // 2. dump the keys
-    let auth_store = AuthController::new(&db_path, &None)
-        .with_context(|| format!("While opening the auth store at {}", db_path.display()))?;
-    let mut dump_keys = dump.create_keys()?;
-    let mut count = 0;
-    for key in auth_store.list_keys()? {
-        dump_keys.push_key(&key)?;
-        count += 1;
-    }
-    dump_keys.flush()?;
-
-    eprintln!("Successfully dumped {count} keys!");
-
-    let rtxn = env.read_txn()?;
-    let all_tasks: Database<OwnedType<BEU32>, SerdeJson<Task>> =
-        try_opening_database(&env, &rtxn, "all-tasks")?;
-    let index_mapping: Database<Str, UuidCodec> =
-        try_opening_database(&env, &rtxn, "index-mapping")?;
-
-    if skip_enqueued_tasks {
-        eprintln!("Skip dumping the enqueued tasks...");
-    } else {
-        eprintln!("Dumping the enqueued tasks...");
-
-        // 3. dump the tasks
-        let mut dump_tasks = dump.create_tasks_queue()?;
-        let mut count = 0;
-        for ret in all_tasks.iter(&rtxn)? {
-            let (_, t) = ret?;
-            let status = t.status;
-            let content_file = t.content_uuid();
-            let mut dump_content_file = dump_tasks.push_task(&t.into())?;
-
-            // 3.1. Dump the `content_file` associated with the task if there is one and the task is not finished yet.
-            if let Some(content_file_uuid) = content_file {
-                if status == Status::Enqueued {
-                    let content_file = file_store.get_update(content_file_uuid)?;
-
-                    let reader =
-                        DocumentsBatchReader::from_reader(content_file).with_context(|| {
-                            format!("While reading content file {:?}", content_file_uuid)
-                        })?;
-
-                    let (mut cursor, documents_batch_index) = reader.into_cursor_and_fields_index();
-                    while let Some(doc) = cursor.next_document().with_context(|| {
-                        format!("While iterating on content file {:?}", content_file_uuid)
-                    })? {
-                        dump_content_file
-                            .push_document(&obkv_to_object(&doc, &documents_batch_index)?)?;
-                    }
-                    dump_content_file.flush()?;
-                    count += 1;
-                }
-            }
-        }
-        dump_tasks.flush()?;
-
-        eprintln!("Successfully dumped {count} enqueued tasks!");
-    }
-
-    eprintln!("Dumping the indexes...");
-
-    // 4. Dump the indexes
-    let mut count = 0;
-    for result in index_mapping.iter(&rtxn)? {
-        let (uid, uuid) = result?;
-        let index_path = db_path.join("indexes").join(uuid.to_string());
-        let index = Index::new(EnvOpenOptions::new(), &index_path).with_context(|| {
-            format!("While trying to open the index at path {:?}", index_path.display())
-        })?;
-
-        let rtxn = index.read_txn()?;
-        let metadata = IndexMetadata {
-            uid: uid.to_owned(),
-            primary_key: index.primary_key(&rtxn)?.map(String::from),
-            created_at: index.created_at(&rtxn)?,
-            updated_at: index.updated_at(&rtxn)?,
-        };
-        let mut index_dumper = dump.create_index(uid, &metadata)?;
-
-        let fields_ids_map = index.fields_ids_map(&rtxn)?;
-        let all_fields: Vec<_> = fields_ids_map.iter().map(|(id, _)| id).collect();
-
-        // 4.1. Dump the documents
-        for ret in index.all_documents(&rtxn)? {
-            let (_id, doc) = ret?;
-            let document = obkv_to_json(&all_fields, &fields_ids_map, doc)?;
-            index_dumper.push_document(&document)?;
-        }
-
-        // 4.2. Dump the settings
-        let settings = meilisearch_types::settings::settings(&index, &rtxn)?;
-        index_dumper.settings(&settings)?;
-        count += 1;
-    }
-
-    eprintln!("Successfully dumped {count} indexes!");
-    // We will not dump experimental feature settings
-    eprintln!("The tool is not dumping experimental features, please set them by hand afterward");
-
-    let dump_uid = started_at.format(format_description!(
-        "[year repr:full][month repr:numerical][day padding:zero]-[hour padding:zero][minute padding:zero][second padding:zero][subsecond digits:3]"
-    )).unwrap();
-
-    let path = dump_dir.join(format!("{}.dump", dump_uid));
-    let file = File::create(&path)?;
-    dump.persist_to(BufWriter::new(file))?;
-
-    eprintln!("Dump exported at path {:?}", path.display());
-
-    Ok(())
-}

meilitool/src/uuid_codec.rs

@@ -1,24 +0,0 @@
-use std::borrow::Cow;
-use std::convert::TryInto;
-
-use meilisearch_types::heed::{BytesDecode, BytesEncode};
-use uuid::Uuid;
-
-/// A heed codec for value of struct Uuid.
-pub struct UuidCodec;
-
-impl<'a> BytesDecode<'a> for UuidCodec {
-    type DItem = Uuid;
-
-    fn bytes_decode(bytes: &'a [u8]) -> Option<Self::DItem> {
-        bytes.try_into().ok().map(Uuid::from_bytes)
-    }
-}
-
-impl BytesEncode<'_> for UuidCodec {
-    type EItem = Uuid;
-
-    fn bytes_encode(item: &Self::EItem) -> Option<Cow<[u8]>> {
-        Some(Cow::Borrowed(item.as_bytes()))
-    }
-}

milli/Cargo.toml

@@ -17,7 +17,7 @@ bincode = "1.3.3"
 bstr = "1.4.0"
 bytemuck = { version = "1.13.1", features = ["extern_crate_alloc"] }
 byteorder = "1.4.3"
-charabia = { version = "0.8.5", default-features = false }
+charabia = { version = "0.8.3", default-features = false }
 concat-arrays = "0.1.2"
 crossbeam-channel = "0.5.8"
 deserr = { version = "0.6.0", features = ["actix-web"]}
@@ -82,7 +82,7 @@ md5 = "0.7.0"
 rand = { version = "0.8.5", features = ["small_rng"] }
 
 [features]
-all-tokenizations = ["charabia/chinese", "charabia/hebrew", "charabia/japanese", "charabia/thai", "charabia/korean", "charabia/greek", "charabia/khmer"]
+all-tokenizations = ["charabia/chinese", "charabia/hebrew", "charabia/japanese", "charabia/thai", "charabia/korean", "charabia/greek"]
 
 # Use POSIX semaphores instead of SysV semaphores in LMDB
 # For more information on this feature, see heed's Cargo.toml
@@ -106,6 +106,3 @@ thai = ["charabia/thai"]
 
 # allow greek specialized tokenization
 greek = ["charabia/greek"]
-
-# allow khmer specialized tokenization
-khmer = ["charabia/khmer"]

milli/examples/search.rs

@@ -58,6 +58,7 @@ fn main() -> Result<(), Box<dyn Error>> {
                 false,
                 &None,
                 &None,
+                &None,
                 GeoSortStrategy::default(),
                 0,
                 20,

milli/examples/settings.rs

@@ -3,7 +3,7 @@ use heed::EnvOpenOptions;
 // use maplit::hashset;
 use milli::{
     update::{IndexerConfig, Settings},
-    Criterion, Index,
+    Index, RankingRule,
 };
 
 fn main() {
@@ -19,13 +19,13 @@ fn main() {
     // builder.set_min_word_len_one_typo(5);
     // builder.set_min_word_len_two_typos(7);
     // builder.set_sortable_fields(hashset! { S("release_date") });
-    builder.set_criteria(vec![
-        Criterion::Words,
-        Criterion::Typo,
-        Criterion::Proximity,
-        Criterion::Attribute,
-        Criterion::Sort,
-        Criterion::Exactness,
+    builder.set_ranking_rules(vec![
+        RankingRule::Words,
+        RankingRule::Typo,
+        RankingRule::Proximity,
+        RankingRule::Attribute,
+        RankingRule::Sort,
+        RankingRule::Exactness,
     ]);
 
     builder.execute(|_| (), || false).unwrap();

milli/src/asc_desc.rs

@@ -8,7 +8,7 @@ use thiserror::Error;
 
 use crate::error::is_reserved_keyword;
 use crate::search::facet::BadGeoError;
-use crate::{CriterionError, Error, UserError};
+use crate::{Error, RankingRuleError, UserError};
 
 /// This error type is never supposed to be shown to the end user.
 /// You must always cast it to a sort error or a criterion error.
@@ -28,23 +28,23 @@ impl From<BadGeoError> for AscDescError {
     }
 }
 
