From 19884162957e6c71eb95d4b5a1f989be442c1a1c Mon Sep 17 00:00:00 2001
From: many <maxime@meilisearch.com>
Date: Tue, 28 Sep 2021 12:05:11 +0200
Subject: [PATCH 1/2] Add failing test related to Meilisearch#1714

---
 milli/src/update/index_documents/mod.rs | 37 +++++++++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/milli/src/update/index_documents/mod.rs b/milli/src/update/index_documents/mod.rs
index b00dbf375..498a2a85d 100644
--- a/milli/src/update/index_documents/mod.rs
+++ b/milli/src/update/index_documents/mod.rs
@@ -981,4 +981,41 @@ mod tests {
         let count = index.number_of_documents(&rtxn).unwrap();
         assert_eq!(count, 4);
     }
+
+    #[test]
+    fn test_meilisearch_1714() {
+        let path = tempfile::tempdir().unwrap();
+        let mut options = EnvOpenOptions::new();
+        options.map_size(10 * 1024 * 1024); // 10 MB
+        let index = Index::new(options, &path).unwrap();
+
+        let content = documents!([
+          {"id": "123", "title": "小化妆包" },
+          {"id": "456", "title": "Ipad 包" }
+        ]);
+
+        let mut wtxn = index.write_txn().unwrap();
+        let builder = IndexDocuments::new(&mut wtxn, &index, 0);
+        builder.execute(content, |_, _| ()).unwrap();
+        wtxn.commit().unwrap();
+
+        let rtxn = index.read_txn().unwrap();
+
+        // Only the first document should match.
+        let count = index.word_docids.get(&rtxn, "化妆包").unwrap().unwrap().len();
+        assert_eq!(count, 1);
+
+        // Only the second document should match.
+        let count = index.word_docids.get(&rtxn, "包").unwrap().unwrap().len();
+        assert_eq!(count, 1);
+
+        let mut search = crate::Search::new(&rtxn, &index);
+        search.query("化妆包");
+        search.authorize_typos(true);
+        search.optional_words(true);
+
+        // only 1 document should be returned
+        let crate::SearchResult { documents_ids, .. } = search.execute().unwrap();
+        assert_eq!(documents_ids.len(), 1);
+    }
 }

From 8046ae4bd570b092a98c424dfbd4f5ac3e0678cc Mon Sep 17 00:00:00 2001
From: many <maxime@meilisearch.com>
Date: Tue, 28 Sep 2021 12:10:43 +0200
Subject: [PATCH 2/2] Count the number of char instead of counting bytes to
 assign the typo tolerance

---
 milli/src/search/query_tree.rs | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/milli/src/search/query_tree.rs b/milli/src/search/query_tree.rs
index 8fa24b9d3..0744231ae 100644
--- a/milli/src/search/query_tree.rs
+++ b/milli/src/search/query_tree.rs
@@ -262,7 +262,7 @@ fn split_best_frequency(ctx: &impl Context, word: &str) -> heed::Result<Option<O
 /// and the provided word length.
 fn typos(word: String, authorize_typos: bool) -> QueryKind {
     if authorize_typos {
-        match word.len() {
+        match word.chars().count() {
             0..=4 => QueryKind::exact(word),
             5..=8 => QueryKind::tolerant(1, word),
             _ => QueryKind::tolerant(2, word),