Merge 41dbdd2d18 into 057fcb3993

* Remove unreachable code

* Add `indices` field to `MatchBounds`

For matches inside arrays, this field holds the indices of the array
elements that matched. For example, searching for `cat` inside
`{ "a": ["dog", "cat", "fox"] }` would return `indices: [1]`. For nested
arrays, this contains multiple indices, starting with the one for the
top-most array. For matches in fields without arrays, `indices` is not
serialized (does not exist) to save space.

crates/meilisearch/src/search/mod.rs

@@ -1735,7 +1735,10 @@ fn format_fields(
     // select the attributes to retrieve
     let displayable_names =
         displayable_ids.iter().map(|&fid| field_ids_map.name(fid).expect("Missing field name"));
-    permissive_json_pointer::map_leaf_values(&mut document, displayable_names, |key, value| {
+    permissive_json_pointer::map_leaf_values(
+        &mut document,
+        displayable_names,
+        |key, array_indices, value| {
             // To get the formatting option of each key we need to see all the rules that applies
             // to the value and merge them together. eg. If a user said he wanted to highlight `doggo`
             // and crop `doggo.name`. `doggo.name` needs to be highlighted + cropped while `doggo.age` is only
@@ -1766,6 +1769,7 @@ fn format_fields(
                 format,
                 &mut infos,
                 compute_matches,
+                array_indices,
                 locales,
             );
 
@@ -1774,7 +1778,8 @@ fn format_fields(
                     matches.insert(key.to_owned(), infos);
                 }
             }
-    });
+        },
+    );
 
     let selectors = formatted_options
         .keys()
@@ -1792,13 +1797,14 @@ fn format_value(
     format_options: Option<FormatOptions>,
     infos: &mut Vec<MatchBounds>,
     compute_matches: bool,
+    array_indices: &[usize],
     locales: Option<&[Language]>,
 ) -> Value {
     match value {
         Value::String(old_string) => {
             let mut matcher = builder.build(&old_string, locales);
             if compute_matches {
-                let matches = matcher.matches();
+                let matches = matcher.matches(array_indices);
                 infos.extend_from_slice(&matches[..]);
             }
 
@@ -1810,51 +1816,15 @@ fn format_value(
                 None => Value::String(old_string),
             }
         }
-        Value::Array(values) => Value::Array(
+        // `map_leaf_values` makes sure this is only called for leaf fields
-            values
+        Value::Array(_) => unreachable!(),
-                .into_iter()
+        Value::Object(_) => unreachable!(),
-                .map(|v| {
-                    format_value(
-                        v,
-                        builder,
-                        format_options.map(|format_options| FormatOptions {
-                            highlight: format_options.highlight,
-                            crop: None,
-                        }),
-                        infos,
-                        compute_matches,
-                        locales,
-                    )
-                })
-                .collect(),
-        ),
-        Value::Object(object) => Value::Object(
-            object
-                .into_iter()
-                .map(|(k, v)| {
-                    (
-                        k,
-                        format_value(
-                            v,
-                            builder,
-                            format_options.map(|format_options| FormatOptions {
-                                highlight: format_options.highlight,
-                                crop: None,
-                            }),
-                            infos,
-                            compute_matches,
-                            locales,
-                        ),
-                    )
-                })
-                .collect(),
-        ),
         Value::Number(number) => {
             let s = number.to_string();
 
             let mut matcher = builder.build(&s, locales);
             if compute_matches {
-                let matches = matcher.matches();
+                let matches = matcher.matches(array_indices);
                 infos.extend_from_slice(&matches[..]);
             }
 

crates/meilisearch/tests/search/formatted.rs

@@ -208,7 +208,10 @@ async fn format_nested() {
                         "doggos.name": [
                           {
                             "start": 0,
-                            "length": 5
+                            "length": 5,
+                            "indices": [
+                              0
+                            ]
                           }
                         ]
                       }

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

@@ -104,6 +104,8 @@ impl FormatOptions {
 pub struct MatchBounds {
     pub start: usize,
     pub length: usize,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub indices: Option<Vec<usize>>,
 }
 
 /// Structure used to analyze a string, compute words that match,
@@ -219,15 +221,20 @@ impl<'t, 'tokenizer> Matcher<'t, 'tokenizer, '_, '_> {
     }
 
     /// Returns boundaries of the words that match the query.
-    pub fn matches(&mut self) -> Vec<MatchBounds> {
+    pub fn matches(&mut self, array_indices: &[usize]) -> Vec<MatchBounds> {
         match &self.matches {
-            None => self.compute_matches().matches(),
+            None => self.compute_matches().matches(array_indices),
             Some((tokens, matches)) => matches
                 .iter()
                 .map(|m| MatchBounds {
                     start: tokens[m.get_first_token_pos()].byte_start,
                     // TODO: Why is this in chars, while start is in bytes?
                     length: m.char_count,
+                    indices: if array_indices.is_empty() {
+                        None
+                    } else {
+                        Some(array_indices.to_owned())
+                    },
                 })
                 .collect(),
         }

