Merge branch 'search-refactor-typo-attributes' into search-refactor

milli/src/search/new/db_cache.rs

@@ -4,10 +4,14 @@ use std::hash::Hash;
 
 use fxhash::FxHashMap;
 use heed::types::ByteSlice;
-use heed::{BytesEncode, Database, RoTxn};
+use heed::{BytesDecode, BytesEncode, Database, RoTxn};
+use roaring::RoaringBitmap;
 
 use super::interner::Interned;
-use crate::{Result, SearchContext};
+use super::Word;
+use crate::{
+    CboRoaringBitmapCodec, CboRoaringBitmapLenCodec, Result, RoaringBitmapCodec, SearchContext,
+};
 
 /// A cache storing pointers to values in the LMDB databases.
 ///
@@ -25,6 +29,7 @@ pub struct DatabaseCache<'ctx> {
     pub word_docids: FxHashMap<Interned<String>, Option<&'ctx [u8]>>,
     pub exact_word_docids: FxHashMap<Interned<String>, Option<&'ctx [u8]>>,
     pub word_prefix_docids: FxHashMap<Interned<String>, Option<&'ctx [u8]>>,
+    pub exact_word_prefix_docids: FxHashMap<Interned<String>, Option<&'ctx [u8]>>,
 
     pub words_fst: Option<fst::Set<Cow<'ctx, [u8]>>>,
     pub word_position_docids: FxHashMap<(Interned<String>, u16), Option<&'ctx [u8]>>,
@@ -64,28 +69,103 @@ impl<'ctx> SearchContext<'ctx> {
         }
     }
 
+    pub fn word_docids(&mut self, word: Word) -> Result<Option<RoaringBitmap>> {
+        match word {
+            Word::Original(word) => {
+                let exact = self.get_db_exact_word_docids(word)?;
+                let tolerant = self.get_db_word_docids(word)?;
+                Ok(match (exact, tolerant) {
+                    (None, None) => None,
+                    (None, Some(tolerant)) => Some(tolerant),
+                    (Some(exact), None) => Some(exact),
+                    (Some(exact), Some(tolerant)) => {
+                        let mut both = exact;
+                        both |= tolerant;
+                        Some(both)
+                    }
+                })
+            }
+            Word::Derived(word) => self.get_db_word_docids(word),
+        }
+    }
+
     /// Retrieve or insert the given value in the `word_docids` database.
-    pub fn get_db_word_docids(&mut self, word: Interned<String>) -> Result<Option<&'ctx [u8]>> {
+    fn get_db_word_docids(&mut self, word: Interned<String>) -> Result<Option<RoaringBitmap>> {
         DatabaseCache::get_value(
             self.txn,
             word,
             self.word_interner.get(word).as_str(),
             &mut self.db_cache.word_docids,
             self.index.word_docids.remap_data_type::<ByteSlice>(),
-        )
+        )?
+        .map(|bytes| RoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
     }
+
+    fn get_db_exact_word_docids(
+        &mut self,
+        word: Interned<String>,
+    ) -> Result<Option<RoaringBitmap>> {
+        DatabaseCache::get_value(
+            self.txn,
+            word,
+            self.word_interner.get(word).as_str(),
+            &mut self.db_cache.exact_word_docids,
+            self.index.exact_word_docids.remap_data_type::<ByteSlice>(),
+        )?
+        .map(|bytes| RoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
+    }
+
+    pub fn word_prefix_docids(&mut self, prefix: Word) -> Result<Option<RoaringBitmap>> {
+        match prefix {
+            Word::Original(prefix) => {
+                let exact = self.get_db_exact_word_prefix_docids(prefix)?;
+                let tolerant = self.get_db_word_prefix_docids(prefix)?;
+                Ok(match (exact, tolerant) {
+                    (None, None) => None,
+                    (None, Some(tolerant)) => Some(tolerant),
+                    (Some(exact), None) => Some(exact),
+                    (Some(exact), Some(tolerant)) => {
+                        let mut both = exact;
+                        both |= tolerant;
+                        Some(both)
+                    }
+                })
+            }
+            Word::Derived(prefix) => self.get_db_word_prefix_docids(prefix),
+        }
+    }
+
     /// Retrieve or insert the given value in the `word_prefix_docids` database.
-    pub fn get_db_word_prefix_docids(
+    fn get_db_word_prefix_docids(
         &mut self,
         prefix: Interned<String>,
-    ) -> Result<Option<&'ctx [u8]>> {
+    ) -> Result<Option<RoaringBitmap>> {
         DatabaseCache::get_value(
             self.txn,
             prefix,
             self.word_interner.get(prefix).as_str(),
             &mut self.db_cache.word_prefix_docids,
             self.index.word_prefix_docids.remap_data_type::<ByteSlice>(),
-        )
+        )?
+        .map(|bytes| RoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
+    }
+
+    fn get_db_exact_word_prefix_docids(
+        &mut self,
+        prefix: Interned<String>,
+    ) -> Result<Option<RoaringBitmap>> {
+        DatabaseCache::get_value(
+            self.txn,
+            prefix,
+            self.word_interner.get(prefix).as_str(),
+            &mut self.db_cache.exact_word_prefix_docids,
+            self.index.exact_word_prefix_docids.remap_data_type::<ByteSlice>(),
+        )?
+        .map(|bytes| RoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
     }
 
