searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-11-30 00:34:26 +01:00
Code Issues Releases Wiki Activity

Merge remote-tracking branch 'origin/main' into facet-levels-refactor

Browse Source
This commit is contained in:
Loïc Lecrenier 2022-10-26 14:03:23 +02:00
commit 2741756248
14 changed files with 414 additions and 136 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

302
benchmarks/benches/indexing.rs
View File

@ -59,7 +59,7 @@ fn setup_settings<'t>(
let sortable_fields = sortable_fields.iter().map(|s| s.to_string()).collect();
builder.set_sortable_fields(sortable_fields);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
}
fn setup_index_with_settings<'t>(
@ -131,8 +131,14 @@ fn indexing_songs_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
@ -169,8 +175,14 @@ fn reindexing_songs_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
@ -185,8 +197,14 @@ fn reindexing_songs_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
@ -225,8 +243,14 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let mut wtxn = index.write_txn().unwrap();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -282,8 +306,14 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let mut wtxn = index.write_txn().unwrap();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS_1_2, "csv");
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -298,8 +328,14 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS_3_4, "csv");
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -307,8 +343,14 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
builder.execute().unwrap();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS_4_4, "csv");
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -345,8 +387,14 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
@ -384,8 +432,14 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
@ -423,8 +477,14 @@ fn indexing_wiki(c: &mut Criterion) {
let indexing_config =
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
@ -461,8 +521,14 @@ fn reindexing_wiki(c: &mut Criterion) {
let indexing_config =
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
@ -478,8 +544,14 @@ fn reindexing_wiki(c: &mut Criterion) {
let indexing_config =
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
@ -518,8 +590,14 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
let mut wtxn = index.write_txn().unwrap();
let indexing_config =
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -576,8 +654,14 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config =
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents =
utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES_1_2, "csv");
@ -594,8 +678,14 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
let indexing_config =
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents =
@ -606,8 +696,14 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
let indexing_config =
IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents =
@ -646,8 +742,14 @@ fn indexing_movies_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::MOVIES, "json");
@ -683,8 +785,14 @@ fn reindexing_movies_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::MOVIES, "json");
@ -699,8 +807,14 @@ fn reindexing_movies_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::MOVIES, "json");
@ -738,8 +852,14 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let mut wtxn = index.write_txn().unwrap();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::MOVIES, "json");
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -794,8 +914,14 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
// as we don't care about the time it takes.
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::MOVIES_1_2, "json");
@ -811,8 +937,14 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::MOVIES_3_4, "json");
@ -821,8 +953,14 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
builder.execute().unwrap();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::MOVIES_4_4, "json");
@ -883,8 +1021,14 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::NESTED_MOVIES, "json");
@ -945,8 +1089,14 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let mut wtxn = index.write_txn().unwrap();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::NESTED_MOVIES, "json");
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -1008,8 +1158,14 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::NESTED_MOVIES, "json");
@ -1046,8 +1202,14 @@ fn indexing_geo(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
@ -1084,8 +1246,14 @@ fn reindexing_geo(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
@ -1101,8 +1269,14 @@ fn reindexing_geo(c: &mut Criterion) {
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let mut wtxn = index.write_txn().unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
@ -1141,8 +1315,14 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
let config = IndexerConfig::default();
let mut wtxn = index.write_txn().unwrap();
let indexing_config = IndexDocumentsConfig::default();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ())
let builder = IndexDocuments::new(
&mut wtxn,
&index,
&config,
indexing_config,
|_| (),
|| false,
)
.unwrap();
let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
let (builder, user_error) = builder.add_documents(documents).unwrap();

5
benchmarks/benches/utils.rs
View File

@ -86,7 +86,7 @@ pub fn base_setup(conf: &Conf) -> Index {
(conf.configure)(&mut builder);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
wtxn.commit().unwrap();
let config = IndexerConfig::default();
@ -96,7 +96,8 @@ pub fn base_setup(conf: &Conf) -> Index {
update_method: IndexDocumentsMethod::ReplaceDocuments,
..Default::default()
};
let builder = IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ()).unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| (), || false).unwrap();
let documents = documents_from(conf.dataset, conf.dataset_format);
let (builder, user_error) = builder.add_documents(documents).unwrap();
user_error.unwrap();

