Merge #5187

5187: Bring back v1.12.0 of pre-release changes into `main` r=irevoire a=curquiza



Co-authored-by: ManyTheFish <many@meilisearch.com>
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
Co-authored-by: Clément Renault <clement@meilisearch.com>
Co-authored-by: meili-bors[bot] <89034592+meili-bors[bot]@users.noreply.github.com>
Co-authored-by: Many the fish <many@meilisearch.com>

crates/milli/src/search/hybrid.rs

@@ -207,7 +207,11 @@ impl<'a> Search<'a> {
                     Ok(embedding) => embedding,
                     Err(error) => {
                         tracing::error!(error=%error, "Embedding failed");
-                        return Ok((keyword_results, Some(0)));
+                        return Ok(return_keyword_results(
+                            self.limit,
+                            self.offset,
+                            keyword_results,
+                        ));
                     }
                 }
             }

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

@@ -274,7 +274,7 @@ impl<'t, 'tokenizer> Matcher<'t, 'tokenizer, '_, '_> {
                     last_match_last_token_position_plus_one
                 } else {
                     // we have matched the end of possible tokens, there's nothing to advance
-                    tokens.len() - 1
+                    tokens.len()
                 }
             };
 

crates/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::index::PrefixSearch;
 use crate::localized_attributes_rules::LocalizedFieldIds;
 use crate::score_details::{ScoreDetails, ScoringStrategy};
 use crate::search::new::distinct::apply_distinct_rule;
@@ -68,6 +69,7 @@ pub struct SearchContext<'ctx> {
     pub term_interner: Interner<QueryTerm>,
     pub phrase_docids: PhraseDocIdsCache,
     pub restricted_fids: Option<RestrictedFids>,
+    pub prefix_search: PrefixSearch,
 }
 
 impl<'ctx> SearchContext<'ctx> {
@@ -85,6 +87,8 @@ impl<'ctx> SearchContext<'ctx> {
             }
         }
 
+        let prefix_search = index.prefix_search(txn)?.unwrap_or_default();
+
         Ok(Self {
             index,
             txn,
@@ -94,9 +98,14 @@ impl<'ctx> SearchContext<'ctx> {
             term_interner: <_>::default(),
             phrase_docids: <_>::default(),
             restricted_fids: None,
+            prefix_search,
         })
     }
 
+    pub fn is_prefix_search_allowed(&self) -> bool {
+        self.prefix_search != PrefixSearch::Disabled
+    }
+
     pub fn attributes_to_search_on(
         &mut self,
         attributes_to_search_on: &'ctx [String],

crates/milli/src/search/new/query_term/parse_query.rs

@@ -28,6 +28,7 @@ pub fn located_query_terms_from_tokens(
     words_limit: Option<usize>,
 ) -> Result<ExtractedTokens> {
     let nbr_typos = number_of_typos_allowed(ctx)?;
+    let allow_prefix_search = ctx.is_prefix_search_allowed();
 
     let mut query_terms = Vec::new();
 
@@ -94,7 +95,7 @@ pub fn located_query_terms_from_tokens(
                         ctx,
                         word,
                         nbr_typos(word),
-                        true,
+                        allow_prefix_search,
                         false,
                     )?;
                     let located_term = LocatedQueryTerm {

crates/milli/src/search/new/resolve_query_graph.rs

@@ -193,15 +193,23 @@ pub fn compute_phrase_docids(
     if words.is_empty() {
         return Ok(RoaringBitmap::new());
     }
-    let mut candidates = RoaringBitmap::new();
+    let mut candidates = None;
     for word in words.iter().flatten().copied() {
         if let Some(word_docids) = ctx.word_docids(None, Word::Original(word))? {
-            candidates |= word_docids;
+            if let Some(candidates) = candidates.as_mut() {
+                *candidates &= word_docids;
+            } else {
+                candidates = Some(word_docids);
+            }
         } else {
             return Ok(RoaringBitmap::new());
         }
     }
 
+    let Some(mut candidates) = candidates else {
+        return Ok(RoaringBitmap::new());
+    };
+
     let winsize = words.len().min(3);
 
     for win in words.windows(winsize) {

crates/milli/src/search/new/tests/integration.rs

@@ -5,6 +5,7 @@ use bumpalo::Bump;
 use heed::EnvOpenOptions;
 use maplit::{btreemap, hashset};
 
+use crate::progress::Progress;
 use crate::update::new::indexer;
 use crate::update::{IndexDocumentsMethod, IndexerConfig, Settings};
 use crate::vector::EmbeddingConfigs;
@@ -72,7 +73,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
             None,
             &mut new_fields_ids_map,
             &|| false,
-            &|_progress| (),
+            Progress::default(),
         )
         .unwrap();
 
@@ -83,6 +84,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     indexer::index(
         &mut wtxn,
         &index,
+        &crate::ThreadPoolNoAbortBuilder::new().build().unwrap(),
         config.grenad_parameters(),
         &db_fields_ids_map,
         new_fields_ids_map,
@@ -90,7 +92,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
         &document_changes,
         embedders,
         &|| false,
-        &|_| (),
+        &Progress::default(),
     )
     .unwrap();