From 0d2e7bcc130ce187721d275752c94ed9f02fb395 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Cl=C3=A9ment=20Renault?= <clement@meilisearch.com>
Date: Thu, 30 Mar 2023 16:10:10 +0200
Subject: [PATCH] Implement the previous way for the exhaustive distinct
 candidates

---
 milli/src/search/mod.rs     | 11 +++++++----
 milli/src/search/new/mod.rs | 16 ++++++++++++++--
 2 files changed, 21 insertions(+), 6 deletions(-)

diff --git a/milli/src/search/mod.rs b/milli/src/search/mod.rs
index 734671990..08803b73f 100644
--- a/milli/src/search/mod.rs
+++ b/milli/src/search/mod.rs
@@ -1,3 +1,9 @@
+use std::fmt;
+
+use levenshtein_automata::{LevenshteinAutomatonBuilder as LevBuilder, DFA};
+use once_cell::sync::Lazy;
+use roaring::bitmap::RoaringBitmap;
+
 pub use self::facet::{FacetDistribution, Filter, DEFAULT_VALUES_PER_FACET};
 pub use self::matches::{
     FormatOptions, MatchBounds, Matcher, MatcherBuilder, MatchingWord, MatchingWords,
@@ -5,10 +11,6 @@ pub use self::matches::{
 use crate::{
     execute_search, AscDesc, DefaultSearchLogger, DocumentId, Index, Result, SearchContext,
 };
-use levenshtein_automata::{LevenshteinAutomatonBuilder as LevBuilder, DFA};
-use once_cell::sync::Lazy;
-use roaring::bitmap::RoaringBitmap;
-use std::fmt;
 
 // Building these factories is not free.
 static LEVDIST0: Lazy<LevBuilder> = Lazy::new(|| LevBuilder::new(0, true));
@@ -112,6 +114,7 @@ impl<'a> Search<'a> {
             &mut ctx,
             &self.query,
             self.terms_matching_strategy,
+            self.exhaustive_number_hits,
             &self.filter,
             &self.sort_criteria,
             self.offset,
diff --git a/milli/src/search/new/mod.rs b/milli/src/search/new/mod.rs
index 4061df162..4d2805fef 100644
--- a/milli/src/search/new/mod.rs
+++ b/milli/src/search/new/mod.rs
@@ -37,6 +37,7 @@ use self::interner::Interner;
 use self::ranking_rules::{BoxRankingRule, RankingRule};
 use self::resolve_query_graph::compute_query_graph_docids;
 use self::sort::Sort;
+use crate::search::new::distinct::{apply_distinct_rule, DistinctOutput};
 use crate::{
     AscDesc, Filter, Index, MatchingWords, Member, Result, SearchResult, TermsMatchingStrategy,
     UserError,
@@ -272,6 +273,7 @@ pub fn execute_search(
     ctx: &mut SearchContext,
     query: &Option<String>,
     terms_matching_strategy: TermsMatchingStrategy,
+    exhaustive_number_hits: bool,
     filters: &Option<Filter>,
     sort_criteria: &Option<Vec<AscDesc>>,
     from: usize,
@@ -333,11 +335,21 @@ pub fn execute_search(
         )?
     };
 
+    // The candidates is the universe unless the exhaustive number of hits
+    // is requested and a distinct attribute is set.
+    let mut candidates = universe;
+    if exhaustive_number_hits {
+        if let Some(f) = ctx.index.distinct_field(ctx.txn)? {
+            if let Some(distinct_fid) = ctx.index.fields_ids_map(ctx.txn)?.id(f) {
+                candidates = apply_distinct_rule(ctx, distinct_fid, &candidates)?.remaining;
+            }
+        }
+    }
+
     Ok(SearchResult {
         // TODO: correct matching words
         matching_words: MatchingWords::default(),
-        // TODO: candidates with distinct
-        candidates: universe,
+        candidates,
         documents_ids,
     })
 }