5
cli/src/main.rs
View File

@ -239,7 +239,7 @@ impl Performer for DocumentAddition {
if let Some(primary) = self.primary {
let mut builder = update::Settings::new(&mut txn, &index, &config);
builder.set_primary_key(primary);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
}
let indexing_config = IndexDocumentsConfig {
@ -260,6 +260,7 @@ impl Performer for DocumentAddition {
&config,
indexing_config,
|step| indexing_callback(step, &bars),
|| false,
)
.unwrap();
let (addition, user_error) = addition.add_documents(reader)?;
@ -517,7 +518,7 @@ impl Performer for SettingsUpdate {
bars.push(bar);
}
update.execute(|step| indexing_callback(step, &bars))?;
update.execute(|step| indexing_callback(step, &bars), || false)?;
txn.commit()?;
Ok(())

2
milli/src/error.rs
View File

@ -56,6 +56,8 @@ pub enum InternalError {
Store(#[from] MdbError),
#[error(transparent)]
Utf8(#[from] str::Utf8Error),
#[error("An indexation process was explicitly aborted.")]
AbortedIndexation,
}
#[derive(Error, Debug)]

38
milli/src/index.rs
View File

@ -1188,6 +1188,7 @@ pub(crate) mod tests {
use tempfile::TempDir;
use crate::documents::DocumentsBatchReader;
use crate::error::{Error, InternalError};
use crate::index::{DEFAULT_MIN_WORD_LEN_ONE_TYPO, DEFAULT_MIN_WORD_LEN_TWO_TYPOS};
use crate::update::{self, IndexDocuments, IndexDocumentsConfig, IndexerConfig, Settings};
use crate::{db_snap, Index};
@ -1237,6 +1238,7 @@ pub(crate) mod tests {
&self.indexer_config,
self.index_documents_config.clone(),
|_| (),
|| false,
)
.unwrap();
let (builder, user_error) = builder.add_documents(documents).unwrap();
@ -1273,11 +1275,45 @@ pub(crate) mod tests {
) -> Result<(), crate::error::Error> {
let mut builder = update::Settings::new(wtxn, &self.inner, &self.indexer_config);
update(&mut builder);
builder.execute(drop)?;
builder.execute(drop, || false)?;
Ok(())
}
}
#[test]
fn aborting_indexation() {
use std::sync::atomic::AtomicBool;
use std::sync::atomic::Ordering::Relaxed;
let index = TempIndex::new();
let mut wtxn = index.inner.write_txn().unwrap();
let should_abort = AtomicBool::new(false);
let builder = IndexDocuments::new(
&mut wtxn,
&index.inner,
&index.indexer_config,
index.index_documents_config.clone(),
|_| (),
|| should_abort.load(Relaxed),
)
.unwrap();
let (builder, user_error) = builder
.add_documents(documents!([
{ "id": 1, "name": "kevin" },
{ "id": 2, "name": "bob", "age": 20 },
{ "id": 2, "name": "bob", "age": 20 },
]))
.unwrap();
user_error.unwrap();
should_abort.store(true, Relaxed);
let err = builder.execute().unwrap_err();
assert!(matches!(err, Error::InternalError(InternalError::AbortedIndexation)));
}
#[test]
fn initial_field_distribution() {
let index = TempIndex::new();

5
milli/src/search/distinct/mod.rs
View File

@ -89,7 +89,7 @@ mod test {
let config = IndexerConfig::default();
let mut update = Settings::new(&mut txn, &index, &config);
update.set_distinct_field(distinct.to_string());
update.execute(|_| ()).unwrap();
update.execute(|_| (), || false).unwrap();
// add documents to the index
let config = IndexerConfig::default();
@ -98,7 +98,8 @@ mod test {
..Default::default()
};
let addition =
IndexDocuments::new(&mut txn, &index, &config, indexing_config, |_| ()).unwrap();
IndexDocuments::new(&mut txn, &index, &config, indexing_config, |_| (), || false)
.unwrap();
let reader =
crate::documents::DocumentsBatchReader::from_reader(Cursor::new(JSON.as_slice()))