-impl From<AscDescError> for CriterionError {
+impl From<AscDescError> for RankingRuleError {
     fn from(error: AscDescError) -> Self {
         match error {
             AscDescError::GeoError(_) => {
-                CriterionError::ReservedNameForSort { name: "_geoPoint".to_string() }
+                RankingRuleError::ReservedNameForSort { name: "_geoPoint".to_string() }
             }
-            AscDescError::InvalidSyntax { name } => CriterionError::InvalidName { name },
+            AscDescError::InvalidSyntax { name } => RankingRuleError::InvalidName { name },
             AscDescError::ReservedKeyword { name } if name.starts_with("_geoPoint") => {
-                CriterionError::ReservedNameForSort { name: "_geoPoint".to_string() }
+                RankingRuleError::ReservedNameForSort { name: "_geoPoint".to_string() }
             }
             AscDescError::ReservedKeyword { name } if name.starts_with("_geoRadius") => {
-                CriterionError::ReservedNameForFilter { name: "_geoRadius".to_string() }
+                RankingRuleError::ReservedNameForFilter { name: "_geoRadius".to_string() }
             }
             AscDescError::ReservedKeyword { name } if name.starts_with("_geoBoundingBox") => {
-                CriterionError::ReservedNameForFilter { name: "_geoBoundingBox".to_string() }
+                RankingRuleError::ReservedNameForFilter { name: "_geoBoundingBox".to_string() }
             }
-            AscDescError::ReservedKeyword { name } => CriterionError::ReservedName { name },
+            AscDescError::ReservedKeyword { name } => RankingRuleError::ReservedName { name },
         }
     }
 }

milli/src/error.rs

@@ -9,7 +9,7 @@ use serde_json::Value;
 use thiserror::Error;
 
 use crate::documents::{self, DocumentsBatchCursorError};
-use crate::{CriterionError, DocumentId, FieldId, Object, SortError};
+use crate::{DocumentId, FieldId, Object, RankingRuleError, SortError};
 
 pub fn is_reserved_keyword(keyword: &str) -> bool {
     ["_geo", "_geoDistance", "_geoPoint", "_geoRadius", "_geoBoundingBox"].contains(&keyword)
@@ -94,7 +94,7 @@ pub enum UserError {
     #[error("A document cannot contain more than 65,535 fields.")]
     AttributeLimitReached,
     #[error(transparent)]
-    CriterionError(#[from] CriterionError),
+    CriterionError(#[from] RankingRuleError),
     #[error("Maximum number of documents reached.")]
     DocumentLimitReached,
     #[error(
@@ -116,6 +116,8 @@ only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and undersco
     InvalidVectorsType { document_id: Value, value: Value },
     #[error("{0}")]
     InvalidFilter(String),
+    #[error("{0}")]
+    InvalidBoostingFilter(String),
     #[error("Invalid type for filter subexpression: expected: {}, found: {1}.", .0.join(", "))]
     InvalidFilterExpression(&'static [&'static str], Value),
     #[error("Attribute `{}` is not sortable. {}",
@@ -280,7 +282,7 @@ error_from_sub_error! {
     ThreadPoolBuildError => InternalError,
     SerializationError => InternalError,
     GeoError => UserError,
-    CriterionError => UserError,
+    RankingRuleError => UserError,
 }
 
 impl<E> From<grenad::Error<E>> for Error

milli/src/index.rs

@@ -25,10 +25,9 @@ use crate::heed_codec::{
 };
 use crate::readable_slices::ReadableSlices;
 use crate::{
-    default_criteria, CboRoaringBitmapCodec, Criterion, DocumentId, ExternalDocumentsIds,
-    FacetDistribution, FieldDistribution, FieldId, FieldIdWordCountCodec, GeoPoint, ObkvCodec,
-    OrderBy, Result, RoaringBitmapCodec, RoaringBitmapLenCodec, Search, U8StrStrCodec, BEU16,
-    BEU32,
+    default_criteria, CboRoaringBitmapCodec, DocumentId, ExternalDocumentsIds, FacetDistribution,
+    FieldDistribution, FieldId, FieldIdWordCountCodec, GeoPoint, ObkvCodec, OrderBy, RankingRule,
+    Result, RoaringBitmapCodec, RoaringBitmapLenCodec, Search, U8StrStrCodec, BEU16, BEU32,
 };
 
 /// The HNSW data-structure that we serialize, fill and search in.
@@ -895,7 +894,7 @@ impl Index {
         let distinct_field = self.distinct_field(rtxn)?;
         let asc_desc_fields =
             self.criteria(rtxn)?.into_iter().filter_map(|criterion| match criterion {
-                Criterion::Asc(field) | Criterion::Desc(field) => Some(field),
+                RankingRule::Asc(field) | RankingRule::Desc(field) => Some(field),
                 _otherwise => None,
             });
 
@@ -1023,17 +1022,17 @@ impl Index {
     pub(crate) fn put_criteria(
         &self,
         wtxn: &mut RwTxn,
-        criteria: &[Criterion],
+        criteria: &[RankingRule],
     ) -> heed::Result<()> {
-        self.main.put::<_, Str, SerdeJson<&[Criterion]>>(wtxn, main_key::CRITERIA_KEY, &criteria)
+        self.main.put::<_, Str, SerdeJson<&[RankingRule]>>(wtxn, main_key::CRITERIA_KEY, &criteria)
     }
 
     pub(crate) fn delete_criteria(&self, wtxn: &mut RwTxn) -> heed::Result<bool> {
         self.main.delete::<_, Str>(wtxn, main_key::CRITERIA_KEY)
     }
 
-    pub fn criteria(&self, rtxn: &RoTxn) -> heed::Result<Vec<Criterion>> {
-        match self.main.get::<_, Str, SerdeJson<Vec<Criterion>>>(rtxn, main_key::CRITERIA_KEY)? {
+    pub fn criteria(&self, rtxn: &RoTxn) -> heed::Result<Vec<RankingRule>> {
+        match self.main.get::<_, Str, SerdeJson<Vec<RankingRule>>>(rtxn, main_key::CRITERIA_KEY)? {
             Some(criteria) => Ok(criteria),
             None => Ok(default_criteria()),
         }

milli/src/lib.rs

@@ -9,7 +9,6 @@ pub static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
 pub mod documents;
 
 mod asc_desc;
-mod criterion;
 pub mod distance;
 mod error;
 mod external_documents_ids;
@@ -18,6 +17,7 @@ mod fields_ids_map;
 pub mod heed_codec;
 pub mod index;
 pub mod proximity;
+mod ranking_rule;
 mod readable_slices;
 pub mod score_details;
 mod search;
@@ -44,7 +44,6 @@ use serde_json::Value;
 pub use {charabia as tokenizer, heed};
 
 pub use self::asc_desc::{AscDesc, AscDescError, Member, SortError};
-pub use self::criterion::{default_criteria, Criterion, CriterionError};
 pub use self::error::{
     Error, FieldIdMapMissingEntry, InternalError, SerializationError, UserError,
 };
@@ -57,6 +56,7 @@ pub use self::heed_codec::{
     UncheckedU8StrStrCodec,
 };
 pub use self::index::Index;
+pub use self::ranking_rule::{default_criteria, RankingRule, RankingRuleError};
 pub use self::search::{
     FacetDistribution, FacetValueHit, Filter, FormatOptions, MatchBounds, MatcherBuilder,
     MatchingWords, OrderBy, Search, SearchForFacetValues, SearchResult, TermsMatchingStrategy,

milli/src/ranking_rule.rs

@@ -7,7 +7,7 @@ use thiserror::Error;
 use crate::{AscDesc, Member};
 
 #[derive(Error, Debug)]
-pub enum CriterionError {
+pub enum RankingRuleError {
     #[error("`{name}` ranking rule is invalid. Valid ranking rules are words, typo, sort, proximity, attribute, exactness and custom ranking rules.")]
     InvalidName { name: String },
     #[error("`{name}` is a reserved keyword and thus can't be used as a ranking rule")]
@@ -25,7 +25,9 @@ pub enum CriterionError {
 }
 
 #[derive(Debug, Serialize, Deserialize, Clone, PartialEq, Eq)]
-pub enum Criterion {
+pub enum RankingRule {
+    /// Sorted by documents matching the given filter and then documents not matching it.
+    FilterBoosting(String),
     /// Sorted by decreasing number of matched query terms.
     /// Query words at the front of an attribute is considered better than if it was at the back.
     Words,
@@ -47,62 +49,68 @@ pub enum Criterion {
     Desc(String),
 }
 
-impl Criterion {
+impl RankingRule {
     /// Returns the field name parameter of this criterion.
     pub fn field_name(&self) -> Option<&str> {
         match self {
-            Criterion::Asc(name) | Criterion::Desc(name) => Some(name),
+            RankingRule::Asc(name) | RankingRule::Desc(name) => Some(name),
             _otherwise => None,
         }
     }
 }
 
-impl FromStr for Criterion {
-    type Err = CriterionError;
+impl FromStr for RankingRule {
+    type Err = RankingRuleError;
 
-    fn from_str(text: &str) -> Result<Criterion, Self::Err> {
+    fn from_str(text: &str) -> Result<RankingRule, Self::Err> {
         match text {
-            "words" => Ok(Criterion::Words),
-            "typo" => Ok(Criterion::Typo),
-            "proximity" => Ok(Criterion::Proximity),
-            "attribute" => Ok(Criterion::Attribute),
-            "sort" => Ok(Criterion::Sort),
-            "exactness" => Ok(Criterion::Exactness),
-            text => match AscDesc::from_str(text)? {
-                AscDesc::Asc(Member::Field(field)) => Ok(Criterion::Asc(field)),
-                AscDesc::Desc(Member::Field(field)) => Ok(Criterion::Desc(field)),
-                AscDesc::Asc(Member::Geo(_)) | AscDesc::Desc(Member::Geo(_)) => {
-                    Err(CriterionError::ReservedNameForSort { name: "_geoPoint".to_string() })?
-                }
+            "words" => Ok(RankingRule::Words),
+            "typo" => Ok(RankingRule::Typo),
+            "proximity" => Ok(RankingRule::Proximity),
+            "attribute" => Ok(RankingRule::Attribute),
+            "sort" => Ok(RankingRule::Sort),
+            "exactness" => Ok(RankingRule::Exactness),
+            text => match AscDesc::from_str(text) {
+                Ok(asc_desc) => match asc_desc {
+                    AscDesc::Asc(Member::Field(field)) => Ok(RankingRule::Asc(field)),
+                    AscDesc::Desc(Member::Field(field)) => Ok(RankingRule::Desc(field)),
+                    AscDesc::Asc(Member::Geo(_)) | AscDesc::Desc(Member::Geo(_)) => {
+                        Err(RankingRuleError::ReservedNameForSort {
+                            name: "_geoPoint".to_string(),
+                        })?
+                    }
+                },
+                Err(err) => Err(err.into()),
             },
         }
     }
 }
 
