searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-11-23 13:24:27 +01:00
Code Issues Releases Wiki Activity

Use the field id docid facet value database when sorting documents

Browse Source
This commit is contained in:
Clément Renault 2020-12-02 21:08:48 +01:00
parent 3cdf14d4c5
commit 026f54dcf7
No known key found for this signature in database
GPG Key ID: 92ADA4E935E71FA4
1 changed files with 56 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
src/search/mod.rs
View File

@ -9,10 +9,12 @@ use levenshtein_automata::DFA;
use levenshtein_automata::LevenshteinAutomatonBuilder as LevBuilder; use levenshtein_automata::LevenshteinAutomatonBuilder as LevBuilder;
use log::debug; use log::debug;
use once_cell::sync::Lazy; use once_cell::sync::Lazy;
use ordered_float::OrderedFloat;
use roaring::bitmap::RoaringBitmap; use roaring::bitmap::RoaringBitmap;
use crate::facet::FacetType; use crate::facet::FacetType;
use crate::heed_codec::facet::{FacetLevelValueF64Codec, FacetLevelValueI64Codec}; use crate::heed_codec::facet::{FacetLevelValueF64Codec, FacetLevelValueI64Codec};
use crate::heed_codec::facet::{FieldDocIdFacetF64Codec, FieldDocIdFacetI64Codec};
use crate::mdfs::Mdfs; use crate::mdfs::Mdfs;
use crate::query_tokens::{QueryTokens, QueryToken}; use crate::query_tokens::{QueryTokens, QueryToken};
use crate::{Index, FieldId, DocumentId, Criterion}; use crate::{Index, FieldId, DocumentId, Criterion};
@ -162,6 +164,22 @@ impl<'a> Search<'a> {
let mut output = Vec::new(); let mut output = Vec::new();
match facet_type { match facet_type {
FacetType::Float => { FacetType::Float => {
if documents_ids.len() <= 1000 {
let db = self.index.field_id_docid_facet_values.remap_key_type::<FieldDocIdFacetF64Codec>();
let mut docids_values = Vec::with_capacity(documents_ids.len() as usize);
for docid in documents_ids {
let left = (field_id, docid, f64::MIN);
let right = (field_id, docid, f64::MAX);
let mut iter = db.range(self.rtxn, &(left..=right))?;
let entry = if ascending { iter.next() } else { iter.last() };
if let Some(((_, _, value), ())) = entry.transpose()? {
docids_values.push((docid, OrderedFloat(value)));
}
}
docids_values.sort_unstable_by_key(|(_, value)| *value);
let iter = docids_values.into_iter().map(|(id, _)| id).take(limit);
if ascending { Ok(iter.collect()) } else { Ok(iter.rev().collect()) }
} else {
let facet_fn = if ascending { let facet_fn = if ascending {
FacetIter::<f64, FacetLevelValueF64Codec>::new FacetIter::<f64, FacetLevelValueF64Codec>::new
} else { } else {
@ -173,8 +191,26 @@ impl<'a> Search<'a> {
output.push(docids); output.push(docids);
if limit_tmp == 0 { break } if limit_tmp == 0 { break }
} }
Ok(output.into_iter().flatten().take(limit).collect())
}
}, },
FacetType::Integer => { FacetType::Integer => {
if documents_ids.len() <= 1000 {
let db = self.index.field_id_docid_facet_values.remap_key_type::<FieldDocIdFacetI64Codec>();
let mut docids_values = Vec::with_capacity(documents_ids.len() as usize);
for docid in documents_ids {
let left = (field_id, docid, i64::MIN);
let right = (field_id, docid, i64::MAX);
let mut iter = db.range(self.rtxn, &(left..=right))?;
let entry = if ascending { iter.next() } else { iter.last() };
if let Some(((_, _, value), ())) = entry.transpose()? {
docids_values.push((docid, value));
}
}
docids_values.sort_unstable_by_key(|(_, value)| *value);
let iter = docids_values.into_iter().map(|(id, _)| id).take(limit);
if ascending { Ok(iter.collect()) } else { Ok(iter.rev().collect()) }
} else {
let facet_fn = if ascending { let facet_fn = if ascending {
FacetIter::<i64, FacetLevelValueI64Codec>::new FacetIter::<i64, FacetLevelValueI64Codec>::new
} else { } else {
@ -186,10 +222,11 @@ impl<'a> Search<'a> {
output.push(docids); output.push(docids);
if limit_tmp == 0 { break } if limit_tmp == 0 { break }
} }
Ok(output.into_iter().flatten().take(limit).collect())
}
}, },
FacetType::String => bail!("criteria facet type must be a number"), FacetType::String => bail!("criteria facet type must be a number"),
} }
Ok(output.into_iter().flatten().take(limit).collect())
} }
pub fn execute(&self) -> anyhow::Result<SearchResult> { pub fn execute(&self) -> anyhow::Result<SearchResult> {