69
milli/src/update/index_documents/mod.rs
View File

@ -32,7 +32,7 @@ pub use self::helpers::{
use self::helpers::{grenad_obkv_into_chunks, GrenadParameters};
pub use self::transform::{Transform, TransformOutput};
use crate::documents::{obkv_to_object, DocumentsBatchReader};
use crate::error::UserError;
use crate::error::{Error, InternalError, UserError};
pub use crate::update::index_documents::helpers::CursorClonableMmap;
use crate::update::{
self, IndexerConfig, PrefixWordPairsProximityDocids, UpdateIndexingStep, WordPrefixDocids,
@ -70,13 +70,14 @@ impl Default for IndexDocumentsMethod {
}
}
pub struct IndexDocuments<'t, 'u, 'i, 'a, F> {
pub struct IndexDocuments<'t, 'u, 'i, 'a, FP, FA> {
wtxn: &'t mut heed::RwTxn<'i, 'u>,
index: &'i Index,
config: IndexDocumentsConfig,
indexer_config: &'a IndexerConfig,
transform: Option<Transform<'a, 'i>>,
progress: F,
progress: FP,
should_abort: FA,
added_documents: u64,
}
@ -90,17 +91,19 @@ pub struct IndexDocumentsConfig {
pub autogenerate_docids: bool,
}
impl<'t, 'u, 'i, 'a, F> IndexDocuments<'t, 'u, 'i, 'a, F>
impl<'t, 'u, 'i, 'a, FP, FA> IndexDocuments<'t, 'u, 'i, 'a, FP, FA>
where
F: Fn(UpdateIndexingStep) + Sync,
FP: Fn(UpdateIndexingStep) + Sync,
FA: Fn() -> bool + Sync,
{
pub fn new(
wtxn: &'t mut heed::RwTxn<'i, 'u>,
index: &'i Index,
indexer_config: &'a IndexerConfig,
config: IndexDocumentsConfig,
progress: F,
) -> Result<IndexDocuments<'t, 'u, 'i, 'a, F>> {
progress: FP,
should_abort: FA,
) -> Result<IndexDocuments<'t, 'u, 'i, 'a, FP, FA>> {
let transform = Some(Transform::new(
wtxn,
index,
@ -114,6 +117,7 @@ where
config,
indexer_config,
progress,
should_abort,
wtxn,
index,
added_documents: 0,
@ -148,12 +152,13 @@ where
Err(user_error) => return Ok((self, Err(user_error))),
};
let indexed_documents = self
.transform
.as_mut()
.expect("Invalid document addition state")
.read_documents(enriched_documents_reader, self.wtxn, &self.progress)?
as u64;
let indexed_documents =
self.transform.as_mut().expect("Invalid document addition state").read_documents(
enriched_documents_reader,
self.wtxn,
&self.progress,
&self.should_abort,
)? as u64;
self.added_documents += indexed_documents;
@ -197,7 +202,8 @@ where
#[logging_timer::time("IndexDocuments::{}")]
pub fn execute_raw(self, output: TransformOutput) -> Result<u64>
where
F: Fn(UpdateIndexingStep) + Sync,
FP: Fn(UpdateIndexingStep) + Sync,
FA: Fn() -> bool + Sync,
{
let TransformOutput {
primary_key,
@ -346,6 +352,10 @@ where
});
for result in lmdb_writer_rx {
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
let typed_chunk = match result? {
TypedChunk::WordDocids { word_docids_reader, exact_word_docids_reader } => {
let cloneable_chunk = unsafe { as_cloneable_grenad(&word_docids_reader)? };
@ -422,17 +432,26 @@ where
word_position_docids: Option<grenad::Reader<CursorClonableMmap>>,
) -> Result<()>
where
F: Fn(UpdateIndexingStep) + Sync,
FP: Fn(UpdateIndexingStep) + Sync,
FA: Fn() -> bool + Sync,
{
// Merged databases are already been indexed, we start from this count;
let mut databases_seen = MERGED_DATABASE_COUNT;
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
databases_seen += 1;
(self.progress)(UpdateIndexingStep::MergeDataIntoFinalDatabase {
databases_seen,
total_databases: TOTAL_POSTING_DATABASE_COUNT,
});
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
let previous_words_prefixes_fst =
self.index.words_prefixes_fst(self.wtxn)?.map_data(|cow| cow.into_owned())?;
@ -446,6 +465,10 @@ where
}
builder.execute()?;
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
let current_prefix_fst = self.index.words_prefixes_fst(self.wtxn)?;
// We retrieve the common words between the previous and new prefix word fst.
@ -473,6 +496,10 @@ where
total_databases: TOTAL_POSTING_DATABASE_COUNT,
});
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
if let Some(word_docids) = word_docids {
execute_word_prefix_docids(
self.wtxn,
@ -499,6 +526,10 @@ where
)?;
}
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
databases_seen += 1;
(self.progress)(UpdateIndexingStep::MergeDataIntoFinalDatabase {
databases_seen,
@ -521,6 +552,10 @@ where
)?;
}
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
databases_seen += 1;
(self.progress)(UpdateIndexingStep::MergeDataIntoFinalDatabase {
databases_seen,
@ -548,6 +583,10 @@ where
)?;
}
if (self.should_abort)() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
databases_seen += 1;
(self.progress)(UpdateIndexingStep::MergeDataIntoFinalDatabase {
databases_seen,