-pub fn default_criteria() -> Vec<Criterion> {
+pub fn default_criteria() -> Vec<RankingRule> {
     vec![
-        Criterion::Words,
-        Criterion::Typo,
-        Criterion::Proximity,
-        Criterion::Attribute,
-        Criterion::Sort,
-        Criterion::Exactness,
+        RankingRule::Words,
+        RankingRule::Typo,
+        RankingRule::Proximity,
+        RankingRule::Attribute,
+        RankingRule::Sort,
+        RankingRule::Exactness,
     ]
 }
 
-impl fmt::Display for Criterion {
+impl fmt::Display for RankingRule {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
-        use Criterion::*;
+        use RankingRule::*;
 
         match self {
             Words => f.write_str("words"),
+            FilterBoosting(_) => write!(f, "filterBoosting"),
             Typo => f.write_str("typo"),
             Proximity => f.write_str("proximity"),
             Attribute => f.write_str("attribute"),
             Sort => f.write_str("sort"),
             Exactness => f.write_str("exactness"),
-            Asc(attr) => write!(f, "{}:asc", attr),
-            Desc(attr) => write!(f, "{}:desc", attr),
+            Asc(attr) => write!(f, "{attr}:asc"),
+            Desc(attr) => write!(f, "{attr}:desc"),
         }
     }
 }
@@ -110,29 +118,29 @@ impl fmt::Display for Criterion {
 #[cfg(test)]
 mod tests {
     use big_s::S;
-    use CriterionError::*;
+    use RankingRuleError::*;
 
     use super::*;
 
     #[test]
     fn parse_criterion() {
         let valid_criteria = [
-            ("words", Criterion::Words),
-            ("typo", Criterion::Typo),
-            ("proximity", Criterion::Proximity),
-            ("attribute", Criterion::Attribute),
-            ("sort", Criterion::Sort),
-            ("exactness", Criterion::Exactness),
-            ("price:asc", Criterion::Asc(S("price"))),
-            ("price:desc", Criterion::Desc(S("price"))),
-            ("price:asc:desc", Criterion::Desc(S("price:asc"))),
-            ("truc:machin:desc", Criterion::Desc(S("truc:machin"))),
-            ("hello-world!:desc", Criterion::Desc(S("hello-world!"))),
-            ("it's spacy over there:asc", Criterion::Asc(S("it's spacy over there"))),
+            ("words", RankingRule::Words),
+            ("typo", RankingRule::Typo),
+            ("proximity", RankingRule::Proximity),
+            ("attribute", RankingRule::Attribute),
+            ("sort", RankingRule::Sort),
+            ("exactness", RankingRule::Exactness),
+            ("price:asc", RankingRule::Asc(S("price"))),
+            ("price:desc", RankingRule::Desc(S("price"))),
+            ("price:asc:desc", RankingRule::Desc(S("price:asc"))),
+            ("truc:machin:desc", RankingRule::Desc(S("truc:machin"))),
+            ("hello-world!:desc", RankingRule::Desc(S("hello-world!"))),
+            ("it's spacy over there:asc", RankingRule::Asc(S("it's spacy over there"))),
         ];
 
         for (input, expected) in valid_criteria {
-            let res = input.parse::<Criterion>();
+            let res = input.parse::<RankingRule>();
             assert!(
                 res.is_ok(),
                 "Failed to parse `{}`, was expecting `{:?}` but instead got `{:?}`",
@@ -167,7 +175,7 @@ mod tests {
         ];
 
         for (input, expected) in invalid_criteria {
-            let res = input.parse::<Criterion>();
+            let res = input.parse::<RankingRule>();
             assert!(
                 res.is_err(),
                 "Should no be able to parse `{}`, was expecting an error but instead got: `{:?}`",

milli/src/score_details.rs

@@ -5,6 +5,7 @@ use crate::distance_between_two_points;
 #[derive(Debug, Clone, PartialEq)]
 pub enum ScoreDetails {
     Words(Words),
+    FilterBoosting(FilterBoosting),
     Typo(Typo),
     Proximity(Rank),
     Fid(Rank),
@@ -23,6 +24,7 @@ impl ScoreDetails {
     pub fn rank(&self) -> Option<Rank> {
         match self {
             ScoreDetails::Words(details) => Some(details.rank()),
+            ScoreDetails::FilterBoosting(_) => None,
             ScoreDetails::Typo(details) => Some(details.rank()),
             ScoreDetails::Proximity(details) => Some(*details),
             ScoreDetails::Fid(details) => Some(*details),
@@ -60,6 +62,11 @@ impl ScoreDetails {
                     details_map.insert("words".into(), words_details);
                     order += 1;
                 }
+                ScoreDetails::FilterBoosting(FilterBoosting { matching }) => {
+                    let sort_details = serde_json::json!({ "matching": matching });
+                    details_map.insert("filterBoosting".into(), sort_details);
+                    order += 1;
+                }
                 ScoreDetails::Typo(typo) => {
                     let typo_details = serde_json::json!({
                         "order": order,
@@ -221,6 +228,11 @@ impl Words {
     }
 }
 
+#[derive(Debug, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
+pub struct FilterBoosting {
+    pub matching: bool,
+}
+
 /// Structure that is super similar to [`Words`], but whose semantics is a bit distinct.
 ///
 /// In exactness, the number of matching words can actually be 0 with a non-zero score,

milli/src/search/facet/filter.rs

@@ -3,7 +3,7 @@ use std::fmt::{Debug, Display};
 use std::ops::Bound::{self, Excluded, Included};
 
 use either::Either;
-pub use filter_parser::{Condition, Error as FPError, FilterCondition, Token};
+pub use filter_parser::{Condition, Error as FPError, FilterCondition, Span, Token};
 use roaring::RoaringBitmap;
 use serde_json::Value;
 

milli/src/search/mod.rs

@@ -11,7 +11,7 @@ use once_cell::sync::Lazy;
 use roaring::bitmap::RoaringBitmap;
 
 pub use self::facet::{FacetDistribution, Filter, OrderBy, DEFAULT_VALUES_PER_FACET};
-pub use self::new::matches::{FormatOptions, MatchBounds, MatcherBuilder, MatchingWords};
+pub use self::new::matches::{FormatOptions, MatchBounds, Matcher, MatcherBuilder, MatchingWords};
 use self::new::PartialSearchResult;
 use crate::error::UserError;
 use crate::heed_codec::facet::{FacetGroupKey, FacetGroupValue};
@@ -38,6 +38,7 @@ pub struct Search<'a> {
     vector: Option<Vec<f32>>,
     // this should be linked to the String in the query
     filter: Option<Filter<'a>>,
+    boosting_filter: Option<Filter<'a>>,
     offset: usize,
     limit: usize,
     sort_criteria: Option<Vec<AscDesc>>,
@@ -57,6 +58,7 @@ impl<'a> Search<'a> {
             query: None,
             vector: None,
             filter: None,
+            boosting_filter: None,
             offset: 0,
             limit: 20,
             sort_criteria: None,
@@ -121,6 +123,11 @@ impl<'a> Search<'a> {
         self
     }
 
+    pub fn boosting_filter(&mut self, condition: Filter<'a>) -> &mut Search<'a> {
+        self.boosting_filter = Some(condition);
+        self
+    }
+
     #[cfg(test)]
     pub fn geo_sort_strategy(&mut self, strategy: new::GeoSortStrategy) -> &mut Search<'a> {
         self.geo_strategy = strategy;
@@ -150,6 +157,7 @@ impl<'a> Search<'a> {
                 self.scoring_strategy,
                 self.exhaustive_number_hits,
                 &self.filter,
+                &self.boosting_filter,
                 &self.sort_criteria,
                 self.geo_strategy,
                 self.offset,
@@ -175,6 +183,7 @@ impl fmt::Debug for Search<'_> {
             query,
             vector: _,
             filter,
+            boosting_filter,
             offset,
             limit,
             sort_criteria,
@@ -191,6 +200,7 @@ impl fmt::Debug for Search<'_> {
             .field("query", query)
             .field("vector", &"[...]")
             .field("filter", filter)
+            .field("boosting_filter", boosting_filter)
             .field("offset", offset)
             .field("limit", limit)
             .field("sort_criteria", sort_criteria)

milli/src/search/new/bucket_sort.rs

@@ -46,8 +46,9 @@ pub fn bucket_sort<'ctx, Q: RankingRuleQueryTrait>(
         if let Some(distinct_fid) = distinct_fid {
             let mut excluded = RoaringBitmap::new();
             let mut results = vec![];
+            let mut skip = 0;
             for docid in universe.iter() {
-                if results.len() >= from + length {
+                if results.len() >= length {
                     break;
                 }
                 if excluded.contains(docid) {
@@ -55,19 +56,16 @@ pub fn bucket_sort<'ctx, Q: RankingRuleQueryTrait>(
                 }
 