     pub fn get_db_word_pair_proximity_docids(
@@ -93,7 +173,7 @@ impl<'ctx> SearchContext<'ctx> {
         word1: Interned<String>,
         word2: Interned<String>,
         proximity: u8,
-    ) -> Result<Option<&'ctx [u8]>> {
+    ) -> Result<Option<RoaringBitmap>> {
         DatabaseCache::get_value(
             self.txn,
             (proximity, word1, word2),
@@ -104,7 +184,32 @@ impl<'ctx> SearchContext<'ctx> {
             ),
             &mut self.db_cache.word_pair_proximity_docids,
             self.index.word_pair_proximity_docids.remap_data_type::<ByteSlice>(),
-        )
+        )?
+        .map(|bytes| CboRoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
+    }
+
+    pub fn get_db_word_pair_proximity_docids_len(
+        &mut self,
+        word1: Interned<String>,
+        word2: Interned<String>,
+        proximity: u8,
+    ) -> Result<Option<u64>> {
+        DatabaseCache::get_value(
+            self.txn,
+            (proximity, word1, word2),
+            &(
+                proximity,
+                self.word_interner.get(word1).as_str(),
+                self.word_interner.get(word2).as_str(),
+            ),
+            &mut self.db_cache.word_pair_proximity_docids,
+            self.index.word_pair_proximity_docids.remap_data_type::<ByteSlice>(),
+        )?
+        .map(|bytes| {
+            CboRoaringBitmapLenCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into())
+        })
+        .transpose()
     }
 
     pub fn get_db_word_prefix_pair_proximity_docids(
@@ -112,7 +217,7 @@ impl<'ctx> SearchContext<'ctx> {
         word1: Interned<String>,
         prefix2: Interned<String>,
         proximity: u8,
-    ) -> Result<Option<&'ctx [u8]>> {
+    ) -> Result<Option<RoaringBitmap>> {
         DatabaseCache::get_value(
             self.txn,
             (proximity, word1, prefix2),
@@ -123,14 +228,16 @@ impl<'ctx> SearchContext<'ctx> {
             ),
             &mut self.db_cache.word_prefix_pair_proximity_docids,
             self.index.word_prefix_pair_proximity_docids.remap_data_type::<ByteSlice>(),
-        )
+        )?
+        .map(|bytes| CboRoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
     }
     pub fn get_db_prefix_word_pair_proximity_docids(
         &mut self,
         left_prefix: Interned<String>,
         right: Interned<String>,
         proximity: u8,
-    ) -> Result<Option<&'ctx [u8]>> {
+    ) -> Result<Option<RoaringBitmap>> {
         DatabaseCache::get_value(
             self.txn,
             (proximity, left_prefix, right),
@@ -141,34 +248,40 @@ impl<'ctx> SearchContext<'ctx> {
             ),
             &mut self.db_cache.prefix_word_pair_proximity_docids,
             self.index.prefix_word_pair_proximity_docids.remap_data_type::<ByteSlice>(),
-        )
+        )?
+        .map(|bytes| CboRoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
     }
 
