mirror of
https://github.com/meilisearch/MeiliSearch
synced 2025-07-03 11:57:07 +02:00
document batch support
reusable transform rework update api add indexer config fix tests review changes Co-authored-by: Clément Renault <clement@meilisearch.com> fmt
This commit is contained in:
parent
74962b2fd9
commit
0c84a40298
18 changed files with 912 additions and 803 deletions
|
@ -6,7 +6,7 @@ use std::path::Path;
|
|||
|
||||
use criterion::{criterion_group, criterion_main, Criterion};
|
||||
use heed::EnvOpenOptions;
|
||||
use milli::update::UpdateBuilder;
|
||||
use milli::update::{IndexDocuments, IndexDocumentsConfig, IndexerConfig, Settings};
|
||||
use milli::Index;
|
||||
|
||||
#[cfg(target_os = "linux")]
|
||||
|
@ -39,9 +39,9 @@ fn indexing_songs_default(c: &mut Criterion) {
|
|||
move || {
|
||||
let index = setup_index();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.settings(&mut wtxn, &index);
|
||||
let mut builder = Settings::new(&mut wtxn, &index, &config);
|
||||
|
||||
builder.set_primary_key("id".to_owned());
|
||||
let displayed_fields =
|
||||
|
@ -66,12 +66,15 @@ fn indexing_songs_default(c: &mut Criterion) {
|
|||
index
|
||||
},
|
||||
move |index| {
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let indexing_config = IndexDocumentsConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let builder = update_builder.index_documents(&mut wtxn, &index);
|
||||
let mut builder =
|
||||
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ());
|
||||
|
||||
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
|
||||
builder.execute(documents, |_| ()).unwrap();
|
||||
builder.add_documents(documents).unwrap();
|
||||
builder.execute().unwrap();
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
index.prepare_for_closing().wait();
|
||||
|
@ -88,9 +91,9 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
|
|||
move || {
|
||||
let index = setup_index();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.settings(&mut wtxn, &index);
|
||||
let mut builder = Settings::new(&mut wtxn, &index, &config);
|
||||
|
||||
builder.set_primary_key("id".to_owned());
|
||||
let displayed_fields =
|
||||
|
@ -112,12 +115,16 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
|
|||
index
|
||||
},
|
||||
move |index| {
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let indexing_config = IndexDocumentsConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let builder = update_builder.index_documents(&mut wtxn, &index);
|
||||
let mut builder =
|
||||
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ());
|
||||
|
||||
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
|
||||
builder.execute(documents, |_| ()).unwrap();
|
||||
|
||||
builder.add_documents(documents).unwrap();
|
||||
builder.execute().unwrap();
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
index.prepare_for_closing().wait();
|
||||
|
@ -134,9 +141,9 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
|
|||
move || {
|
||||
let index = setup_index();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.settings(&mut wtxn, &index);
|
||||
let mut builder = Settings::new(&mut wtxn, &index, &config);
|
||||
|
||||
builder.set_primary_key("id".to_owned());
|
||||
let displayed_fields =
|
||||
|
@ -154,12 +161,15 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
|
|||
index
|
||||
},
|
||||
move |index| {
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let indexing_config = IndexDocumentsConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let builder = update_builder.index_documents(&mut wtxn, &index);
|
||||
let mut builder =
|
||||
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ());
|
||||
|
||||
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
|
||||
builder.execute(documents, |_| ()).unwrap();
|
||||
builder.add_documents(documents).unwrap();
|
||||
builder.execute().unwrap();
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
index.prepare_for_closing().wait();
|
||||
|
@ -176,9 +186,9 @@ fn indexing_wiki(c: &mut Criterion) {
|
|||
move || {
|
||||
let index = setup_index();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.settings(&mut wtxn, &index);
|
||||
let mut builder = Settings::new(&mut wtxn, &index, &config);
|
||||
|
||||
builder.set_primary_key("id".to_owned());
|
||||
let displayed_fields =
|
||||
|
@ -195,13 +205,16 @@ fn indexing_wiki(c: &mut Criterion) {
|
|||
index
|
||||
},
|
||||
move |index| {
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let indexing_config =
|
||||
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.index_documents(&mut wtxn, &index);
|
||||
builder.enable_autogenerate_docids();