                 distinct_single_docid(ctx.index, ctx.txn, distinct_fid, docid, &mut excluded)?;
+                skip += 1;
+                if skip <= from {
+                    continue;
+                }
+
                 results.push(docid);
             }
 
             let mut all_candidates = universe - excluded;
             all_candidates.extend(results.iter().copied());
-            // drain the results of the skipped elements
-            // this **must** be done **after** writing the entire results in `all_candidates` to ensure
-            // e.g. estimatedTotalHits is correct.
-            if results.len() >= from {
-                results.drain(..from);
-            } else {
-                results.clear();
-            }
 
             return Ok(BucketSortOutput {
                 scores: vec![Default::default(); results.len()],

milli/src/search/new/filter_boosting.rs

@@ -0,0 +1,79 @@
+use roaring::RoaringBitmap;
+
+use super::logger::SearchLogger;
+use super::{RankingRule, RankingRuleOutput, RankingRuleQueryTrait, SearchContext};
+use crate::score_details::{self, ScoreDetails};
+use crate::{Filter, Result};
+
+pub struct FilterBoosting<'f, Query> {
+    filter: Filter<'f>,
+    original_query: Option<Query>,
+    matching: Option<RankingRuleOutput<Query>>,
+    non_matching: Option<RankingRuleOutput<Query>>,
+}
+
+impl<'f, Query> FilterBoosting<'f, Query> {
+    pub fn new(filter: Filter<'f>) -> Result<Self> {
+        Ok(Self { filter, original_query: None, matching: None, non_matching: None })
+    }
+}
+
+impl<'ctx, 'f, Query: RankingRuleQueryTrait> RankingRule<'ctx, Query>
+    for FilterBoosting<'f, Query>
+{
+    fn id(&self) -> String {
+        // TODO improve this
+        let Self { filter: original_expression, .. } = self;
+        format!("boost:{original_expression:?}")
+    }
+
+    fn start_iteration(
+        &mut self,
+        ctx: &mut SearchContext<'ctx>,
+        _logger: &mut dyn SearchLogger<Query>,
+        parent_candidates: &RoaringBitmap,
+        parent_query: &Query,
+    ) -> Result<()> {
+        let universe_matching = match self.filter.evaluate(ctx.txn, ctx.index) {
+            Ok(documents) => documents,
+            Err(e) => return Err(e), // TODO manage the invalid_search_boosting_filter
+        };
+        let matching = parent_candidates & universe_matching;
+        let non_matching = parent_candidates - &matching;
+
+        self.original_query = Some(parent_query.clone());
+
+        self.matching = Some(RankingRuleOutput {
+            query: parent_query.clone(),
+            candidates: matching,
+            score: ScoreDetails::FilterBoosting(score_details::FilterBoosting { matching: true }),
+        });
+
+        self.non_matching = Some(RankingRuleOutput {
+            query: parent_query.clone(),
+            candidates: non_matching,
+            score: ScoreDetails::FilterBoosting(score_details::FilterBoosting { matching: false }),
+        });
+
+        Ok(())
+    }
+
+    fn next_bucket(
+        &mut self,
+        _ctx: &mut SearchContext<'ctx>,
+        _logger: &mut dyn SearchLogger<Query>,
+        _universe: &RoaringBitmap,
+    ) -> Result<Option<RankingRuleOutput<Query>>> {
+        Ok(self.matching.take().or_else(|| self.non_matching.take()))
+    }
+
+    fn end_iteration(
+        &mut self,
+        _ctx: &mut SearchContext<'ctx>,
+        _logger: &mut dyn SearchLogger<Query>,
+    ) {
+        self.original_query = None;
+        self.matching = None;
+        self.non_matching = None;
+    }
+}

milli/src/search/new/matches/mod.rs

@@ -512,6 +512,7 @@ mod tests {
                 false,
                 &None,
                 &None,
+                &None,
                 crate::search::new::GeoSortStrategy::default(),
                 0,
                 100,

milli/src/search/new/mod.rs

@@ -15,6 +15,7 @@ mod resolve_query_graph;
 mod small_bitmap;
 
 mod exact_attribute;
+mod filter_boosting;
 mod sort;
 
 #[cfg(test)]
@@ -26,6 +27,7 @@ use bucket_sort::{bucket_sort, BucketSortOutput};
 use charabia::TokenizerBuilder;
 use db_cache::DatabaseCache;
 use exact_attribute::ExactAttribute;
+use filter_boosting::FilterBoosting;
 use graph_based_ranking_rule::{Exactness, Fid, Position, Proximity, Typo};
 use heed::RoTxn;
 use instant_distance::Search;
@@ -190,25 +192,30 @@ fn resolve_universe(
 }
 
 /// Return the list of initialised ranking rules to be used for a placeholder search.
-fn get_ranking_rules_for_placeholder_search<'ctx>(
+fn get_ranking_rules_for_placeholder_search<'ctx, 'f: 'ctx>(
     ctx: &SearchContext<'ctx>,
     sort_criteria: &Option<Vec<AscDesc>>,
     geo_strategy: geo_sort::Strategy,
+    boosting_filter: &Option<Filter<'f>>,
 ) -> Result<Vec<BoxRankingRule<'ctx, PlaceholderQuery>>> {
     let mut sort = false;
     let mut sorted_fields = HashSet::new();
     let mut geo_sorted = false;
-    let mut ranking_rules: Vec<BoxRankingRule<PlaceholderQuery>> = vec![];
+    let mut ranking_rules: Vec<BoxRankingRule<_>> = match boosting_filter {
+        Some(filter) => vec![Box::new(FilterBoosting::new(filter.clone())?)],
+        None => Vec::new(),
+    };
     let settings_ranking_rules = ctx.index.criteria(ctx.txn)?;
     for rr in settings_ranking_rules {
         match rr {
             // These rules need a query to have an effect; ignore them in placeholder search
-            crate::Criterion::Words
-            | crate::Criterion::Typo
-            | crate::Criterion::Attribute
-            | crate::Criterion::Proximity
-            | crate::Criterion::Exactness => continue,
-            crate::Criterion::Sort => {
+            crate::RankingRule::FilterBoosting(_)
+            | crate::RankingRule::Words
+            | crate::RankingRule::Typo
+            | crate::RankingRule::Attribute
+            | crate::RankingRule::Proximity
+            | crate::RankingRule::Exactness => continue,
+            crate::RankingRule::Sort => {
                 if sort {
                     continue;
                 }
@@ -222,14 +229,14 @@ fn get_ranking_rules_for_placeholder_search<'ctx>(
                 )?;
                 sort = true;
             }
-            crate::Criterion::Asc(field_name) => {
+            crate::RankingRule::Asc(field_name) => {
                 if sorted_fields.contains(&field_name) {
                     continue;
                 }
                 sorted_fields.insert(field_name.clone());
                 ranking_rules.push(Box::new(Sort::new(ctx.index, ctx.txn, field_name, true)?));
             }
-            crate::Criterion::Desc(field_name) => {
+            crate::RankingRule::Desc(field_name) => {
                 if sorted_fields.contains(&field_name) {
                     continue;
                 }
@@ -242,11 +249,12 @@ fn get_ranking_rules_for_placeholder_search<'ctx>(
 }
 
 /// Return the list of initialised ranking rules to be used for a query graph search.
-fn get_ranking_rules_for_query_graph_search<'ctx>(
+fn get_ranking_rules_for_query_graph_search<'ctx, 'f: 'ctx>(
     ctx: &SearchContext<'ctx>,
     sort_criteria: &Option<Vec<AscDesc>>,
     geo_strategy: geo_sort::Strategy,
     terms_matching_strategy: TermsMatchingStrategy,
+    boosting_filter: &Option<Filter<'f>>,
 ) -> Result<Vec<BoxRankingRule<'ctx, QueryGraph>>> {
     // query graph search
     let mut words = false;
@@ -263,15 +271,18 @@ fn get_ranking_rules_for_query_graph_search<'ctx>(
         words = true;
     }
 
