[#6778] updated the excerpt modifier to properly account for multibyte characters

CHANGELOG.md

@@ -4,6 +4,8 @@
 
 - Updated the default `COUNT` list request query to use `rowid` when possible to minimize the need of having the `id` field in a covering index.
 
+- Updated the excerpt modifier to properly account for multibyte characters ([#6778](https://github.com/pocketbase/pocketbase/issues/6778)).
+
 
 ## v0.27.1
 

tools/picker/excerpt_modifier.go

@@ -135,10 +135,14 @@ func (m *excerptModifier) Modify(value any) (any, error) {
 	result := strings.TrimSpace(builder.String())
 
 	if len(result) > m.max {
-		result = strings.TrimSpace(result[:m.max])
-
-		if m.withEllipsis {
-			result += "..."
+		// note: casted to []rune to properly account for multi-byte chars
+		runes := []rune(result)
+		if len(runes) > m.max {
+			result = string(runes[:m.max])
+			result = strings.TrimSpace(result)
+			if m.withEllipsis {
+				result += "..."
+			}
 		}
 	}
 

tools/picker/excerpt_modifier_test.go

@@ -134,6 +134,20 @@ func TestExcerptModifierModify(t *testing.T) {
 			html,
 			plainText,
 		},
+
+		// multibyte chars
+		{
+			"mutibyte chars <= max",
+			[]string{"4", "t"},
+			"аб\nв ",
+			"аб в",
+		},
+		{
+			"mutibyte chars > max",
+			[]string{"3", "t"},
+			"аб\nв ",
+			"аб...",
+		},
 	}
 
 	for _, s := range scenarios {