     pub fn get_db_word_position_docids(
         &mut self,
         word: Interned<String>,
         position: u16,
-    ) -> Result<Option<&'ctx [u8]>> {
+    ) -> Result<Option<RoaringBitmap>> {
         DatabaseCache::get_value(
             self.txn,
             (word, position),
             &(self.word_interner.get(word).as_str(), position),
             &mut self.db_cache.word_position_docids,
             self.index.word_position_docids.remap_data_type::<ByteSlice>(),
-        )
+        )?
+        .map(|bytes| CboRoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
     }
 
     pub fn get_db_word_fid_docids(
         &mut self,
         word: Interned<String>,
         fid: u16,
-    ) -> Result<Option<&'ctx [u8]>> {
+    ) -> Result<Option<RoaringBitmap>> {
         DatabaseCache::get_value(
             self.txn,
             (word, fid),
             &(self.word_interner.get(word).as_str(), fid),
             &mut self.db_cache.word_fid_docids,
             self.index.word_fid_docids.remap_data_type::<ByteSlice>(),
-        )
+        )?
+        .map(|bytes| CboRoaringBitmapCodec::bytes_decode(bytes).ok_or(heed::Error::Decoding.into()))
+        .transpose()
     }
 }

milli/src/search/new/exact_attribute.rs

@@ -1,11 +1,10 @@
-use heed::BytesDecode;
 use roaring::{MultiOps, RoaringBitmap};
 
 use super::query_graph::QueryGraph;
 use super::ranking_rules::{RankingRule, RankingRuleOutput};
 use crate::search::new::query_graph::QueryNodeData;
 use crate::search::new::query_term::ExactTerm;
-use crate::{CboRoaringBitmapCodec, Result, SearchContext, SearchLogger};
+use crate::{Result, SearchContext, SearchLogger};
 
 /// A ranking rule that produces 3 disjoint buckets:
 ///
@@ -161,10 +160,8 @@ impl State {
                 // Note: Since the position is stored bucketed in word_position_docids, for queries with a lot of
                 // longer phrases we'll be losing on precision here.
                 let bucketed_position = crate::bucketed_position(position + offset);
-                let word_position_docids = CboRoaringBitmapCodec::bytes_decode(
+                let word_position_docids =
-                    ctx.get_db_word_position_docids(*word, bucketed_position)?.unwrap_or_default(),
+                    ctx.get_db_word_position_docids(*word, bucketed_position)?.unwrap_or_default();
-                )
-                .unwrap_or_default();
                 candidates &= word_position_docids;
                 if candidates.is_empty() {
                     return Ok(State::Empty(query_graph.clone()));
@@ -191,11 +188,7 @@ impl State {
                     // ignore stop words words in phrases
                     .flatten()
                     .map(|word| -> Result<_> {
-                        Ok(ctx
+                        Ok(ctx.get_db_word_fid_docids(*word, fid)?.unwrap_or_default())
-                            .get_db_word_fid_docids(*word, fid)?
-                            .map(CboRoaringBitmapCodec::bytes_decode)
-                            .unwrap_or_default()
-                            .unwrap_or_default())
                     }),
             )?;
             intersection &= &candidates;

milli/src/search/new/logger/visual.rs

@@ -427,14 +427,14 @@ fill: \"#B6E2D3\"
                 )?;
 
                 for w in term_subset.all_single_words_except_prefix_db(ctx)? {
-                    let w = ctx.word_interner.get(w);
+                    let w = ctx.word_interner.get(w.interned());
                     writeln!(file, "{w}: word")?;
                 }
                 for p in term_subset.all_phrases(ctx)? {
                     writeln!(file, "{}: phrase", p.description(ctx))?;
                 }
                 if let Some(w) = term_subset.use_prefix_db(ctx) {
-                    let w = ctx.word_interner.get(w);
+                    let w = ctx.word_interner.get(w.interned());
                     writeln!(file, "{w}: prefix db")?;
                 }
 

milli/src/search/new/mod.rs

@@ -50,6 +50,8 @@ use ranking_rules::{BoxRankingRule, RankingRule};
 use resolve_query_graph::compute_query_graph_docids;
 use sort::Sort;
 