-    let mut ranking_rules: Vec<BoxRankingRule<QueryGraph>> = vec![];
+    let mut ranking_rules: Vec<BoxRankingRule<QueryGraph>> = match boosting_filter {
+        Some(filter) => vec![Box::new(FilterBoosting::new(filter.clone())?)],
+        None => Vec::new(),
+    };
     let settings_ranking_rules = ctx.index.criteria(ctx.txn)?;
     for rr in settings_ranking_rules {
         // Add Words before any of: typo, proximity, attribute
         match rr {
-            crate::Criterion::Typo
-            | crate::Criterion::Attribute
-            | crate::Criterion::Proximity
-            | crate::Criterion::Exactness => {
+            crate::RankingRule::Typo
+            | crate::RankingRule::Attribute
+            | crate::RankingRule::Proximity
+            | crate::RankingRule::Exactness => {
                 if !words {
                     ranking_rules.push(Box::new(Words::new(terms_matching_strategy)));
                     words = true;
@@ -280,28 +291,33 @@ fn get_ranking_rules_for_query_graph_search<'ctx>(
             _ => {}
         }
         match rr {
-            crate::Criterion::Words => {
+            crate::RankingRule::Words => {
                 if words {
                     continue;
                 }
                 ranking_rules.push(Box::new(Words::new(terms_matching_strategy)));
                 words = true;
             }
-            crate::Criterion::Typo => {
+            crate::RankingRule::FilterBoosting(_) => {
+                // that is not possible to define the filterBoosting ranking rule by hand,
+                // or by using the seetings. It is always inserted by the engine itself.
+                continue;
+            }
+            crate::RankingRule::Typo => {
                 if typo {
                     continue;
                 }
                 typo = true;
                 ranking_rules.push(Box::new(Typo::new(None)));
             }
-            crate::Criterion::Proximity => {
+            crate::RankingRule::Proximity => {
                 if proximity {
                     continue;
                 }
                 proximity = true;
                 ranking_rules.push(Box::new(Proximity::new(None)));
             }
-            crate::Criterion::Attribute => {
+            crate::RankingRule::Attribute => {
                 if attribute {
                     continue;
                 }
@@ -309,7 +325,7 @@ fn get_ranking_rules_for_query_graph_search<'ctx>(
                 ranking_rules.push(Box::new(Fid::new(None)));
                 ranking_rules.push(Box::new(Position::new(None)));
             }
-            crate::Criterion::Sort => {
+            crate::RankingRule::Sort => {
                 if sort {
                     continue;
                 }
@@ -323,7 +339,7 @@ fn get_ranking_rules_for_query_graph_search<'ctx>(
                 )?;
                 sort = true;
             }
-            crate::Criterion::Exactness => {
+            crate::RankingRule::Exactness => {
                 if exactness {
                     continue;
                 }
@@ -331,14 +347,15 @@ fn get_ranking_rules_for_query_graph_search<'ctx>(
                 ranking_rules.push(Box::new(Exactness::new()));
                 exactness = true;
             }
-            crate::Criterion::Asc(field_name) => {
+            crate::RankingRule::Asc(field_name) => {
+                // TODO Question: Why would it be invalid to sort price:asc, typo, price:desc?
                 if sorted_fields.contains(&field_name) {
                     continue;
                 }
                 sorted_fields.insert(field_name.clone());
                 ranking_rules.push(Box::new(Sort::new(ctx.index, ctx.txn, field_name, true)?));
             }
-            crate::Criterion::Desc(field_name) => {
+            crate::RankingRule::Desc(field_name) => {
                 if sorted_fields.contains(&field_name) {
                     continue;
                 }
@@ -406,14 +423,15 @@ fn resolve_sort_criteria<'ctx, Query: RankingRuleQueryTrait>(
 }
 
 #[allow(clippy::too_many_arguments)]
-pub fn execute_search(
-    ctx: &mut SearchContext,
+pub fn execute_search<'ctx, 'f: 'ctx>(
+    ctx: &mut SearchContext<'ctx>,
     query: &Option<String>,
     vector: &Option<Vec<f32>>,
     terms_matching_strategy: TermsMatchingStrategy,
     scoring_strategy: ScoringStrategy,
     exhaustive_number_hits: bool,
-    filters: &Option<Filter>,
+    filter: &Option<Filter>,
+    boosting_filter: &Option<Filter<'f>>,
     sort_criteria: &Option<Vec<AscDesc>>,
     geo_strategy: geo_sort::Strategy,
     from: usize,
@@ -422,8 +440,8 @@ pub fn execute_search(
     placeholder_search_logger: &mut dyn SearchLogger<PlaceholderQuery>,
     query_graph_logger: &mut dyn SearchLogger<QueryGraph>,
 ) -> Result<PartialSearchResult> {
-    let mut universe = if let Some(filters) = filters {
-        filters.evaluate(ctx.txn, ctx.index)?
+    let mut universe = if let Some(filter) = filter {
+        filter.evaluate(ctx.txn, ctx.index)?
     } else {
         ctx.index.documents_ids(ctx.txn)?
     };
@@ -434,18 +452,7 @@ pub fn execute_search(
         let mut search = Search::default();
         let docids = match ctx.index.vector_hnsw(ctx.txn)? {
             Some(hnsw) => {
-                if let Some(expected_size) = hnsw.iter().map(|(_, point)| point.len()).next() {
-                    if vector.len() != expected_size {
-                        return Err(UserError::InvalidVectorDimensions {
-                            expected: expected_size,
-                            found: vector.len(),
-                        }
-                        .into());
-                    }
-                }
-
                 let vector = NDotProductPoint::new(vector.clone());
-
                 let neighbors = hnsw.search(&vector, &mut search);
 
                 let mut docids = Vec::new();
@@ -527,6 +534,7 @@ pub fn execute_search(
             sort_criteria,
             geo_strategy,
             terms_matching_strategy,
+            boosting_filter,
         )?;
 
         universe =
@@ -543,8 +551,13 @@ pub fn execute_search(
             query_graph_logger,
         )?
     } else {
-        let ranking_rules =
-            get_ranking_rules_for_placeholder_search(ctx, sort_criteria, geo_strategy)?;
+        let ranking_rules = get_ranking_rules_for_placeholder_search(
+            ctx,
+            sort_criteria,
+            geo_strategy,
+            boosting_filter,
+        )?;
+
         bucket_sort(
             ctx,
             ranking_rules,
@@ -591,7 +604,8 @@ fn check_sort_criteria(ctx: &SearchContext, sort_criteria: Option<&Vec<AscDesc>>
 
     // We check that the sort ranking rule exists and throw an
     // error if we try to use it and that it doesn't.
-    let sort_ranking_rule_missing = !ctx.index.criteria(ctx.txn)?.contains(&crate::Criterion::Sort);
+    let sort_ranking_rule_missing =
+        !ctx.index.criteria(ctx.txn)?.contains(&crate::RankingRule::Sort);
     if sort_ranking_rule_missing {
         return Err(UserError::SortRankingRuleMissing.into());
     }

milli/src/search/new/ranking_rule_graph/mod.rs

@@ -29,7 +29,7 @@ use std::hash::Hash;
 pub use cheapest_paths::PathVisitor;
 pub use condition_docids_cache::ConditionDocIdsCache;
 pub use dead_ends_cache::DeadEndsCache;
-pub use exactness::ExactnessGraph;
+pub use exactness::{ExactnessCondition, ExactnessGraph};
 pub use fid::{FidCondition, FidGraph};
 pub use position::{PositionCondition, PositionGraph};
 pub use proximity::{ProximityCondition, ProximityGraph};

milli/src/search/new/tests/attribute_fid.rs

@@ -1,5 +1,5 @@
 use crate::index::tests::TempIndex;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -12,7 +12,7 @@ fn create_index() -> TempIndex {
                 "description".to_owned(),
                 "plot".to_owned(),
             ]);
-            s.set_criteria(vec![Criterion::Attribute]);
+            s.set_ranking_rules(vec![RankingRule::Attribute]);
         })
         .unwrap();
 

milli/src/search/new/tests/attribute_position.rs

@@ -1,5 +1,5 @@
 use crate::index::tests::TempIndex;
-use crate::{db_snap, Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{db_snap, RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -12,7 +12,7 @@ fn create_index() -> TempIndex {
                 "text2".to_owned(),
                 "other".to_owned(),
             ]);
-            s.set_criteria(vec![Criterion::Attribute]);
+            s.set_ranking_rules(vec![RankingRule::Attribute]);
         })
         .unwrap();
 

milli/src/search/new/tests/distinct.rs

@@ -19,7 +19,7 @@ use maplit::hashset;
 
 use super::collect_field_values;
 use crate::index::tests::TempIndex;
-use crate::{AscDesc, Criterion, Index, Member, Search, SearchResult, TermsMatchingStrategy};
+use crate::{AscDesc, Index, Member, RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -30,7 +30,7 @@ fn create_index() -> TempIndex {
             s.set_searchable_fields(vec!["text".to_owned()]);
             s.set_sortable_fields(hashset! { S("rank1"), S("letter") });
             s.set_distinct_field("letter".to_owned());
-            s.set_criteria(vec![Criterion::Words]);
+            s.set_ranking_rules(vec![RankingRule::Words]);
         })
         .unwrap();
 
@@ -252,7 +252,7 @@ fn test_distinct_placeholder_sort() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Sort]);
+            s.set_ranking_rules(vec![RankingRule::Sort]);
         })
         .unwrap();
 
@@ -387,7 +387,7 @@ fn test_distinct_words() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words]);
+            s.set_ranking_rules(vec![RankingRule::Words]);
         })
         .unwrap();
 
@@ -440,7 +440,11 @@ fn test_distinct_sort_words() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Sort, Criterion::Words, Criterion::Desc(S("rank1"))]);
+            s.set_ranking_rules(vec![
+                RankingRule::Sort,
+                RankingRule::Words,
+                RankingRule::Desc(S("rank1")),
+            ]);
         })
         .unwrap();
 
@@ -513,7 +517,7 @@ fn test_distinct_all_candidates() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Sort]);
+            s.set_ranking_rules(vec![RankingRule::Sort]);
         })
         .unwrap();
 
@@ -536,7 +540,7 @@ fn test_distinct_typo() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words, Criterion::Typo]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Typo]);
         })
         .unwrap();
 

milli/src/search/new/tests/exactness.rs

@@ -21,7 +21,7 @@ Then these rules will only work with
 
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index_simple_ordered() -> TempIndex {
     let index = TempIndex::new();
@@ -30,7 +30,7 @@ fn create_index_simple_ordered() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness]);
+            s.set_ranking_rules(vec![RankingRule::Exactness]);
         })
         .unwrap();
 
@@ -89,7 +89,7 @@ fn create_index_simple_reversed() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness]);
+            s.set_ranking_rules(vec![RankingRule::Exactness]);
         })
         .unwrap();
 