crates/permissive-json-pointer/src/lib.rs

@@ -45,7 +45,7 @@ fn contained_in(selector: &str, key: &str) -> bool {
 /// map_leaf_values(
 ///     value.as_object_mut().unwrap(),
 ///     ["jean.race.name"],
-///     |key, value| match (value, key) {
+///     |key, _array_indices, value| match (value, key) {
 ///         (Value::String(name), "jean.race.name") => *name = "patou".to_string(),
 ///         _ => unreachable!(),
 ///     },
@@ -66,17 +66,18 @@ fn contained_in(selector: &str, key: &str) -> bool {
 pub fn map_leaf_values<'a>(
     value: &mut Map<String, Value>,
     selectors: impl IntoIterator<Item = &'a str>,
-    mut mapper: impl FnMut(&str, &mut Value),
+    mut mapper: impl FnMut(&str, &[usize], &mut Value),
 ) {
     let selectors: Vec<_> = selectors.into_iter().collect();
-    map_leaf_values_in_object(value, &selectors, "", &mut mapper);
+    map_leaf_values_in_object(value, &selectors, "", &[], &mut mapper);
 }
 
 pub fn map_leaf_values_in_object(
     value: &mut Map<String, Value>,
     selectors: &[&str],
     base_key: &str,
-    mapper: &mut impl FnMut(&str, &mut Value),
+    array_indices: &[usize],
+    mapper: &mut impl FnMut(&str, &[usize], &mut Value),
 ) {
     for (key, value) in value.iter_mut() {
         let base_key = if base_key.is_empty() {
@@ -94,12 +95,12 @@ pub fn map_leaf_values_in_object(
         if should_continue {
             match value {
                 Value::Object(object) => {
-                    map_leaf_values_in_object(object, selectors, &base_key, mapper)
+                    map_leaf_values_in_object(object, selectors, &base_key, array_indices, mapper)
                 }
                 Value::Array(array) => {
-                    map_leaf_values_in_array(array, selectors, &base_key, mapper)
+                    map_leaf_values_in_array(array, selectors, &base_key, array_indices, mapper)
                 }
-                value => mapper(&base_key, value),
+                value => mapper(&base_key, array_indices, value),
             }
         }
     }
@@ -109,13 +110,24 @@ pub fn map_leaf_values_in_array(
     values: &mut [Value],
     selectors: &[&str],
     base_key: &str,
-    mapper: &mut impl FnMut(&str, &mut Value),
+    base_array_indices: &[usize],
+    mapper: &mut impl FnMut(&str, &[usize], &mut Value),
 ) {
-    for value in values.iter_mut() {
+    // This avoids allocating twice
+    let mut array_indices = Vec::with_capacity(base_array_indices.len() + 1);
+    array_indices.extend_from_slice(base_array_indices);
+    array_indices.push(0);
+
+    for (i, value) in values.iter_mut().enumerate() {
+        *array_indices.last_mut().unwrap() = i;
         match value {
-            Value::Object(object) => map_leaf_values_in_object(object, selectors, base_key, mapper),
+            Value::Object(object) => {
-            Value::Array(array) => map_leaf_values_in_array(array, selectors, base_key, mapper),
+                map_leaf_values_in_object(object, selectors, base_key, &array_indices, mapper)
-            value => mapper(base_key, value),
+            }
+            Value::Array(array) => {
+                map_leaf_values_in_array(array, selectors, base_key, &array_indices, mapper)
+            }
+            value => mapper(base_key, &array_indices, value),
         }
     }
 }
@@ -743,12 +755,14 @@ mod tests {
             }
         });
 
-        map_leaf_values(value.as_object_mut().unwrap(), ["jean.race.name"], |key, value| {
+        map_leaf_values(
-            match (value, key) {
+            value.as_object_mut().unwrap(),
+            ["jean.race.name"],
+            |key, _, value| match (value, key) {
                 (Value::String(name), "jean.race.name") => *name = S("patou"),
                 _ => unreachable!(),
-            }
+            },
-        });
+        );
 
         assert_eq!(
             value,
@@ -775,7 +789,7 @@ mod tests {
         });
 
         let mut calls = 0;
-        map_leaf_values(value.as_object_mut().unwrap(), ["jean"], |key, value| {
+        map_leaf_values(value.as_object_mut().unwrap(), ["jean"], |key, _, value| {
             calls += 1;
             match (value, key) {
                 (Value::String(name), "jean.race.name") => *name = S("patou"),
@@ -798,4 +812,52 @@ mod tests {
             })
         );
     }
+
+    #[test]
+    fn map_array() {
+        let mut value: Value = json!({
+            "no_array": "peter",
+            "simple": ["foo", "bar"],
+            "nested": [
+                {
+                    "a": [
+                        ["cat", "dog"],
+                        ["fox", "bear"],
+                    ],
+                    "b": "hi",
+                },
+                {
+                    "a": ["green", "blue"],
+                },
+            ],
+        });
+
+        map_leaf_values(
+            value.as_object_mut().unwrap(),
+            ["no_array", "simple", "nested"],
+            |_key, array_indices, value| {
+                *value = format!("{array_indices:?}").into();
+            },
+        );
+
+        assert_eq!(
+            value,
+            json!({
+                "no_array": "[]",
+                "simple": ["[0]", "[1]"],
+                "nested": [
+                    {
+                        "a": [
+                            ["[0, 0, 0]", "[0, 0, 1]"],
+                            ["[0, 1, 0]", "[0, 1, 1]"],
+                        ],
+                        "b": "[0]",
+                    },
+                    {
+                        "a": ["[1, 0]", "[1, 1]"],
+                    },
+                ],
+            })
+        );
+    }
 }