+use self::interner::Interned;
+
 /// A structure used throughout the execution of a search query.
 pub struct SearchContext<'ctx> {
     pub index: &'ctx Index,
@@ -75,6 +77,21 @@ impl<'ctx> SearchContext<'ctx> {
     }
 }
 
+#[derive(Clone, Copy, PartialEq, PartialOrd, Ord, Eq)]
+pub enum Word {
+    Original(Interned<String>),
+    Derived(Interned<String>),
+}
+
+impl Word {
+    pub fn interned(&self) -> Interned<String> {
+        match self {
+            Word::Original(word) => *word,
+            Word::Derived(word) => *word,
+        }
+    }
+}
+
 /// Apply the [`TermsMatchingStrategy`] to the query graph and resolve it.
 #[allow(clippy::too_many_arguments)]
 fn resolve_maximally_reduced_query_graph(

milli/src/search/new/query_term/compute_derivations.rs

@@ -1,17 +1,17 @@
-use fst::automaton::Str;
-use fst::{Automaton, IntoStreamer, Streamer};
-use heed::types::DecodeIgnore;
-use heed::BytesDecode;
 use std::borrow::Cow;
 use std::collections::BTreeSet;
 use std::ops::ControlFlow;
 
+use fst::automaton::Str;
+use fst::{Automaton, IntoStreamer, Streamer};
+use heed::types::DecodeIgnore;
+
 use super::*;
 use crate::search::fst_utils::{Complement, Intersection, StartsWith, Union};
 use crate::search::new::query_term::TwoTypoTerm;
 use crate::search::new::{limits, SearchContext};
 use crate::search::{build_dfa, get_first};
-use crate::{CboRoaringBitmapLenCodec, Result, MAX_WORD_LENGTH};
+use crate::{Result, MAX_WORD_LENGTH};
 
 #[derive(Debug, Clone, Copy, PartialEq, Eq, Hash)]
 pub enum NumberOfTypos {
@@ -177,6 +177,7 @@ pub fn partially_initialized_term_from_word(
     word: &str,
     max_typo: u8,
     is_prefix: bool,
+    is_ngram: bool,
 ) -> Result<QueryTerm> {
     let word_interned = ctx.word_interner.insert(word.to_owned());
 
@@ -197,12 +198,19 @@ pub fn partially_initialized_term_from_word(
     let fst = ctx.index.words_fst(ctx.txn)?;
 
     let use_prefix_db = is_prefix
-        && ctx
+        && (ctx
             .index
             .word_prefix_docids
             .remap_data_type::<DecodeIgnore>()
             .get(ctx.txn, word)?
-            .is_some();
+            .is_some()
+            || (!is_ngram
+                && ctx
+                    .index
+                    .exact_word_prefix_docids
+                    .remap_data_type::<DecodeIgnore>()
+                    .get(ctx.txn, word)?
+                    .is_some()));
     let use_prefix_db = if use_prefix_db { Some(word_interned) } else { None };
 
     let mut zero_typo = None;
@@ -385,9 +393,7 @@ fn split_best_frequency(
         let left = ctx.word_interner.insert(left.to_owned());
         let right = ctx.word_interner.insert(right.to_owned());
 
-        if let Some(docid_bytes) = ctx.get_db_word_pair_proximity_docids(left, right, 1)? {
+        if let Some(frequency) = ctx.get_db_word_pair_proximity_docids_len(left, right, 1)? {
-            let frequency =
-                CboRoaringBitmapLenCodec::bytes_decode(docid_bytes).ok_or(heed::Error::Decoding)?;
             if best.map_or(true, |(old, _, _)| frequency > old) {
                 best = Some((frequency, left, right));
             }

milli/src/search/new/query_term/mod.rs

@@ -3,18 +3,18 @@ mod ntypo_subset;
 mod parse_query;
 mod phrase;
 
-use super::interner::{DedupInterner, Interned};
-use super::{limits, SearchContext};
-use crate::Result;
 use std::collections::BTreeSet;
 use std::ops::RangeInclusive;
 
+use compute_derivations::partially_initialized_term_from_word;
 use either::Either;
 pub use ntypo_subset::NTypoTermSubset;
 pub use parse_query::{located_query_terms_from_string, make_ngram, number_of_typos_allowed};
 pub use phrase::Phrase;
 
-use compute_derivations::partially_initialized_term_from_word;
+use super::interner::{DedupInterner, Interned};
+use super::{limits, SearchContext, Word};
+use crate::Result;
 
 /// A set of word derivations attached to a location in the search query.
 #[derive(Clone, PartialEq, Eq, Hash)]
@@ -159,12 +159,12 @@ impl QueryTermSubset {
         self.two_typo_subset.intersect(&other.two_typo_subset);
     }
 
-    pub fn use_prefix_db(&self, ctx: &SearchContext) -> Option<Interned<String>> {
+    pub fn use_prefix_db(&self, ctx: &SearchContext) -> Option<Word> {
         let original = ctx.term_interner.get(self.original);
         let Some(use_prefix_db) = original.zero_typo.use_prefix_db else {
             return None
         };
-        match &self.zero_typo_subset {
+        let word = match &self.zero_typo_subset {
             NTypoTermSubset::All => Some(use_prefix_db),
             NTypoTermSubset::Subset { words, phrases: _ } => {
                 // TODO: use a subset of prefix words instead
@@ -175,12 +175,19 @@ impl QueryTermSubset {
                 }
             }
             NTypoTermSubset::Nothing => None,
+        };
+        word.map(|word| {
+            if original.ngram_words.is_some() {
+                Word::Derived(word)
+            } else {
+                Word::Original(word)
             }
+        })
     }
     pub fn all_single_words_except_prefix_db(
         &self,
         ctx: &mut SearchContext,
-    ) -> Result<BTreeSet<Interned<String>>> {
+    ) -> Result<BTreeSet<Word>> {
         let mut result = BTreeSet::default();
         // TODO: a compute_partially funtion
         if !self.one_typo_subset.is_empty() || !self.two_typo_subset.is_empty() {
@@ -197,8 +204,20 @@ impl QueryTermSubset {
                     synonyms: _,
                     use_prefix_db: _,
                 } = &original.zero_typo;
-                result.extend(zero_typo.iter().copied());
+                result.extend(zero_typo.iter().copied().map(|w| {
-                result.extend(prefix_of.iter().copied());
+                    if original.ngram_words.is_some() {
+                        Word::Derived(w)
+                    } else {
+                        Word::Original(w)
+                    }
+                }));
+                result.extend(prefix_of.iter().copied().map(|w| {
+                    if original.ngram_words.is_some() {
+                        Word::Derived(w)
+                    } else {
+                        Word::Original(w)
+                    }
+                }));
             }
             NTypoTermSubset::Subset { words, phrases: _ } => {
                 let ZeroTypoTerm {
@@ -210,10 +229,20 @@ impl QueryTermSubset {
                 } = &original.zero_typo;
                 if let Some(zero_typo) = zero_typo {
                     if words.contains(zero_typo) {
-                        result.insert(*zero_typo);
+                        if original.ngram_words.is_some() {
+                            result.insert(Word::Derived(*zero_typo));
+                        } else {
+                            result.insert(Word::Original(*zero_typo));
                         }
                     }
-                result.extend(prefix_of.intersection(words).copied());
+                }
+                result.extend(prefix_of.intersection(words).copied().map(|w| {
+                    if original.ngram_words.is_some() {
+                        Word::Derived(w)
+                    } else {
+                        Word::Original(w)
+                    }
+                }));
             }
             NTypoTermSubset::Nothing => {}
         }
@@ -223,13 +252,13 @@ impl QueryTermSubset {
                 let Lazy::Init(OneTypoTerm { split_words: _, one_typo }) = &original.one_typo else {
                     panic!()
                 };
-                result.extend(one_typo.iter().copied())
+                result.extend(one_typo.iter().copied().map(Word::Derived))
             }
             NTypoTermSubset::Subset { words, phrases: _ } => {
                 let Lazy::Init(OneTypoTerm { split_words: _, one_typo }) = &original.one_typo else {
                     panic!()
                 };
-                result.extend(one_typo.intersection(words));
+                result.extend(one_typo.intersection(words).copied().map(Word::Derived));
             }
             NTypoTermSubset::Nothing => {}
         };
@@ -239,13 +268,13 @@ impl QueryTermSubset {
                 let Lazy::Init(TwoTypoTerm { two_typos }) = &original.two_typo else {
                     panic!()
                 };
-                result.extend(two_typos.iter().copied());
+                result.extend(two_typos.iter().copied().map(Word::Derived));
             }
             NTypoTermSubset::Subset { words, phrases: _ } => {
                 let Lazy::Init(TwoTypoTerm { two_typos }) = &original.two_typo else {
                     panic!()
                 };
-                result.extend(two_typos.intersection(words));
+                result.extend(two_typos.intersection(words).copied().map(Word::Derived));
             }
             NTypoTermSubset::Nothing => {}
         };

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

@@ -1,8 +1,8 @@
-use charabia::{normalizer::NormalizedTokenIter, SeparatorKind, TokenKind};
+use charabia::normalizer::NormalizedTokenIter;
-
+use charabia::{SeparatorKind, TokenKind};
-use crate::{Result, SearchContext, MAX_WORD_LENGTH};
 
 use super::*;
+use crate::{Result, SearchContext, MAX_WORD_LENGTH};
 
 /// Convert the tokenised search query into a list of located query terms.
 // TODO: checking if the positions are correct for phrases, separators, ngrams
@@ -51,6 +51,7 @@ pub fn located_query_terms_from_string(
                                 word,
                                 nbr_typos(word),
                                 false,
+                                false,
                             )?;
                             let located_term = LocatedQueryTerm {
                                 value: ctx.term_interner.push(term),
@@ -62,8 +63,13 @@ pub fn located_query_terms_from_string(
                     }
                 } else {
                     let word = token.lemma();
-                    let term =
+                    let term = partially_initialized_term_from_word(
-                        partially_initialized_term_from_word(ctx, word, nbr_typos(word), true)?;
+                        ctx,
+                        word,
+                        nbr_typos(word),
+                        true,
+                        false,
+                    )?;
                     let located_term = LocatedQueryTerm {
                         value: ctx.term_interner.push(term),
                         positions: position..=position,
@@ -195,7 +201,8 @@ pub fn make_ngram(
     let max_nbr_typos =
         number_of_typos_allowed(ngram_str.as_str()).saturating_sub(terms.len() as u8 - 1);
 
-    let mut term = partially_initialized_term_from_word(ctx, &ngram_str, max_nbr_typos, is_prefix)?;
+    let mut term =
+        partially_initialized_term_from_word(ctx, &ngram_str, max_nbr_typos, is_prefix, true)?;
 
     // Now add the synonyms
     let index_synonyms = ctx.index.synonyms(ctx.txn)?;

milli/src/search/new/ranking_rule_graph/exactness/mod.rs

@@ -1,10 +1,10 @@
-use heed::BytesDecode;
 use roaring::RoaringBitmap;
 
 use super::{ComputedCondition, RankingRuleGraphTrait};
 use crate::search::new::interner::{DedupInterner, Interned};
 use crate::search::new::query_term::{ExactTerm, LocatedQueryTermSubset};
-use crate::{Result, RoaringBitmapCodec, SearchContext};
+use crate::search::new::Word;
+use crate::{Result, SearchContext};
 
 #[derive(Clone, PartialEq, Eq, Hash)]
 pub enum ExactnessCondition {
@@ -27,8 +27,8 @@ fn compute_docids(
     let mut candidates = match exact_term {
         ExactTerm::Phrase(phrase) => ctx.get_phrase_docids(phrase)?.clone(),
         ExactTerm::Word(word) => {
-            if let Some(word_candidates) = ctx.get_db_word_docids(word)? {
+            if let Some(word_candidates) = ctx.word_docids(Word::Original(word))? {
-                RoaringBitmapCodec::bytes_decode(word_candidates).ok_or(heed::Error::Decoding)?
+                word_candidates
             } else {
                 return Ok(Default::default());
             }

milli/src/search/new/ranking_rule_graph/proximity/compute_docids.rs

@@ -2,7 +2,6 @@
 
 use std::collections::BTreeSet;
 