@@ -147,7 +147,7 @@ fn create_index_simple_random() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness]);
+            s.set_ranking_rules(vec![RankingRule::Exactness]);
         })
         .unwrap();
 
@@ -201,7 +201,7 @@ fn create_index_attribute_starts_with() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness]);
+            s.set_ranking_rules(vec![RankingRule::Exactness]);
         })
         .unwrap();
 
@@ -251,7 +251,7 @@ fn create_index_simple_ordered_with_typos() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness]);
+            s.set_ranking_rules(vec![RankingRule::Exactness]);
         })
         .unwrap();
 
@@ -350,7 +350,11 @@ fn create_index_with_varying_proximities() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness, Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![
+                RankingRule::Exactness,
+                RankingRule::Words,
+                RankingRule::Proximity,
+            ]);
         })
         .unwrap();
 
@@ -404,7 +408,7 @@ fn create_index_with_typo_and_prefix() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness]);
+            s.set_ranking_rules(vec![RankingRule::Exactness]);
         })
         .unwrap();
 
@@ -442,7 +446,11 @@ fn create_index_all_equal_except_proximity_between_ignored_terms() -> TempIndex
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Exactness, Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![
+                RankingRule::Exactness,
+                RankingRule::Words,
+                RankingRule::Proximity,
+            ]);
         })
         .unwrap();
 
@@ -698,7 +706,7 @@ fn test_exactness_after_words() {
 
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words, Criterion::Exactness]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Exactness]);
         })
         .unwrap();
 
@@ -747,7 +755,7 @@ fn test_words_after_exactness() {
 
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Exactness, Criterion::Words]);
+            s.set_ranking_rules(vec![RankingRule::Exactness, RankingRule::Words]);
         })
         .unwrap();
 
@@ -796,7 +804,11 @@ fn test_proximity_after_exactness() {
 
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Exactness, Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![
+                RankingRule::Exactness,
+                RankingRule::Words,
+                RankingRule::Proximity,
+            ]);
         })
         .unwrap();
 
@@ -834,7 +846,11 @@ fn test_proximity_after_exactness() {
 
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Exactness, Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![
+                RankingRule::Exactness,
+                RankingRule::Words,
+                RankingRule::Proximity,
+            ]);
         })
         .unwrap();
 
@@ -868,7 +884,11 @@ fn test_exactness_followed_by_typo_prefer_no_typo_prefix() {
 
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Exactness, Criterion::Words, Criterion::Typo]);
+            s.set_ranking_rules(vec![
+                RankingRule::Exactness,
+                RankingRule::Words,
+                RankingRule::Typo,
+            ]);
         })
         .unwrap();
 
@@ -904,7 +924,11 @@ fn test_typo_followed_by_exactness() {
 
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words, Criterion::Typo, Criterion::Exactness]);
+            s.set_ranking_rules(vec![
+                RankingRule::Words,
+                RankingRule::Typo,
+                RankingRule::Exactness,
+            ]);
         })
         .unwrap();
 

milli/src/search/new/tests/geo_sort.rs

@@ -9,7 +9,7 @@ use maplit::hashset;
 use crate::index::tests::TempIndex;
 use crate::score_details::ScoreDetails;
 use crate::search::new::tests::collect_field_values;
-use crate::{AscDesc, Criterion, GeoSortStrategy, Member, Search, SearchResult};
+use crate::{AscDesc, GeoSortStrategy, Member, RankingRule, Search, SearchResult};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -18,7 +18,7 @@ fn create_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_sortable_fields(hashset! { S("_geo") });
-            s.set_criteria(vec![Criterion::Words, Criterion::Sort]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Sort]);
         })
         .unwrap();
     index

milli/src/search/new/tests/integration.rs

@@ -6,10 +6,10 @@ use maplit::{btreemap, hashset};
 
 use crate::documents::{DocumentsBatchBuilder, DocumentsBatchReader};
 use crate::update::{IndexDocuments, IndexDocumentsConfig, IndexerConfig, Settings};
-use crate::{db_snap, Criterion, Index, Object};
+use crate::{db_snap, Index, Object, RankingRule};
 pub const CONTENT: &str = include_str!("../../../../tests/assets/test_set.ndjson");
 
-pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
+pub fn setup_search_index_with_criteria(criteria: &[RankingRule]) -> Index {
     let path = tempfile::tempdir().unwrap();
     let mut options = EnvOpenOptions::new();
     options.map_size(10 * 1024 * 1024); // 10 MB
@@ -20,7 +20,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
 
     let mut builder = Settings::new(&mut wtxn, &index, &config);
 
-    builder.set_criteria(criteria.to_vec());
+    builder.set_ranking_rules(criteria.to_vec());
     builder.set_filterable_fields(hashset! {
         S("tag"),
         S("asc_desc_rank"),
@@ -70,6 +70,6 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
 
 #[test]
 fn snapshot_integration_dataset() {
-    let index = setup_search_index_with_criteria(&[Criterion::Attribute]);
+    let index = setup_search_index_with_criteria(&[RankingRule::Attribute]);
     db_snap!(index, word_position_docids, @"3c9347a767bceef3beb31465f1e5f3ae");
 }

milli/src/search/new/tests/ngram_split_words.rs

@@ -19,7 +19,7 @@ This module tests the following properties:
 
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -28,7 +28,7 @@ fn create_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Words]);
+            s.set_ranking_rules(vec![RankingRule::Words]);
         })
         .unwrap();
 

milli/src/search/new/tests/proximity.rs

@@ -19,7 +19,7 @@ use std::collections::BTreeMap;
 
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_simple_index() -> TempIndex {
     let index = TempIndex::new();
@@ -28,7 +28,7 @@ fn create_simple_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Proximity]);
         })
         .unwrap();
 
@@ -94,7 +94,7 @@ fn create_edge_cases_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Proximity]);
         })
         .unwrap();
 

milli/src/search/new/tests/proximity_typo.rs

@@ -8,7 +8,7 @@ implemented.
 
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -17,7 +17,11 @@ fn create_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Words, Criterion::Proximity, Criterion::Typo]);
+            s.set_ranking_rules(vec![
+                RankingRule::Words,
+                RankingRule::Proximity,
+                RankingRule::Typo,
+            ]);
         })
         .unwrap();
 

milli/src/search/new/tests/sort.rs

@@ -17,7 +17,7 @@ use maplit::hashset;
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
 use crate::{
-    score_details, AscDesc, Criterion, Member, Search, SearchResult, TermsMatchingStrategy,
+    score_details, AscDesc, Member, RankingRule, Search, SearchResult, TermsMatchingStrategy,
 };
 
 fn create_index() -> TempIndex {
@@ -28,7 +28,7 @@ fn create_index() -> TempIndex {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
             s.set_sortable_fields(hashset! { S("rank"), S("vague"), S("letter") });
-            s.set_criteria(vec![Criterion::Sort]);
+            s.set_ranking_rules(vec![RankingRule::Sort]);
         })
         .unwrap();
 
@@ -331,7 +331,7 @@ fn test_redacted() {
         .update_settings(|s| {
             s.set_displayed_fields(vec!["text".to_owned(), "vague".to_owned()]);
             s.set_sortable_fields(hashset! { S("rank"), S("vague"), S("letter") });
-            s.set_criteria(vec![Criterion::Sort]);
+            s.set_ranking_rules(vec![RankingRule::Sort]);
         })
         .unwrap();
 

milli/src/search/new/tests/typo.rs

@@ -22,7 +22,7 @@ use std::collections::BTreeMap;
 
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -31,7 +31,7 @@ fn create_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Words]);
+            s.set_ranking_rules(vec![RankingRule::Words]);
         })
         .unwrap();
 
@@ -457,7 +457,7 @@ fn test_typo_ranking_rule_not_preceded_by_words_ranking_rule() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Typo]);
+            s.set_ranking_rules(vec![RankingRule::Typo]);
         })
         .unwrap();
 
@@ -495,7 +495,7 @@ fn test_typo_ranking_rule_not_preceded_by_words_ranking_rule() {
 
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words, Criterion::Typo]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Typo]);
         })
         .unwrap();
 
@@ -540,7 +540,7 @@ fn test_typo_bucketing() {
     drop(txn);
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Typo]);
+            s.set_ranking_rules(vec![RankingRule::Typo]);
         })
         .unwrap();
     let txn = index.read_txn().unwrap();
@@ -589,7 +589,7 @@ fn test_typo_synonyms() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Typo]);
+            s.set_ranking_rules(vec![RankingRule::Typo]);
 
             let mut synonyms = BTreeMap::new();
             synonyms.insert("lackadaisical".to_owned(), vec!["lazy".to_owned()]);

milli/src/search/new/tests/typo_proximity.rs

@@ -17,7 +17,7 @@ because the typo ranking rule before it only used the derivation `beautiful`.
 
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -26,7 +26,11 @@ fn create_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Words, Criterion::Typo, Criterion::Proximity]);
+            s.set_ranking_rules(vec![
+                RankingRule::Words,
+                RankingRule::Typo,
+                RankingRule::Proximity,
+            ]);
         })
         .unwrap();
 

milli/src/search/new/tests/words_tms.rs

@@ -14,7 +14,7 @@ account by the proximity ranking rule.
 
 use crate::index::tests::TempIndex;
 use crate::search::new::tests::collect_field_values;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
 
 fn create_index() -> TempIndex {
     let index = TempIndex::new();
@@ -23,7 +23,7 @@ fn create_index() -> TempIndex {
         .update_settings(|s| {
             s.set_primary_key("id".to_owned());
             s.set_searchable_fields(vec!["text".to_owned()]);
-            s.set_criteria(vec![Criterion::Words]);
+            s.set_ranking_rules(vec![RankingRule::Words]);
         })
         .unwrap();
 
@@ -265,7 +265,7 @@ fn test_words_proximity_tms_last_simple() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Proximity]);
         })
         .unwrap();
 
@@ -346,7 +346,7 @@ fn test_words_proximity_tms_last_phrase() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Proximity]);
         })
         .unwrap();
 
@@ -416,7 +416,7 @@ fn test_words_tms_all() {
     let index = create_index();
     index
         .update_settings(|s| {
-            s.set_criteria(vec![Criterion::Words, Criterion::Proximity]);
+            s.set_ranking_rules(vec![RankingRule::Words, RankingRule::Proximity]);
         })
         .unwrap();
 

milli/src/update/index_documents/helpers/mod.rs

@@ -14,7 +14,7 @@ pub use grenad_helpers::{
 };
 pub use merge_functions::{
     concat_u32s_array, keep_first, keep_latest_obkv, merge_btreeset_string,
-    merge_cbo_roaring_bitmaps, merge_obkvs_and_operations, merge_roaring_bitmaps,
+    merge_cbo_roaring_bitmaps, merge_obkvs_and_operations, merge_roaring_bitmaps, merge_two_obkvs,
     serialize_roaring_bitmap, MergeFn,
 };
 

milli/src/update/index_documents/mod.rs

@@ -20,7 +20,10 @@ use slice_group_by::GroupBy;
 use typed_chunk::{write_typed_chunk_into_index, TypedChunk};
 
 use self::enrich::enrich_documents_batch;
-pub use self::enrich::{extract_finite_float_from_value, DocumentId};
+pub use self::enrich::{
+    extract_finite_float_from_value, validate_document_id, validate_document_id_value,
+    validate_geo_from_json, DocumentId,
+};
 pub use self::helpers::{
     as_cloneable_grenad, create_sorter, create_writer, fst_stream_into_hashset,
     fst_stream_into_vec, merge_btreeset_string, merge_cbo_roaring_bitmaps, merge_roaring_bitmaps,

milli/src/update/settings.rs

@@ -9,9 +9,9 @@ use time::OffsetDateTime;
 
 use super::index_documents::{IndexDocumentsConfig, Transform};
 use super::IndexerConfig;
-use crate::criterion::Criterion;
 use crate::error::UserError;
 use crate::index::{DEFAULT_MIN_WORD_LEN_ONE_TYPO, DEFAULT_MIN_WORD_LEN_TWO_TYPOS};
+use crate::ranking_rule::RankingRule;
 use crate::update::index_documents::IndexDocumentsMethod;
 use crate::update::{IndexDocuments, UpdateIndexingStep};
 use crate::{FieldsIdsMap, Index, OrderBy, Result};
@@ -110,7 +110,7 @@ pub struct Settings<'a, 't, 'u, 'i> {
     displayed_fields: Setting<Vec<String>>,
     filterable_fields: Setting<HashSet<String>>,
     sortable_fields: Setting<HashSet<String>>,
-    criteria: Setting<Vec<Criterion>>,
+    ranking_rules: Setting<Vec<RankingRule>>,
     stop_words: Setting<BTreeSet<String>>,
     non_separator_tokens: Setting<BTreeSet<String>>,
     separator_tokens: Setting<BTreeSet<String>>,
@@ -142,7 +142,7 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
             displayed_fields: Setting::NotSet,
             filterable_fields: Setting::NotSet,
             sortable_fields: Setting::NotSet,
-            criteria: Setting::NotSet,
+            ranking_rules: Setting::NotSet,
             stop_words: Setting::NotSet,
             non_separator_tokens: Setting::NotSet,
             separator_tokens: Setting::NotSet,
@@ -194,12 +194,12 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
         self.sortable_fields = Setting::Reset;
     }
 
-    pub fn reset_criteria(&mut self) {
-        self.criteria = Setting::Reset;
+    pub fn reset_ranking_rules(&mut self) {
+        self.ranking_rules = Setting::Reset;
     }
 
-    pub fn set_criteria(&mut self, criteria: Vec<Criterion>) {
-        self.criteria = Setting::Set(criteria);
+    pub fn set_ranking_rules(&mut self, ranking_rules: Vec<RankingRule>) {
+        self.ranking_rules = Setting::Set(ranking_rules);
     }
 
     pub fn reset_stop_words(&mut self) {
@@ -696,7 +696,7 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
     }
 
     fn update_criteria(&mut self) -> Result<()> {
-        match &self.criteria {
+        match &self.ranking_rules {
             Setting::Set(criteria) => {
                 self.index.put_criteria(self.wtxn, criteria)?;
             }
@@ -924,7 +924,7 @@ mod tests {
     use crate::error::Error;
     use crate::index::tests::TempIndex;
     use crate::update::{ClearDocuments, DeleteDocuments};
-    use crate::{Criterion, Filter, SearchResult};
+    use crate::{Filter, RankingRule, SearchResult};
 
     #[test]
     fn set_and_reset_searchable_fields() {
@@ -1167,7 +1167,7 @@ mod tests {
         index
             .update_settings(|settings| {
                 settings.set_displayed_fields(vec![S("name")]);
-                settings.set_criteria(vec![Criterion::Asc("age".to_owned())]);
+                settings.set_ranking_rules(vec![RankingRule::Asc("age".to_owned())]);
             })
             .unwrap();
 
@@ -1473,7 +1473,7 @@ mod tests {
             .update_settings(|settings| {
                 settings.set_displayed_fields(vec!["hello".to_string()]);
                 settings.set_filterable_fields(hashset! { S("age"), S("toto") });
-                settings.set_criteria(vec![Criterion::Asc(S("toto"))]);
+                settings.set_ranking_rules(vec![RankingRule::Asc(S("toto"))]);
             })
             .unwrap();
 
@@ -1482,7 +1482,7 @@ mod tests {
         assert_eq!(&["hello"][..], index.displayed_fields(&rtxn).unwrap().unwrap());
         // since no documents have been pushed the primary key is still unset
         assert!(index.primary_key(&rtxn).unwrap().is_none());
-        assert_eq!(vec![Criterion::Asc("toto".to_string())], index.criteria(&rtxn).unwrap());
+        assert_eq!(vec![RankingRule::Asc("toto".to_string())], index.criteria(&rtxn).unwrap());
         drop(rtxn);
 
         // We set toto and age as searchable to force reordering of the fields
@@ -1495,7 +1495,7 @@ mod tests {
         let rtxn = index.read_txn().unwrap();
         assert_eq!(&["hello"][..], index.displayed_fields(&rtxn).unwrap().unwrap());
         assert!(index.primary_key(&rtxn).unwrap().is_none());
-        assert_eq!(vec![Criterion::Asc("toto".to_string())], index.criteria(&rtxn).unwrap());
+        assert_eq!(vec![RankingRule::Asc("toto".to_string())], index.criteria(&rtxn).unwrap());
     }
 
     #[test]
@@ -1507,7 +1507,7 @@ mod tests {
             .update_settings(|settings| {
                 settings.set_displayed_fields(vec!["hello".to_string()]);
                 // It is only Asc(toto), there is a facet database but it is denied to filter with toto.
-                settings.set_criteria(vec![Criterion::Asc(S("toto"))]);
+                settings.set_ranking_rules(vec![RankingRule::Asc(S("toto"))]);
             })
             .unwrap();
 
@@ -1715,7 +1715,7 @@ mod tests {
                     displayed_fields,
                     filterable_fields,
                     sortable_fields,
-                    criteria,
+                    ranking_rules: criteria,
                     stop_words,
                     non_separator_tokens,
                     separator_tokens,

milli/tests/search/distinct.rs

@@ -2,8 +2,8 @@ use std::collections::HashSet;
 
 use big_s::S;
 use milli::update::Settings;
-use milli::{Criterion, Search, SearchResult, TermsMatchingStrategy};
-use Criterion::*;
+use milli::{RankingRule, Search, SearchResult, TermsMatchingStrategy};
+use RankingRule::*;
 
 use crate::search::{self, EXTERNAL_DOCUMENTS_IDS};
 
@@ -202,7 +202,7 @@ test_distinct!(
     EXTERNAL_DOCUMENTS_IDS.len(),
     1,
     vec![],
-    3
+    2
 );
 test_distinct!(
     // testing: https://github.com/meilisearch/meilisearch/issues/4078
@@ -212,7 +212,7 @@ test_distinct!(
     1,
     2,
     vec![],
-    3
+    1
 );
 test_distinct!(
     // testing: https://github.com/meilisearch/meilisearch/issues/4078
@@ -222,7 +222,7 @@ test_distinct!(
     EXTERNAL_DOCUMENTS_IDS.len(),
     2,
     vec![],
-    7
+    5
 );
 test_distinct!(
     // testing: https://github.com/meilisearch/meilisearch/issues/4078
@@ -232,5 +232,5 @@ test_distinct!(
     2,
     4,
     vec![],
-    7
+    3
 );

milli/tests/search/filters.rs

@@ -1,6 +1,6 @@
 use either::{Either, Left, Right};
-use milli::{Criterion, Filter, Search, SearchResult, TermsMatchingStrategy};
-use Criterion::*;
+use milli::{Filter, RankingRule, Search, SearchResult, TermsMatchingStrategy};
+use RankingRule::*;
 
 use crate::search::{self, EXTERNAL_DOCUMENTS_IDS};
 

milli/tests/search/mod.rs

@@ -8,7 +8,7 @@ use heed::EnvOpenOptions;
 use maplit::{btreemap, hashset};
 use milli::documents::{DocumentsBatchBuilder, DocumentsBatchReader};
 use milli::update::{IndexDocuments, IndexDocumentsConfig, IndexerConfig, Settings};
-use milli::{AscDesc, Criterion, DocumentId, Index, Member, Object, TermsMatchingStrategy};
+use milli::{AscDesc, DocumentId, Index, Member, Object, RankingRule, TermsMatchingStrategy};
 use serde::{Deserialize, Deserializer};
 use slice_group_by::GroupBy;
 
