searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-12-04 18:45:46 +01:00
Code Issues Releases Wiki Activity

Fix word pair proximity

Browse Source
This commit is contained in:
ManyTheFish 2023-09-26 10:08:36 +02:00
parent 04fd44b5e2
commit 27161bcd05
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
milli/src/update/index_documents/extract/extract_word_pair_proximity_docids.rs
View File

@ -1,4 +1,3 @@
use std::cmp::Ordering;
use std::collections::{HashMap, VecDeque}; use std::collections::{HashMap, VecDeque};
use std::fs::File; use std::fs::File;
use std::{cmp, io}; use std::{cmp, io};
@ -11,7 +10,7 @@ use super::helpers::{
}; };
use crate::error::SerializationError; use crate::error::SerializationError;
use crate::index::db_name::DOCID_WORD_POSITIONS; use crate::index::db_name::DOCID_WORD_POSITIONS;
use crate::proximity::{positions_proximity, MAX_DISTANCE}; use crate::proximity::{index_proximity, MAX_DISTANCE};
use crate::{DocumentId, Result}; use crate::{DocumentId, Result};
/// Extracts the best proximity between pairs of words and the documents ids where this pair appear. /// Extracts the best proximity between pairs of words and the documents ids where this pair appear.
@ -70,7 +69,7 @@ pub fn extract_word_pair_proximity_docids<R: io::Read + io::Seek>(
for (position, word) in KvReaderU16::new(&value).iter() { for (position, word) in KvReaderU16::new(&value).iter() {
// drain the proximity window until the head word is considered close to the word we are inserting. // drain the proximity window until the head word is considered close to the word we are inserting.
while word_positions.get(0).map_or(false, |(_w, p)| { while word_positions.get(0).map_or(false, |(_w, p)| {
positions_proximity(*p as u32, position as u32) > MAX_DISTANCE index_proximity(*p as u32, position as u32) >= MAX_DISTANCE
}) { }) {
word_positions_into_word_pair_proximity( word_positions_into_word_pair_proximity(
&mut word_positions, &mut word_positions,
@ -108,6 +107,7 @@ fn document_word_positions_into_sorter(
word_pair_proximity: &HashMap<(String, String), u8>, word_pair_proximity: &HashMap<(String, String), u8>,
word_pair_proximity_docids_sorter: &mut grenad::Sorter<MergeFn>, word_pair_proximity_docids_sorter: &mut grenad::Sorter<MergeFn>,
) -> Result<()> { ) -> Result<()> {
puffin::profile_function!();
let mut key_buffer = Vec::new(); let mut key_buffer = Vec::new();
for ((w1, w2), prox) in word_pair_proximity { for ((w1, w2), prox) in word_pair_proximity {
key_buffer.clear(); key_buffer.clear();
@ -126,9 +126,10 @@ fn word_positions_into_word_pair_proximity(
word_positions: &mut VecDeque<(String, u16)>, word_positions: &mut VecDeque<(String, u16)>,
word_pair_proximity: &mut HashMap<(String, String), u8>, word_pair_proximity: &mut HashMap<(String, String), u8>,
) -> Result<()> { ) -> Result<()> {
puffin::profile_function!();
let (head_word, head_position) = word_positions.pop_front().unwrap(); let (head_word, head_position) = word_positions.pop_front().unwrap();
for (word, position) in word_positions.iter() { for (word, position) in word_positions.iter() {
let prox = positions_proximity(head_position as u32, *position as u32) as u8; let prox = index_proximity(head_position as u32, *position as u32) as u8;
word_pair_proximity word_pair_proximity
.entry((head_word.clone(), word.clone())) .entry((head_word.clone(), word.clone()))
.and_modify(|p| { .and_modify(|p| {