-use heed::BytesDecode;
 use roaring::RoaringBitmap;
 
 use super::ProximityCondition;
@@ -10,8 +9,8 @@ use crate::search::new::interner::Interned;
 use crate::search::new::query_term::{Phrase, QueryTermSubset};
 use crate::search::new::ranking_rule_graph::ComputedCondition;
 use crate::search::new::resolve_query_graph::compute_query_term_subset_docids;
-use crate::search::new::SearchContext;
+use crate::search::new::{SearchContext, Word};
-use crate::{CboRoaringBitmapCodec, Result, RoaringBitmapCodec};
+use crate::Result;
 
 pub fn compute_docids(
     ctx: &mut SearchContext,
@@ -55,8 +54,8 @@ pub fn compute_docids(
         {
             compute_prefix_edges(
                 ctx,
-                left_word,
+                left_word.interned(),
-                right_prefix,
+                right_prefix.interned(),
                 left_phrase,
                 forward_proximity,
                 backward_proximity,
@@ -92,9 +91,7 @@ pub fn compute_docids(
                 if universe.is_disjoint(ctx.get_phrase_docids(left_phrase)?) {
                     continue;
                 }
-            } else if let Some(lw_bytes) = ctx.get_db_word_docids(left_word)? {
+            } else if let Some(left_word_docids) = ctx.word_docids(left_word)? {
-                let left_word_docids =
-                    RoaringBitmapCodec::bytes_decode(lw_bytes).ok_or(heed::Error::Decoding)?;
                 if universe.is_disjoint(&left_word_docids) {
                     continue;
                 }
@@ -104,7 +101,7 @@ pub fn compute_docids(
         for (right_word, right_phrase) in right_derivs {
             compute_non_prefix_edges(
                 ctx,
-                left_word,
+                left_word.interned(),
                 right_word,
                 left_phrase,
                 right_phrase,
@@ -155,7 +152,7 @@ fn compute_prefix_edges(
     if let Some(new_docids) =
         ctx.get_db_word_prefix_pair_proximity_docids(left_word, right_prefix, forward_proximity)?
     {
-        let new_docids = &universe & CboRoaringBitmapCodec::deserialize_from(new_docids)?;
+        let new_docids = &universe & new_docids;
         if !new_docids.is_empty() {
             used_left_words.insert(left_word);
             used_right_prefix.insert(right_prefix);
@@ -170,7 +167,7 @@ fn compute_prefix_edges(
             left_word,
             backward_proximity,
         )? {
-            let new_docids = &universe & CboRoaringBitmapCodec::deserialize_from(new_docids)?;
+            let new_docids = &universe & new_docids;
             if !new_docids.is_empty() {
                 used_left_words.insert(left_word);
                 used_right_prefix.insert(right_prefix);
@@ -217,7 +214,7 @@ fn compute_non_prefix_edges(
     if let Some(new_docids) =
         ctx.get_db_word_pair_proximity_docids(word1, word2, forward_proximity)?
     {
-        let new_docids = &universe & CboRoaringBitmapCodec::deserialize_from(new_docids)?;
+        let new_docids = &universe & new_docids;
         if !new_docids.is_empty() {
             used_left_words.insert(word1);
             used_right_words.insert(word2);
@@ -231,7 +228,7 @@ fn compute_non_prefix_edges(
         if let Some(new_docids) =
             ctx.get_db_word_pair_proximity_docids(word2, word1, backward_proximity)?
         {
-            let new_docids = &universe & CboRoaringBitmapCodec::deserialize_from(new_docids)?;
+            let new_docids = &universe & new_docids;
             if !new_docids.is_empty() {
                 used_left_words.insert(word2);
                 used_right_words.insert(word1);
@@ -246,7 +243,7 @@ fn compute_non_prefix_edges(
 fn last_words_of_term_derivations(
     ctx: &mut SearchContext,
     t: &QueryTermSubset,
-) -> Result<BTreeSet<(Option<Interned<Phrase>>, Interned<String>)>> {
+) -> Result<BTreeSet<(Option<Interned<Phrase>>, Word)>> {
     let mut result = BTreeSet::new();
 
     for w in t.all_single_words_except_prefix_db(ctx)? {
@@ -256,7 +253,7 @@ fn last_words_of_term_derivations(
         let phrase = ctx.phrase_interner.get(p);
         let last_term_of_phrase = phrase.words.last().unwrap();
         if let Some(last_word) = last_term_of_phrase {
-            result.insert((Some(p), *last_word));
+            result.insert((Some(p), Word::Original(*last_word)));
         }
     }
 