@@ -27,7 +27,7 @@ pub const EXTERNAL_DOCUMENTS_IDS: &[&str; 17] =
 
 pub const CONTENT: &str = include_str!("../assets/test_set.ndjson");
 
-pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
+pub fn setup_search_index_with_criteria(criteria: &[RankingRule]) -> Index {
     let path = tempfile::tempdir().unwrap();
     let mut options = EnvOpenOptions::new();
     options.map_size(10 * 1024 * 1024); // 10 MB
@@ -38,7 +38,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
 
     let mut builder = Settings::new(&mut wtxn, &index, &config);
 
-    builder.set_criteria(criteria.to_vec());
+    builder.set_ranking_rules(criteria.to_vec());
     builder.set_filterable_fields(hashset! {
         S("tag"),
         S("asc_desc_rank"),
@@ -95,7 +95,7 @@ pub fn internal_to_external_ids(index: &Index, internal_ids: &[DocumentId]) -> V
 }
 
 pub fn expected_order(
-    criteria: &[Criterion],
+    criteria: &[RankingRule],
     optional_words: TermsMatchingStrategy,
     sort_by: &[AscDesc],
 ) -> Vec<TestDocument> {
@@ -107,47 +107,56 @@ pub fn expected_order(
         let mut new_groups = Vec::new();
         for group in groups.iter_mut() {
             match criterion {
-                Criterion::Attribute => {
+                RankingRule::Attribute => {
                     group.sort_by_key(|d| d.attribute_rank);
                     new_groups
                         .extend(group.linear_group_by_key(|d| d.attribute_rank).map(Vec::from));
                 }
-                Criterion::Exactness => {
+                RankingRule::Exactness => {
                     group.sort_by_key(|d| d.exact_rank);
                     new_groups.extend(group.linear_group_by_key(|d| d.exact_rank).map(Vec::from));
                 }
-                Criterion::Proximity => {
+                RankingRule::Proximity => {
                     group.sort_by_key(|d| d.proximity_rank);
                     new_groups
                         .extend(group.linear_group_by_key(|d| d.proximity_rank).map(Vec::from));
                 }
-                Criterion::Sort if sort_by == [AscDesc::Asc(Member::Field(S("tag")))] => {
+                RankingRule::Sort if sort_by == [AscDesc::Asc(Member::Field(S("tag")))] => {
                     group.sort_by_key(|d| d.sort_by_rank);
                     new_groups.extend(group.linear_group_by_key(|d| d.sort_by_rank).map(Vec::from));
                 }
-                Criterion::Sort if sort_by == [AscDesc::Desc(Member::Field(S("tag")))] => {
+                RankingRule::Sort if sort_by == [AscDesc::Desc(Member::Field(S("tag")))] => {
                     group.sort_by_key(|d| Reverse(d.sort_by_rank));
                     new_groups.extend(group.linear_group_by_key(|d| d.sort_by_rank).map(Vec::from));
                 }
-                Criterion::Typo => {
+                RankingRule::Typo => {
                     group.sort_by_key(|d| d.typo_rank);
                     new_groups.extend(group.linear_group_by_key(|d| d.typo_rank).map(Vec::from));
                 }
-                Criterion::Words => {
+                RankingRule::Words => {
                     group.sort_by_key(|d| d.word_rank);
                     new_groups.extend(group.linear_group_by_key(|d| d.word_rank).map(Vec::from));
                 }
-                Criterion::Asc(field_name) if field_name == "asc_desc_rank" => {
+                RankingRule::Asc(field_name) if field_name == "asc_desc_rank" => {
                     group.sort_by_key(|d| d.asc_desc_rank);
                     new_groups
                         .extend(group.linear_group_by_key(|d| d.asc_desc_rank).map(Vec::from));
                 }
-                Criterion::Desc(field_name) if field_name == "asc_desc_rank" => {
+                RankingRule::Desc(field_name) if field_name == "asc_desc_rank" => {
                     group.sort_by_key(|d| Reverse(d.asc_desc_rank));
                     new_groups
                         .extend(group.linear_group_by_key(|d| d.asc_desc_rank).map(Vec::from));
                 }
-                Criterion::Asc(_) | Criterion::Desc(_) | Criterion::Sort => {
+                RankingRule::FilterBoosting(filter) => {
+                    // move the matching documents first, then the ones that don't match
+                    group.sort_by_key(|d| if execute_filter(filter, d).is_some() { 0 } else { 1 });
+                    new_groups.extend(
+                        group
+                            .linear_group_by_key(|d| execute_filter(filter, d).is_some())
+                            .map(Vec::from),
+                    );
+                }
+                RankingRule::Asc(_) | RankingRule::Desc(_) | RankingRule::Sort => {
                     new_groups.push(group.clone())
                 }
             }

milli/tests/search/phrase_search.rs

@@ -1,7 +1,7 @@
 use milli::update::{IndexerConfig, Settings};
-use milli::{Criterion, Index, Search, TermsMatchingStrategy};
+use milli::{Index, RankingRule, Search, TermsMatchingStrategy};
 
-use crate::search::Criterion::{Attribute, Exactness, Proximity};
+use crate::search::RankingRule::{Attribute, Exactness, Proximity};
 
 fn set_stop_words(index: &Index, stop_words: &[&str]) {
     let mut wtxn = index.write_txn().unwrap();
@@ -14,7 +14,7 @@ fn set_stop_words(index: &Index, stop_words: &[&str]) {
     wtxn.commit().unwrap();
 }
 
-fn test_phrase_search_with_stop_words_given_criteria(criteria: &[Criterion]) {
+fn test_phrase_search_with_stop_words_given_criteria(criteria: &[RankingRule]) {
     let index = super::setup_search_index_with_criteria(criteria);
 
     // Add stop_words

milli/tests/search/query_criteria.rs

@@ -7,9 +7,9 @@ use itertools::Itertools;
 use maplit::hashset;
 use milli::documents::{DocumentsBatchBuilder, DocumentsBatchReader};
 use milli::update::{IndexDocuments, IndexDocumentsConfig, IndexerConfig, Settings};
-use milli::{AscDesc, Criterion, Index, Member, Search, SearchResult, TermsMatchingStrategy};
+use milli::{AscDesc, Index, Member, RankingRule, Search, SearchResult, TermsMatchingStrategy};
 use rand::Rng;
-use Criterion::*;
+use RankingRule::*;
 
 use crate::search::{self, EXTERNAL_DOCUMENTS_IDS};
 
@@ -88,7 +88,7 @@ test_criterion!(
 
 #[test]
 fn criteria_mixup() {
-    use Criterion::*;
+    use RankingRule::*;
     let index = search::setup_search_index_with_criteria(&[
         Words,
         Attribute,
@@ -233,7 +233,7 @@ fn criteria_mixup() {
         //update criteria
         let mut wtxn = index.write_txn().unwrap();
         let mut builder = Settings::new(&mut wtxn, &index, &config);
-        builder.set_criteria(criteria.clone());
+        builder.set_ranking_rules(criteria.clone());
         builder.execute(|_| (), || false).unwrap();
         wtxn.commit().unwrap();
 
@@ -324,7 +324,7 @@ fn criteria_ascdesc() {
 
         let mut wtxn = index.write_txn().unwrap();
         let mut builder = Settings::new(&mut wtxn, &index, &config);
-        builder.set_criteria(vec![criterion.clone()]);
+        builder.set_ranking_rules(vec![criterion.clone()]);
         builder.execute(|_| (), || false).unwrap();
         wtxn.commit().unwrap();
 

milli/tests/search/sort.rs

@@ -1,5 +1,5 @@
 use big_s::S;
-use milli::Criterion::{Attribute, Exactness, Proximity, Typo, Words};
+use milli::RankingRule::{Attribute, Exactness, Proximity, Typo, Words};
 use milli::{AscDesc, Error, Member, Search, TermsMatchingStrategy, UserError};
 
 use crate::search::{self, EXTERNAL_DOCUMENTS_IDS};

milli/tests/search/typo_tolerance.rs

@@ -2,10 +2,10 @@ use std::collections::BTreeSet;
 
 use heed::EnvOpenOptions;
 use milli::update::{IndexDocuments, IndexDocumentsConfig, IndexerConfig, Settings};
-use milli::{Criterion, Index, Search, TermsMatchingStrategy};
+use milli::{Index, RankingRule, Search, TermsMatchingStrategy};
 use serde_json::json;
 use tempfile::tempdir;
-use Criterion::*;
+use RankingRule::*;
 
 #[test]
 fn test_typo_tolerance_one_typo() {