|
||||
let mut builder =
|
||||
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ());
|
||||
|
||||
let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
|
||||
builder.execute(documents, |_| ()).unwrap();
|
||||
builder.add_documents(documents).unwrap();
|
||||
builder.execute().unwrap();
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
index.prepare_for_closing().wait();
|
||||
|
@ -218,9 +231,9 @@ fn indexing_movies_default(c: &mut Criterion) {
|
|||
move || {
|
||||
let index = setup_index();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.settings(&mut wtxn, &index);
|
||||
let mut builder = Settings::new(&mut wtxn, &index, &config);
|
||||
|
||||
builder.set_primary_key("id".to_owned());
|
||||
let displayed_fields = ["title", "poster", "overview", "release_date", "genres"]
|
||||
|
@ -242,12 +255,15 @@ fn indexing_movies_default(c: &mut Criterion) {
|
|||
index
|
||||
},
|
||||
move |index| {
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let indexing_config = IndexDocumentsConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let builder = update_builder.index_documents(&mut wtxn, &index);
|
||||
let mut builder =
|
||||
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ());
|
||||
|
||||
let documents = utils::documents_from(datasets_paths::MOVIES, "json");
|
||||
builder.execute(documents, |_| ()).unwrap();
|
||||
builder.add_documents(documents).unwrap();
|
||||
builder.execute().unwrap();
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
index.prepare_for_closing().wait();
|
||||
|
@ -264,9 +280,9 @@ fn indexing_geo(c: &mut Criterion) {
|
|||
move || {
|
||||
let index = setup_index();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.settings(&mut wtxn, &index);
|
||||
let mut builder = Settings::new(&mut wtxn, &index, &config);
|
||||
|
||||
builder.set_primary_key("geonameid".to_owned());
|
||||
let displayed_fields =
|
||||
|
@ -293,12 +309,15 @@ fn indexing_geo(c: &mut Criterion) {
|
|||
index
|
||||
},
|
||||
move |index| {
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let indexing_config = IndexDocumentsConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let builder = update_builder.index_documents(&mut wtxn, &index);
|
||||
let mut builder =
|
||||
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ());
|
||||
|
||||
let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
|
||||
builder.execute(documents, |_| ()).unwrap();
|
||||
builder.add_documents(documents).unwrap();
|
||||
builder.execute().unwrap();
|
||||
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
|
|
|
@ -8,7 +8,9 @@ use std::path::Path;
|
|||
use criterion::BenchmarkId;
|
||||
use heed::EnvOpenOptions;
|
||||
use milli::documents::DocumentBatchReader;
|
||||
use milli::update::{IndexDocumentsMethod, Settings, UpdateBuilder};
|
||||
use milli::update::{
|
||||
IndexDocuments, IndexDocumentsConfig, IndexDocumentsMethod, IndexerConfig, Settings,
|
||||
};
|
||||
use milli::{Filter, Index};
|
||||
use serde_json::{Map, Value};
|
||||
|
||||
|
@ -65,9 +67,9 @@ pub fn base_setup(conf: &Conf) -> Index {
|
|||
options.max_readers(10);
|
||||
let index = Index::new(options, conf.database_name).unwrap();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.settings(&mut wtxn, &index);
|
||||
let mut builder = Settings::new(&mut wtxn, &index, &config);
|
||||
|
||||
if let Some(primary_key) = conf.primary_key {
|
||||
builder.set_primary_key(primary_key.to_string());
|
||||
|
@ -87,16 +89,19 @@ pub fn base_setup(conf: &Conf) -> Index {
|
|||
builder.execute(|_| ()).unwrap();
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
let update_builder = UpdateBuilder::new();
|
||||
let config = IndexerConfig::default();
|
||||
let mut wtxn = index.write_txn().unwrap();
|
||||
let mut builder = update_builder.index_documents(&mut wtxn, &index);
|
||||
if let None = conf.primary_key {
|
||||
builder.enable_autogenerate_docids();
|
||||
}
|
||||
let indexing_config = IndexDocumentsConfig {
|
||||
autogenerate_docids: conf.primary_key.is_none(),
|
||||
update_method: IndexDocumentsMethod::ReplaceDocuments,
|
||||
..Default::default()
|
||||
};
|
||||
let mut builder = IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ());
|
||||
let documents = documents_from(conf.dataset, conf.dataset_format);
|
||||
|
||||
builder.index_documents_method(IndexDocumentsMethod::ReplaceDocuments);
|
||||
builder.execute(documents, |_| ()).unwrap();
|
||||
builder.add_documents(documents).unwrap();
|
||||
|
||||
builder.execute().unwrap();
|
||||
wtxn.commit().unwrap();
|
||||
|
||||
index
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue