Merge #4881

4881: Infer locales from index settings r=curquiza a=ManyTheFish

# Pull Request

## Related issue
Fixes #4828
Fixes #4816
## What does this PR do?
- Add some test using `AttributesToSearchOn`
- Make the search infer the language based on the index settings when the `locales` filed is not precise


CI is now working:
https://github.com/meilisearch/meilisearch/actions/runs/10490050545/job/29055955667



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

meilisearch/tests/search/locales.rs

@@ -386,12 +386,39 @@ async fn force_locales() {
             |response, code| {
                 snapshot!(response, @r###"
                 {
-                  "hits": [],
+                  "hits": [
+                    {
+                      "name_zh": "进击的巨人",
+                      "author_zh": "諫山創",
+                      "description_zh": "进击的巨人是日本的漫画系列，由諫山 創作画。",
+                      "id": 853,
+                      "_vectors": {
+                        "manual": [
+                          1.0,
+                          2.0,
+                          3.0
+                        ]
+                      },
+                      "_formatted": {
+                        "name_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>",
+                        "author_zh": "諫山創",
+                        "description_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>是日本的漫画系列，由諫山 創作画。",
+                        "id": "853",
+                        "_vectors": {
+                          "manual": [
+                            "1.0",
+                            "2.0",
+                            "3.0"
+                          ]
+                        }
+                      }
+                    }
+                  ],
                   "query": "\"进击的巨人\"",
                   "processingTimeMs": "[duration]",
                   "limit": 20,
                   "offset": 0,
-                  "estimatedTotalHits": 0
+                  "estimatedTotalHits": 1
                 }
                 "###);
                 snapshot!(code, @"200 OK");
@@ -483,12 +510,39 @@ async fn force_locales_with_pattern() {
             |response, code| {
                 snapshot!(response, @r###"
                 {
-                  "hits": [],
+                  "hits": [
+                    {
+                      "name_zh": "进击的巨人",
+                      "author_zh": "諫山創",
+                      "description_zh": "进击的巨人是日本的漫画系列，由諫山 創作画。",
+                      "id": 853,
+                      "_vectors": {
+                        "manual": [
+                          1.0,
+                          2.0,
+                          3.0
+                        ]
+                      },
+                      "_formatted": {
+                        "name_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>",
+                        "author_zh": "諫山創",
+                        "description_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>是日本的漫画系列，由諫山 創作画。",
+                        "id": "853",
+                        "_vectors": {
+                          "manual": [
+                            "1.0",
+                            "2.0",
+                            "3.0"
+                          ]
+                        }
+                      }
+                    }
+                  ],
                   "query": "\"进击的巨人\"",
                   "processingTimeMs": "[duration]",
                   "limit": 20,
                   "offset": 0,
-                  "estimatedTotalHits": 0
+                  "estimatedTotalHits": 1
                 }
                 "###);
                 snapshot!(code, @"200 OK");
@@ -761,6 +815,275 @@ async fn force_different_locales_with_pattern() {
         .await;
 }
 
+#[actix_rt::test]
+async fn auto_infer_locales_at_search_with_attributes_to_search_on() {
+    let server = Server::new().await;
+
+    let index = server.index("test");
+    let documents = DOCUMENTS.clone();
+    let (response, _) = index
+        .update_settings(
+            json!({
+                "searchableAttributes": ["name_en", "name_ja", "name_zh", "author_en", "author_ja", "author_zh", "description_en", "description_ja", "description_zh"],
+                "localizedAttributes": [
+                    // force japanese
+                    {"attributePatterns": ["*_zh"], "locales": ["jpn"]},
+                    // force chinese
+                    {"attributePatterns": ["*_ja"], "locales": ["cmn"]},
+                    // any language
+                    {"attributePatterns": ["*_en"], "locales": []}
+                ]
+            }),
+        )
+        .await;
+    snapshot!(response, @r###"
+    {
+      "taskUid": 0,
+      "indexUid": "test",
+      "status": "enqueued",
+      "type": "settingsUpdate",
+      "enqueuedAt": "[date]"
+    }
+    "###);
+    index.add_documents(documents, None).await;
+    index.wait_task(1).await;
+
+    // auto infer any language
+    index
+        .search(
+            json!({"q": "\"进击的巨人\"", "attributesToHighlight": ["*"]}),
+            |response, code| {
+                snapshot!(response, @r###"
+                {
+                  "hits": [],
+                  "query": "\"进击的巨人\"",
+                  "processingTimeMs": "[duration]",
+                  "limit": 20,
+                  "offset": 0,
+                  "estimatedTotalHits": 0
+                }
+                "###);
+                snapshot!(code, @"200 OK");
+            },
+        )
+        .await;
+
+    // should infer chinese
+    index
+            .search(
+                json!({"q": "\"进击的巨人\"", "attributesToHighlight": ["*"], "attributesToSearchOn": ["name_zh", "description_zh"]}),
+                |response, code| {
+                    snapshot!(response, @r###"
+                    {
+                      "hits": [
+                        {
+                          "name_zh": "进击的巨人",
+                          "author_zh": "諫山創",
+                          "description_zh": "进击的巨人是日本的漫画系列，由諫山 創作画。",
+                          "id": 853,
+                          "_vectors": {
+                            "manual": [
+                              1.0,
+                              2.0,
+                              3.0
+                            ]
+                          },
+                          "_formatted": {
+                            "name_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>",
+                            "author_zh": "諫山創",
+                            "description_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>是日本的漫画系列，由諫山 創作画。",
+                            "id": "853",
+                            "_vectors": {
+                              "manual": [
+                                "1.0",
+                                "2.0",
+                                "3.0"
+                              ]
+                            }
+                          }
+                        }
+                      ],
+                      "query": "\"进击的巨人\"",
+                      "processingTimeMs": "[duration]",
+                      "limit": 20,
+                      "offset": 0,
+                      "estimatedTotalHits": 1
+                    }
+                    "###);
+                    snapshot!(code, @"200 OK");
+                },
+            )
+            .await;
+}
+
+#[actix_rt::test]
+async fn auto_infer_locales_at_search() {
+    let server = Server::new().await;
+
+    let index = server.index("test");
+    let documents = DOCUMENTS.clone();
+    let (response, _) = index
+        .update_settings(
+            json!({
+                "searchableAttributes": ["name_en", "name_ja", "name_zh", "author_en", "author_ja", "author_zh", "description_en", "description_ja", "description_zh"],
+                "localizedAttributes": [
+                    // force japanese
+                    {"attributePatterns": ["*"], "locales": ["jpn"]},
+                ]
+            }),
+        )
+        .await;
+    snapshot!(response, @r###"
+    {
+      "taskUid": 0,
+      "indexUid": "test",
+      "status": "enqueued",
+      "type": "settingsUpdate",
+      "enqueuedAt": "[date]"
+    }
+    "###);
+    index.add_documents(documents, None).await;
+    index.wait_task(1).await;
+
+    index
+        .search(
+            json!({"q": "\"进击的巨人\"", "attributesToHighlight": ["*"]}),
+            |response, code| {
+                snapshot!(response, @r###"
+                {
+                  "hits": [
+                    {
+                      "name_zh": "进击的巨人",
+                      "author_zh": "諫山創",
+                      "description_zh": "进击的巨人是日本的漫画系列，由諫山 創作画。",
+                      "id": 853,
+                      "_vectors": {
+                        "manual": [
+                          1.0,
+                          2.0,
+                          3.0
+                        ]
+                      },
+                      "_formatted": {
+                        "name_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>",
+                        "author_zh": "諫山創",
+                        "description_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>是日本的漫画系列，由諫山 創作画。",
+                        "id": "853",
+                        "_vectors": {
+                          "manual": [
+                            "1.0",
+                            "2.0",
+                            "3.0"
+                          ]
+                        }
+                      }
+                    }
+                  ],
+                  "query": "\"进击的巨人\"",
+                  "processingTimeMs": "[duration]",
+                  "limit": 20,
+                  "offset": 0,
+                  "estimatedTotalHits": 1
+                }
+                "###);
+                snapshot!(code, @"200 OK");
+            },
+        )
+        .await;
+
+    index
+            .search(
+                json!({"q": "\"进击的巨人\"", "attributesToHighlight": ["*"]}),
+                |response, code| {
+                    snapshot!(response, @r###"
+                    {
+                      "hits": [
+                        {
+                          "name_zh": "进击的巨人",
+                          "author_zh": "諫山創",
+                          "description_zh": "进击的巨人是日本的漫画系列，由諫山 創作画。",
+                          "id": 853,
+                          "_vectors": {
+                            "manual": [
+                              1.0,
+                              2.0,
+                              3.0
+                            ]
+                          },
+                          "_formatted": {
+                            "name_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>",
+                            "author_zh": "諫山創",
+                            "description_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>是日本的漫画系列，由諫山 創作画。",
+                            "id": "853",
+                            "_vectors": {
+                              "manual": [
+                                "1.0",
+                                "2.0",
+                                "3.0"
+                              ]
+                            }
+                          }
+                        }
+                      ],
+                      "query": "\"进击的巨人\"",
+                      "processingTimeMs": "[duration]",
+                      "limit": 20,
+                      "offset": 0,
+                      "estimatedTotalHits": 1
+                    }
+                    "###);
+                    snapshot!(code, @"200 OK");
+                },
+            )
+            .await;
+
+    index
+        .search(
+            json!({"q": "\"进击的巨人\"", "attributesToHighlight": ["*"]}),
+            |response, code| {
+                snapshot!(response, @r###"
+                {
+                  "hits": [
+                    {
+                      "name_zh": "进击的巨人",
+                      "author_zh": "諫山創",
+                      "description_zh": "进击的巨人是日本的漫画系列，由諫山 創作画。",
+                      "id": 853,
+                      "_vectors": {
+                        "manual": [
+                          1.0,
+                          2.0,
+                          3.0
+                        ]
+                      },
+                      "_formatted": {
+                        "name_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>",
+                        "author_zh": "諫山創",
+                        "description_zh": "<em>进</em><em>击</em><em>的</em><em>巨人</em>是日本的漫画系列，由諫山 創作画。",
+                        "id": "853",
+                        "_vectors": {
+                          "manual": [
+                            "1.0",
+                            "2.0",
+                            "3.0"
+                          ]
+                        }
+                      }
+                    }
+                  ],
+                  "query": "\"进击的巨人\"",
+                  "processingTimeMs": "[duration]",
+                  "limit": 20,
+                  "offset": 0,
+                  "estimatedTotalHits": 1
+                }
+                "###);
+                snapshot!(code, @"200 OK");
+            },
+        )
+        .await;
+}
+
 #[actix_rt::test]
 async fn force_different_locales_with_pattern_nested() {
     let server = Server::new().await;

