Merge #592

592: Fix(smart-crop): Ensure that matches are all highlighted r=loiclec a=ManyTheFish

In the case of only 1 query word was matching in an attribute,
and this same word was matching several times in the crop window of the same attribute,
then only the first match was highlighted.
Now we ensure that the computed "best matches interval" contains all the words that match in the same crop window.

related to meilisearch/meilisearch#2627



Co-authored-by: ManyTheFish <[email protected]>

benchmarks/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "benchmarks"
-version = "0.31.1"
+version = "0.31.2"
 edition = "2018"
 publish = false
 

cli/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "cli"
-version = "0.31.1"
+version = "0.31.2"
 edition = "2018"
 description = "A CLI to interact with a milli index"
 publish = false

filter-parser/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "filter-parser"
-version = "0.31.1"
+version = "0.31.2"
 edition = "2021"
 description = "The parser for the Meilisearch filter syntax"
 publish = false

flatten-serde-json/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "flatten-serde-json"
-version = "0.31.1"
+version = "0.31.2"
 edition = "2021"
 description = "Flatten serde-json objects like elastic search"
 readme = "README.md"

helpers/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "helpers"
-version = "0.31.1"
+version = "0.31.2"
 authors = ["Clément Renault <[email protected]>"]
 edition = "2018"
 description = "A small tool to do operations on the database"

http-ui/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "http-ui"
 description = "The HTTP user interface of the milli search engine"
-version = "0.31.1"
+version = "0.31.2"
 authors = ["Clément Renault <[email protected]>"]
 edition = "2018"
 publish = false

infos/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "infos"
-version = "0.31.1"
+version = "0.31.2"
 authors = ["Clément Renault <[email protected]>"]
 edition = "2018"
 publish = false

json-depth-checker/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "json-depth-checker"
-version = "0.31.1"
+version = "0.31.2"
 edition = "2021"
 description = "A library that indicates if a JSON must be flattened"
 publish = false

milli/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "milli"
-version = "0.31.1"
+version = "0.31.2"
 authors = ["Kerollmops <[email protected]>"]
 edition = "2018"
 

milli/src/search/matches/mod.rs

@@ -359,11 +359,11 @@ impl<'t, A: AsRef<[u8]>> Matcher<'t, '_, A> {
         if matches.len() > 1 {
             // positions of the first and the last match of the best matches interval in `matches`.
             let mut best_interval = (0, 0);
-            let mut best_interval_score = self.match_interval_score(&matches[0..=0]);
+            let mut best_interval_score = None;
             // current interval positions.
             let mut interval_first = 0;
             let mut interval_last = 0;
-            for (index, next_match) in matches.iter().enumerate().skip(1) {
+            for (index, next_match) in matches.iter().enumerate() {
                 // if next match would make interval gross more than crop_size,
                 // we compare the current interval with the best one,
                 // then we increase `interval_first` until next match can be added.
@@ -372,10 +372,15 @@ impl<'t, A: AsRef<[u8]>> Matcher<'t, '_, A> {
                         self.match_interval_score(&matches[interval_first..=interval_last]);
 
                     // keep interval if it's the best
-                    if interval_score > best_interval_score {
-                        best_interval = (interval_first, interval_last);
-                        best_interval_score = interval_score;
-                    }
+                    best_interval_score = match best_interval_score.take() {
+                        Some(best_interval_score) if interval_score <= best_interval_score => {
+                            Some(best_interval_score)
+                        }
+                        _ => {
+                            best_interval = (interval_first, interval_last);
+                            Some(interval_score)
+                        }
+                    };
 
                     // advance start of the interval while interval is longer than crop_size.
                     while next_match.word_position - matches[interval_first].word_position
@@ -390,7 +395,9 @@ impl<'t, A: AsRef<[u8]>> Matcher<'t, '_, A> {
             // compute the last interval score and compare it to the best one.
             let interval_score =
                 self.match_interval_score(&matches[interval_first..=interval_last]);
-            if interval_score > best_interval_score {
+            if best_interval_score
+                .map_or(true, |best_interval_score| interval_score > best_interval_score)
+            {
                 best_interval = (interval_first, interval_last);
             }
 
@@ -743,6 +750,19 @@ mod tests {
             &matcher.format(format_options),
             "…void void void void void <em>split</em> <em>the</em> <em>world</em> void void"
         );
+
+        // testing https://github.com/meilisearch/meilisearch/issues/2627
+        let matching_words = vec![(vec![MatchingWord::new("test".to_string(), 0, true)], vec![0])];
+        let matching_words = MatchingWords::new(matching_words);
+        let builder = MatcherBuilder::from_matching_words(matching_words);
+
+        let text = "Lorem ipsum dolor test sit amet, consetetur sadipscing elitr, sed test diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat.";
+        let mut matcher = builder.build(text);
+        let format_options = FormatOptions { highlight: true, crop: Some(20) };
+        assert_eq!(
+            &matcher.format(format_options),
+            "Lorem ipsum dolor <em>test</em> sit amet, consetetur sadipscing elitr, sed <em>test</em> diam nonumy eirmod tempor invidunt ut labore et dolore…"
+        );
     }
 
     #[test]