Merge #4257

4257: Change proximity precision settings r=dureuill a=ManyTheFish

- [x] Add proximity_precision value into the analytics
- [x] Change the naming of `attributeScale` and `wordScale` into `byAttribute` and `byWord`
- [x] Remove proximityPrecision from the experimental feature

Co-authored-by: ManyTheFish <many@meilisearch.com>
Co-authored-by: Many the fish <many@meilisearch.com>

index-scheduler/src/batch.rs

@@ -1354,9 +1354,6 @@ impl IndexScheduler {
                     if matches!(checked_settings.embedders, milli::update::Setting::Set(_)) {
                         self.features().check_vector("Passing `embedders` in settings")?
                     }
-                    if checked_settings.proximity_precision.set().is_some() {
-                        self.features.features().check_proximity_precision()?;
-                    }
                     task.details = Some(Details::SettingsUpdate { settings: Box::new(settings) });
                     apply_settings_to_builder(&checked_settings, &mut builder);
 

index-scheduler/src/features.rs

@@ -81,19 +81,6 @@ impl RoFeatures {
             .into())
         }
     }
-
-    pub fn check_proximity_precision(&self) -> Result<()> {
-        if self.runtime.proximity_precision {
-            Ok(())
-        } else {
-            Err(FeatureNotEnabledError {
-                disabled_action: "Using `proximityPrecision` index setting",
-                feature: "proximity precision",
-                issue_link: "https://github.com/orgs/meilisearch/discussions/710",
-            }
-            .into())
-        }
-    }
 }
 
 impl FeatureData {

meilisearch-types/src/features.rs

@@ -7,7 +7,6 @@ pub struct RuntimeTogglableFeatures {
     pub vector_store: bool,
     pub metrics: bool,
     pub export_puffin_reports: bool,
-    pub proximity_precision: bool,
 }
 
 #[derive(Default, Debug, Clone, Copy)]

meilisearch-types/src/settings.rs

@@ -724,23 +724,23 @@ impl From<RankingRuleView> for Criterion {
 #[serde(deny_unknown_fields, rename_all = "camelCase")]
 #[deserr(error = DeserrJsonError<InvalidSettingsProximityPrecision>, rename_all = camelCase, deny_unknown_fields)]
 pub enum ProximityPrecisionView {
-    WordScale,
-    AttributeScale,
+    ByWord,
+    ByAttribute,
 }
 
 impl From<ProximityPrecision> for ProximityPrecisionView {
     fn from(value: ProximityPrecision) -> Self {
         match value {
-            ProximityPrecision::WordScale => ProximityPrecisionView::WordScale,
-            ProximityPrecision::AttributeScale => ProximityPrecisionView::AttributeScale,
+            ProximityPrecision::ByWord => ProximityPrecisionView::ByWord,
+            ProximityPrecision::ByAttribute => ProximityPrecisionView::ByAttribute,
         }
     }
 }
 impl From<ProximityPrecisionView> for ProximityPrecision {
     fn from(value: ProximityPrecisionView) -> Self {
         match value {
-            ProximityPrecisionView::WordScale => ProximityPrecision::WordScale,
-            ProximityPrecisionView::AttributeScale => ProximityPrecision::AttributeScale,
+            ProximityPrecisionView::ByWord => ProximityPrecision::ByWord,
+            ProximityPrecisionView::ByAttribute => ProximityPrecision::ByAttribute,
         }
     }
 }

meilisearch/src/routes/features.rs

@@ -48,8 +48,6 @@ pub struct RuntimeTogglableFeatures {
     pub metrics: Option<bool>,
     #[deserr(default)]
     pub export_puffin_reports: Option<bool>,
-    #[deserr(default)]
-    pub proximity_precision: Option<bool>,
 }
 
 async fn patch_features(
@@ -72,10 +70,6 @@ async fn patch_features(
             .0
             .export_puffin_reports
             .unwrap_or(old_features.export_puffin_reports),
-        proximity_precision: new_features
-            .0
-            .proximity_precision
-            .unwrap_or(old_features.proximity_precision),
     };
 