meilisearch/tests/search/mod.rs

@@ -7,6 +7,7 @@ mod facet_search;
 mod formatted;
 mod geo;
 mod hybrid;
+#[cfg(not(feature = "chinese-pinyin"))]
 mod locales;
 mod matching_strategy;
 mod multi;
@@ -392,6 +393,7 @@ async fn negative_special_cases_search() {
 }
 
 #[cfg(feature = "default")]
+#[cfg(not(feature = "chinese-pinyin"))]
 #[actix_rt::test]
 async fn test_kanji_language_detection() {
     let server = Server::new().await;

milli/src/localized_attributes_rules.rs

@@ -90,6 +90,21 @@ impl LocalizedFieldIds {
     pub fn locales(&self, fields_id: FieldId) -> Option<&[Language]> {
         self.field_id_to_locales.get(&fields_id).map(Vec::as_slice)
     }
+
+    pub fn all_locales(&self) -> Vec<Language> {
+        let mut locales = Vec::new();
+        for field_locales in self.field_id_to_locales.values() {
+            if !field_locales.is_empty() {
+                locales.extend(field_locales);
+            } else {
+                // If a field has no locales, we consider it as not localized
+                return Vec::new();
+            }
+        }
+        locales.sort();
+        locales.dedup();
+        locales
+    }
 }
 
 #[cfg(test)]

milli/src/search/mod.rs

@@ -360,6 +360,7 @@ mod test {
     use super::*;
 
     #[cfg(feature = "japanese")]
+    #[cfg(not(feature = "chinese-pinyin"))]
     #[test]
     fn test_kanji_language_detection() {
         use crate::index::tests::TempIndex;

milli/src/search/new/mod.rs

@@ -49,6 +49,7 @@ pub use self::geo_sort::Strategy as GeoSortStrategy;
 use self::graph_based_ranking_rule::Words;
 use self::interner::Interned;
 use self::vector_sort::VectorSort;
+use crate::localized_attributes_rules::LocalizedFieldIds;
 use crate::score_details::{ScoreDetails, ScoringStrategy};
 use crate::search::new::distinct::apply_distinct_rule;
 use crate::vector::Embedder;
@@ -671,9 +672,44 @@ pub fn execute_search(
             tokbuilder.words_dict(dictionary);
         }
 
-        if let Some(locales) = locales {
+        let db_locales;
-            tokbuilder.allow_list(locales);
+        match locales {
-        }
+            Some(locales) => {
+                if !locales.is_empty() {
+                    tokbuilder.allow_list(locales);
+                }
+            }
+            None => {
+                // If no locales are specified, we use the locales specified in the localized attributes rules
+                let localized_attributes_rules = ctx.index.localized_attributes_rules(ctx.txn)?;
+                let fields_ids_map = ctx.index.fields_ids_map(ctx.txn)?;
+                let searchable_fields = ctx.index.searchable_fields_ids(ctx.txn)?;
+
+                let localized_fields = match &ctx.restricted_fids {
+                    // if AttributeToSearchOn is set, use the restricted list of ids
+                    Some(restricted_fids) => {
+                        let iter = restricted_fids
+                            .exact
+                            .iter()
+                            .chain(restricted_fids.tolerant.iter())
+                            .map(|(fid, _)| *fid);
+
+                        LocalizedFieldIds::new(&localized_attributes_rules, &fields_ids_map, iter)
+                    }
+                    // Otherwise use the full list of ids coming from the index searchable fields
+                    None => LocalizedFieldIds::new(
+                        &localized_attributes_rules,
+                        &fields_ids_map,
+                        searchable_fields.into_iter(),
+                    ),
+                };
+
+                db_locales = localized_fields.all_locales();
+                if !db_locales.is_empty() {
+                    tokbuilder.allow_list(&db_locales);
+                }
+            }
+        };
 
         let tokenizer = tokbuilder.build();
         drop(entered);

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

@@ -6,6 +6,7 @@ pub mod exactness;
 pub mod geo_sort;
 pub mod integration;
 #[cfg(feature = "all-tokenizations")]
+#[cfg(not(feature = "chinese-pinyin"))]
 pub mod language;
 pub mod ngram_split_words;
 pub mod proximity;