push a first version of the benchmark for the typo

milli/Cargo.toml

@@ -61,5 +61,5 @@ rand = "0.8.3"
 default = []
 
 [[bench]]
-name = "search"
+name = "typo"
 harness = false

milli/benches/README.md

@@ -0,0 +1,8 @@
+Benchmarks
+==========
+
+For our benchmark we are using a small subset of the dataset songs.csv. It was generated with this command:
+```
+xsv sample --seed 42 song.csv -o smol_songs.csv
+```
+The original songs.csv datasets is available [here](https://meili-datasets.s3.fr-par.scw.cloud/songs.csv.gz)

milli/benches/typo.rs

@@ -1,22 +1,27 @@
-use std::time::Duration;
+mod utils;
 
-use heed::EnvOpenOptions;
+use std::time::Duration;
-use milli::Index;
 use criterion::{criterion_group, criterion_main, BenchmarkId};
 
-fn bench_search(c: &mut criterion::Criterion) {
+fn bench_typo(c: &mut criterion::Criterion) {
-    let database = "books-4cpu.mmdb";
+    let index = utils::base_setup(Some(vec!["typo".to_string()]));
+
     let queries = [
-        "minogue kylie",
+        "mongus ",
-        "minogue kylie live",
+        "thelonius monk ",
+        "Disnaylande ",
+        "the white striper ",
+        "indochie ",
+        "indochien ",
+        "klub des loopers ",
+        "fear of the duck ",
+        "michel depech ",
+        "stromal ",
+        "dire straights ",
+        "Arethla Franklin ",
     ];
 
-    let mut options = EnvOpenOptions::new();
+    let mut group = c.benchmark_group("typo");
-    options.map_size(100 * 1024 * 1024 * 1024); // 100 GB
-    options.max_readers(10);
-    let index = Index::new(options, database).unwrap();
-
-    let mut group = c.benchmark_group("search");
     group.sample_size(10);
     group.measurement_time(Duration::from_secs(12));
 
@@ -32,5 +37,5 @@ fn bench_search(c: &mut criterion::Criterion) {
     group.finish();
 }
 
-criterion_group!(benches, bench_search);
+criterion_group!(benches, bench_typo);
 criterion_main!(benches);

milli/benches/utils.rs

@@ -0,0 +1,41 @@
+use std::{fs::{File, create_dir_all}};
+
+use heed::EnvOpenOptions;
+use milli::{Index, update::{IndexDocumentsMethod, UpdateBuilder, UpdateFormat}};
+
+pub fn base_setup(criteria: Option<Vec<String>>) -> Index {
+    let database = "songs.mmdb";
+    create_dir_all(&database).unwrap();
+
+    let mut options = EnvOpenOptions::new();
+    options.map_size(100 * 1024 * 1024 * 1024); // 100 GB
+    options.max_readers(10);
+    let index = Index::new(options, database).unwrap();
+
+    let update_builder = UpdateBuilder::new(0);
+    let mut wtxn = index.write_txn().unwrap();
+    let mut builder = update_builder.settings(&mut wtxn, &index);
+
+    if let Some(criteria) = criteria {
+        builder.reset_faceted_fields();
+        builder.reset_criteria();
+        builder.reset_stop_words();
+
+        builder.set_criteria(criteria);
+    }
+
+    builder.execute(|_, _| ()).unwrap();
+    wtxn.commit().unwrap();
+
+    let update_builder = UpdateBuilder::new(0);
+    let mut wtxn = index.write_txn().unwrap();
+    let mut builder = update_builder.index_documents(&mut wtxn, &index);
+    builder.update_format(UpdateFormat::Csv);
+    builder.index_documents_method(IndexDocumentsMethod::ReplaceDocuments);
+    // we called from cargo the current directory is supposed to be milli/milli
+    let reader = File::open("benches/smol_songs.csv").unwrap();
+    builder.execute(reader, |_, _| ()).unwrap();
+    wtxn.commit().unwrap();
+
+    index
+}