     // explicitly destructure for analytics rather than using the `Serialize` implementation, because
@@ -86,7 +80,6 @@ async fn patch_features(
         vector_store,
         metrics,
         export_puffin_reports,
-        proximity_precision,
     } = new_features;
 
     analytics.publish(
@@ -96,7 +89,6 @@ async fn patch_features(
             "vector_store": vector_store,
             "metrics": metrics,
             "export_puffin_reports": export_puffin_reports,
-            "proximity_precision": proximity_precision,
         }),
         Some(&req),
     );

meilisearch/src/routes/indexes/settings.rs

@@ -453,6 +453,7 @@ make_setting_route!(
             json!({
                 "proximity_precision": {
                     "set": precision.is_some(),
+                    "value": precision,
                 }
             }),
             Some(req),

meilisearch/tests/dumps/mod.rs

@@ -1860,8 +1860,7 @@ async fn import_dump_v6_containing_experimental_features() {
       "scoreDetails": false,
       "vectorStore": false,
       "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 
@@ -1890,7 +1889,7 @@ async fn import_dump_v6_containing_experimental_features() {
       "dictionary": [],
       "synonyms": {},
       "distinctAttribute": null,
-      "proximityPrecision": "attributeScale",
+      "proximityPrecision": "byAttribute",
       "typoTolerance": {
         "enabled": true,
         "minWordSizeForTypos": {

meilisearch/tests/features/mod.rs

@@ -21,8 +21,7 @@ async fn experimental_features() {
       "scoreDetails": false,
       "vectorStore": false,
       "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 
@@ -34,8 +33,7 @@ async fn experimental_features() {
       "scoreDetails": false,
       "vectorStore": true,
       "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 
@@ -47,8 +45,7 @@ async fn experimental_features() {
       "scoreDetails": false,
       "vectorStore": true,
       "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 
@@ -61,8 +58,7 @@ async fn experimental_features() {
       "scoreDetails": false,
       "vectorStore": true,
       "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 
@@ -75,8 +71,7 @@ async fn experimental_features() {
       "scoreDetails": false,
       "vectorStore": true,
       "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 }
@@ -96,8 +91,7 @@ async fn experimental_feature_metrics() {
       "scoreDetails": false,
       "vectorStore": false,
       "metrics": true,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 
@@ -152,7 +146,7 @@ async fn errors() {
     meili_snap::snapshot!(code, @"400 Bad Request");
     meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
     {
-      "message": "Unknown field `NotAFeature`: expected one of `scoreDetails`, `vectorStore`, `metrics`, `exportPuffinReports`, `proximityPrecision`",
+      "message": "Unknown field `NotAFeature`: expected one of `scoreDetails`, `vectorStore`, `metrics`, `exportPuffinReports`",
       "code": "bad_request",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#bad_request"

meilisearch/tests/search/hybrid.rs

@@ -16,8 +16,7 @@ async fn index_with_documents<'a>(server: &'a Server, documents: &Value) -> Inde
       "scoreDetails": false,
       "vectorStore": true,
       "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": false
+      "exportPuffinReports": false
     }
     "###);
 

meilisearch/tests/settings/proximity_settings.rs

@@ -27,17 +27,6 @@ static DOCUMENTS: Lazy<crate::common::Value> = Lazy::new(|| {
 #[actix_rt::test]
 async fn attribute_scale_search() {
     let server = Server::new().await;
-    let (response, code) = server.set_features(json!({"proximityPrecision": true})).await;
-    meili_snap::snapshot!(code, @"200 OK");
-    meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
-    {
-      "scoreDetails": false,
-      "vectorStore": false,
-      "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": true
-    }
-    "###);
     let index = server.index("test");
 
     index.add_documents(DOCUMENTS.clone(), None).await;
@@ -45,7 +34,7 @@ async fn attribute_scale_search() {
 
