searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-11-23 05:14:27 +01:00
Code Issues Releases Wiki Activity

Take the words-fst into account when retrieving the biggests values

Browse Source
This commit is contained in:
Kerollmops 2020-08-26 14:36:22 +02:00
parent 32da07ccee
commit ba2eb0d7ad
No known key found for this signature in database
GPG Key ID: 92ADA4E935E71FA4
1 changed files with 7 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
src/bin/infos.rs
View File

@ -185,6 +185,7 @@ fn biggest_value_sizes(index: &Index, rtxn: &heed::RoTxn, limit: usize) -> anyho
use std::convert::TryInto; use std::convert::TryInto;
use heed::types::{Str, ByteSlice}; use heed::types::{Str, ByteSlice};
let main_name = "main";
let word_positions_name = "word_positions"; let word_positions_name = "word_positions";
let word_position_docids_name = "word_position_docids"; let word_position_docids_name = "word_position_docids";
let word_attribute_docids_name = "word_attribute_docids"; let word_attribute_docids_name = "word_attribute_docids";
@ -192,6 +193,11 @@ fn biggest_value_sizes(index: &Index, rtxn: &heed::RoTxn, limit: usize) -> anyho
let mut heap = BinaryHeap::with_capacity(limit + 1); let mut heap = BinaryHeap::with_capacity(limit + 1);
if limit > 0 { if limit > 0 {
if let Some(fst) = index.fst(rtxn)? {
heap.push(Reverse((fst.as_fst().as_bytes().len(), format!("words-fst"), main_name)));
if heap.len() > limit { heap.pop(); }
}
for result in index.word_positions.as_polymorph().iter::<_, Str, ByteSlice>(rtxn)? { for result in index.word_positions.as_polymorph().iter::<_, Str, ByteSlice>(rtxn)? {
let (word, value) = result?; let (word, value) = result?;
heap.push(Reverse((value.len(), word.to_string(), word_positions_name))); heap.push(Reverse((value.len(), word.to_string(), word_positions_name)));
@ -223,7 +229,7 @@ fn biggest_value_sizes(index: &Index, rtxn: &heed::RoTxn, limit: usize) -> anyho
let stdout = io::stdout(); let stdout = io::stdout();
let mut wtr = csv::Writer::from_writer(stdout.lock()); let mut wtr = csv::Writer::from_writer(stdout.lock());
wtr.write_record(&["size", "key_name", "database_name"])?; wtr.write_record(&["database_name", "key_name", "size"])?;
for Reverse((size, key_name, database_name)) in heap.into_sorted_vec() { for Reverse((size, key_name, database_name)) in heap.into_sorted_vec() {
wtr.write_record(&[database_name.to_string(), key_name, size.to_string()])?; wtr.write_record(&[database_name.to_string(), key_name, size.to_string()])?;