rust/tantivy
1
0
Fork 0
mirror of https://github.com/quickwit-oss/tantivy.git synced 2026-05-31 23:50:41 +00:00
Code Issues Packages Projects Releases Wiki Activity

Split buffered refill from scorer removal

Browse Source
This commit is contained in:
pascal
2026-05-31 12:20:07 +02:00
parent 34fec8b23e
commit 24a97dbe69
1 changed files with 28 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
src/query/union/buffered_union.rs
View File

@@ -11,23 +11,6 @@ use crate::{DocId, Score};
const HORIZON_NUM_TINYBITSETS: usize = HORIZON as usize / 64;
const HORIZON: u32 = 64u32 * 64u32;
// `drain_filter` is not stable yet.
// This function is similar except that it does is not unstable, and
// it does not keep the original vector ordering.
//
// Elements are dropped and not yielded.
fn unordered_drain_filter<T, P>(v: &mut Vec<T>, mut predicate: P)
where P: FnMut(&mut T) -> bool {
let mut i = 0;
while i < v.len() {
if predicate(&mut v[i]) {
v.swap_remove(i);
} else {
i += 1;
}
}
}
/// Creates a `DocSet` that iterate through the union of two or more `DocSet`s.
pub struct BufferedUnionScorer<TScorer, TScoreCombiner = DoNothingCombiner> {
/// Active scorers (already filtered of `TERMINATED`).
@@ -55,28 +38,37 @@ pub struct BufferedUnionScorer<TScorer, TScoreCombiner = DoNothingCombiner> {
num_docs: u32,
}
fn refill_scorer<TScorer: Scorer, TScoreCombiner: ScoreCombiner>(
scorer: &mut TScorer,
bitsets: &mut [TinySet; HORIZON_NUM_TINYBITSETS],
score_combiner: &mut [TScoreCombiner; HORIZON as usize],
min_doc: DocId,
horizon: DocId,
) {
loop {
let doc = scorer.doc();
if doc >= horizon {
break;
}
// add this document
let delta = doc - min_doc;
bitsets[(delta / 64) as usize].insert_mut(delta % 64u32);
score_combiner[delta as usize].update(scorer);
scorer.advance();
}
}
fn refill<TScorer: Scorer, TScoreCombiner: ScoreCombiner>(
scorers: &mut Vec<TScorer>,
bitsets: &mut [TinySet; HORIZON_NUM_TINYBITSETS],
score_combiner: &mut [TScoreCombiner; HORIZON as usize],
min_doc: DocId,
) {
unordered_drain_filter(scorers, |scorer| {
let horizon = min_doc + HORIZON;
loop {
let doc = scorer.doc();
if doc >= horizon {
break;
}
// add this document
let delta = doc - min_doc;
bitsets[(delta / 64) as usize].insert_mut(delta % 64u32);
score_combiner[delta as usize].update(scorer);
scorer.advance();
}
// remove the docset if it has been entirely consumed.
scorer.doc() == TERMINATED
});
let horizon = min_doc + HORIZON;
for scorer in scorers.iter_mut() {
refill_scorer(scorer, bitsets, score_combiner, min_doc, horizon);
}
scorers.retain(|scorer| scorer.doc() != TERMINATED);
}
impl<TScorer: Scorer, TScoreCombiner: ScoreCombiner> BufferedUnionScorer<TScorer, TScoreCombiner> {
@@ -247,12 +239,12 @@ where
// The target is outside of the buffered horizon.
// advance all docsets to a doc >= to the target.
unordered_drain_filter(&mut self.docsets, |docset| {
for docset in &mut self.docsets {
if docset.doc() < target {
docset.seek(target);
}
docset.doc() == TERMINATED
});
}
self.docsets.retain(|docset| docset.doc() != TERMINATED);
// at this point all of the docsets
// are positioned on a doc >= to the target.