     let (response, code) = index
         .update_settings(json!({
-            "proximityPrecision": "attributeScale",
+            "proximityPrecision": "byAttribute",
             "rankingRules": ["words", "typo", "proximity"],
         }))
         .await;
@@ -111,17 +100,6 @@ async fn attribute_scale_search() {
 #[actix_rt::test]
 async fn attribute_scale_phrase_search() {
     let server = Server::new().await;
-    let (response, code) = server.set_features(json!({"proximityPrecision": true})).await;
-    meili_snap::snapshot!(code, @"200 OK");
-    meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
-    {
-      "scoreDetails": false,
-      "vectorStore": false,
-      "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": true
-    }
-    "###);
     let index = server.index("test");
 
     index.add_documents(DOCUMENTS.clone(), None).await;
@@ -129,7 +107,7 @@ async fn attribute_scale_phrase_search() {
 
     let (_response, _code) = index
         .update_settings(json!({
-            "proximityPrecision": "attributeScale",
+            "proximityPrecision": "byAttribute",
             "rankingRules": ["words", "typo", "proximity"],
         }))
         .await;
@@ -190,17 +168,6 @@ async fn attribute_scale_phrase_search() {
 #[actix_rt::test]
 async fn word_scale_set_and_reset() {
     let server = Server::new().await;
-    let (response, code) = server.set_features(json!({"proximityPrecision": true})).await;
-    meili_snap::snapshot!(code, @"200 OK");
-    meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
-    {
-      "scoreDetails": false,
-      "vectorStore": false,
-      "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": true
-    }
-    "###);
     let index = server.index("test");
 
     index.add_documents(DOCUMENTS.clone(), None).await;
@@ -209,7 +176,7 @@ async fn word_scale_set_and_reset() {
     // Set and reset the setting ensuring the swap between the 2 settings is applied.
     let (_response, _code) = index
         .update_settings(json!({
-            "proximityPrecision": "attributeScale",
+            "proximityPrecision": "byAttribute",
             "rankingRules": ["words", "typo", "proximity"],
         }))
         .await;
@@ -217,7 +184,7 @@ async fn word_scale_set_and_reset() {
 
     let (_response, _code) = index
         .update_settings(json!({
-            "proximityPrecision": "wordScale",
+            "proximityPrecision": "byWord",
             "rankingRules": ["words", "typo", "proximity"],
         }))
         .await;
@@ -316,17 +283,6 @@ async fn word_scale_set_and_reset() {
 #[actix_rt::test]
 async fn attribute_scale_default_ranking_rules() {
     let server = Server::new().await;
-    let (response, code) = server.set_features(json!({"proximityPrecision": true})).await;
-    meili_snap::snapshot!(code, @"200 OK");
-    meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
-    {
-      "scoreDetails": false,
-      "vectorStore": false,
-      "metrics": false,
-      "exportPuffinReports": false,
-      "proximityPrecision": true
-    }
-    "###);
     let index = server.index("test");
 
     index.add_documents(DOCUMENTS.clone(), None).await;
@@ -334,7 +290,7 @@ async fn attribute_scale_default_ranking_rules() {
 
     let (response, code) = index
         .update_settings(json!({
-            "proximityPrecision": "attributeScale"
+            "proximityPrecision": "byAttribute"
         }))
         .await;
     assert_eq!("202", code.as_str(), "{:?}", response);

milli/src/proximity.rs

@@ -32,6 +32,6 @@ pub fn path_proximity(path: &[Position]) -> u32 {
 #[serde(rename_all = "camelCase")]
 pub enum ProximityPrecision {
     #[default]
-    WordScale,
-    AttributeScale,
+    ByWord,
+    ByAttribute,
 }

milli/src/search/new/db_cache.rs

@@ -299,9 +299,9 @@ impl<'ctx> SearchContext<'ctx> {
         proximity: u8,
     ) -> Result<Option<RoaringBitmap>> {
         match self.index.proximity_precision(self.txn)?.unwrap_or_default() {
-            ProximityPrecision::AttributeScale => {
+            ProximityPrecision::ByAttribute => {
                 // Force proximity to 0 because:
-                // in AttributeScale, there are only 2 possible distances:
+                // in ByAttribute, there are only 2 possible distances:
                 // 1. words in same attribute: in that the DB contains (0, word1, word2)
                 // 2. words in different attributes: no DB entry for these two words.
                 let proximity = 0;
@@ -345,19 +345,17 @@ impl<'ctx> SearchContext<'ctx> {
 