@@ -269,7 +266,7 @@ fn first_word_of_term_iter(
     let mut result = BTreeSet::new();
     let all_words = t.all_single_words_except_prefix_db(ctx)?;
     for w in all_words {
-        result.insert((w, None));
+        result.insert((w.interned(), None));
     }
     for p in t.all_phrases(ctx)? {
         let phrase = ctx.phrase_interner.get(p);

milli/src/search/new/resolve_query_graph.rs

@@ -3,16 +3,15 @@
 use std::collections::VecDeque;
 
 use fxhash::FxHashMap;
-use heed::BytesDecode;
 use roaring::RoaringBitmap;
 
 use super::interner::Interned;
 use super::query_graph::QueryNodeData;
 use super::query_term::{Phrase, QueryTermSubset};
 use super::small_bitmap::SmallBitmap;
-use super::{QueryGraph, SearchContext};
+use super::{QueryGraph, SearchContext, Word};
 use crate::search::new::query_term::LocatedQueryTermSubset;
-use crate::{CboRoaringBitmapCodec, Result, RoaringBitmapCodec};
+use crate::Result;
 
 #[derive(Default)]
 pub struct PhraseDocIdsCache {
@@ -36,8 +35,8 @@ pub fn compute_query_term_subset_docids(
 ) -> Result<RoaringBitmap> {
     let mut docids = RoaringBitmap::new();
     for word in term.all_single_words_except_prefix_db(ctx)? {
-        if let Some(word_docids) = ctx.get_db_word_docids(word)? {
+        if let Some(word_docids) = ctx.word_docids(word)? {
-            docids |= RoaringBitmapCodec::bytes_decode(word_docids).ok_or(heed::Error::Decoding)?;
+            docids |= word_docids;
         }
     }
     for phrase in term.all_phrases(ctx)? {
@@ -45,9 +44,8 @@ pub fn compute_query_term_subset_docids(
     }
 
     if let Some(prefix) = term.use_prefix_db(ctx) {
-        if let Some(prefix_docids) = ctx.get_db_word_prefix_docids(prefix)? {
+        if let Some(prefix_docids) = ctx.word_prefix_docids(prefix)? {
-            docids |=
+            docids |= prefix_docids;
-                RoaringBitmapCodec::bytes_decode(prefix_docids).ok_or(heed::Error::Decoding)?;
         }
     }
 
@@ -127,9 +125,8 @@ pub fn compute_phrase_docids(
     }
     if words.len() == 1 {
         if let Some(word) = &words[0] {
-            if let Some(word_docids) = ctx.get_db_word_docids(*word)? {
+            if let Some(word_docids) = ctx.word_docids(Word::Original(*word))? {
-                return RoaringBitmapCodec::bytes_decode(word_docids)
+                return Ok(word_docids);
-                    .ok_or(heed::Error::Decoding.into());
             } else {
                 return Ok(RoaringBitmap::new());
             }
@@ -158,7 +155,7 @@ pub fn compute_phrase_docids(
             {
                 if dist == 0 {
                     match ctx.get_db_word_pair_proximity_docids(s1, s2, 1)? {
-                        Some(m) => bitmaps.push(CboRoaringBitmapCodec::deserialize_from(m)?),
+                        Some(m) => bitmaps.push(m),
                         // If there are no documents for this pair, there will be no
                         // results for the phrase query.
                         None => return Ok(RoaringBitmap::new()),
@@ -169,7 +166,7 @@ pub fn compute_phrase_docids(
                         if let Some(m) =
                             ctx.get_db_word_pair_proximity_docids(s1, s2, dist as u8 + 1)?
                         {
-                            bitmap |= CboRoaringBitmapCodec::deserialize_from(m)?;
+                            bitmap |= m;
                         }
                     }
                     if bitmap.is_empty() {