                 Ok(docids)
             }
-            ProximityPrecision::WordScale => {
-                DatabaseCache::get_value::<_, _, CboRoaringBitmapCodec>(
-                    self.txn,
-                    (proximity, word1, word2),
-                    &(
-                        proximity,
-                        self.word_interner.get(word1).as_str(),
-                        self.word_interner.get(word2).as_str(),
-                    ),
-                    &mut self.db_cache.word_pair_proximity_docids,
-                    self.index.word_pair_proximity_docids.remap_data_type::<Bytes>(),
-                )
-            }
+            ProximityPrecision::ByWord => DatabaseCache::get_value::<_, _, CboRoaringBitmapCodec>(
+                self.txn,
+                (proximity, word1, word2),
+                &(
+                    proximity,
+                    self.word_interner.get(word1).as_str(),
+                    self.word_interner.get(word2).as_str(),
+                ),
+                &mut self.db_cache.word_pair_proximity_docids,
+                self.index.word_pair_proximity_docids.remap_data_type::<Bytes>(),
+            ),
         }
     }
 
@@ -368,10 +366,10 @@ impl<'ctx> SearchContext<'ctx> {
         proximity: u8,
     ) -> Result<Option<u64>> {
         match self.index.proximity_precision(self.txn)?.unwrap_or_default() {
-            ProximityPrecision::AttributeScale => Ok(self
+            ProximityPrecision::ByAttribute => Ok(self
                 .get_db_word_pair_proximity_docids(word1, word2, proximity)?
                 .map(|d| d.len())),
-            ProximityPrecision::WordScale => {
+            ProximityPrecision::ByWord => {
                 DatabaseCache::get_value::<_, _, CboRoaringBitmapLenCodec>(
                     self.txn,
                     (proximity, word1, word2),
@@ -394,9 +392,9 @@ impl<'ctx> SearchContext<'ctx> {
         mut proximity: u8,
     ) -> Result<Option<RoaringBitmap>> {
         let proximity_precision = self.index.proximity_precision(self.txn)?.unwrap_or_default();
-        if proximity_precision == ProximityPrecision::AttributeScale {
+        if proximity_precision == ProximityPrecision::ByAttribute {
             // Force proximity to 0 because:
-            // in AttributeScale, there are only 2 possible distances:
+            // in ByAttribute, there are only 2 possible distances:
             // 1. words in same attribute: in that the DB contains (0, word1, word2)
             // 2. words in different attributes: no DB entry for these two words.
             proximity = 0;
@@ -408,7 +406,7 @@ impl<'ctx> SearchContext<'ctx> {
             docids.clone()
         } else {
             let prefix_docids = match proximity_precision {
-                ProximityPrecision::AttributeScale => {
+                ProximityPrecision::ByAttribute => {
                     // Compute the distance at the attribute level and store it in the cache.
                     let fids = if let Some(fids) = self.index.searchable_fields_ids(self.txn)? {
                         fids
@@ -429,7 +427,7 @@ impl<'ctx> SearchContext<'ctx> {
                     }
                     prefix_docids
                 }
-                ProximityPrecision::WordScale => {
+                ProximityPrecision::ByWord => {
                     // compute docids using prefix iter and store the result in the cache.
                     let key = U8StrStrCodec::bytes_encode(&(
                         proximity,

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

@@ -157,7 +157,7 @@ pub(crate) fn data_from_obkv_documents(
         });
     }
 
-    if proximity_precision == ProximityPrecision::WordScale {
+    if proximity_precision == ProximityPrecision::ByWord {
         spawn_extraction_task::<_, _, Vec<grenad::Reader<BufReader<File>>>>(
             docid_word_positions_chunks.clone(),
             indexer,