12
milli/src/update/index_documents/transform.rs
View File

@ -138,15 +138,17 @@ impl<'a, 'i> Transform<'a, 'i> {
})
}
pub fn read_documents<R, F>(
pub fn read_documents<R, FP, FA>(
&mut self,
reader: EnrichedDocumentsBatchReader<R>,
wtxn: &mut heed::RwTxn,
progress_callback: F,
progress_callback: FP,
should_abort: FA,
) -> Result<usize>
where
R: Read + Seek,
F: Fn(UpdateIndexingStep) + Sync,
FP: Fn(UpdateIndexingStep) + Sync,
FA: Fn() -> bool + Sync,
{
let (mut cursor, fields_index) = reader.into_cursor_and_fields_index();
@ -165,6 +167,10 @@ impl<'a, 'i> Transform<'a, 'i> {
while let Some(enriched_document) = cursor.next_enriched_document()? {
let EnrichedDocument { document, document_id } = enriched_document;
if should_abort() {
return Err(Error::InternalError(InternalError::AbortedIndexation));
}
// drop_and_reuse is called instead of .clear() to communicate to the compiler that field_buffer
// does not keep references from the cursor between loop iterations
let mut field_buffer_cache = drop_and_reuse(field_buffer);

20
milli/src/update/settings.rs
View File

@ -266,9 +266,15 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
self.pagination_max_total_hits = Setting::Reset;
}
fn reindex<F>(&mut self, cb: &F, old_fields_ids_map: FieldsIdsMap) -> Result<()>
fn reindex<FP, FA>(
&mut self,
progress_callback: &FP,
should_abort: &FA,
old_fields_ids_map: FieldsIdsMap,
) -> Result<()>
where
F: Fn(UpdateIndexingStep) + Sync,
FP: Fn(UpdateIndexingStep) + Sync,
FA: Fn() -> bool + Sync,
{
let fields_ids_map = self.index.fields_ids_map(self.wtxn)?;
// if the settings are set before any document update, we don't need to do anything, and
@ -302,7 +308,8 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
self.index,
self.indexer_config,
IndexDocumentsConfig::default(),
&cb,
&progress_callback,
&should_abort,
)?;
indexing_builder.execute_raw(output)?;
@ -657,9 +664,10 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
Ok(())
}
pub fn execute<F>(mut self, progress_callback: F) -> Result<()>
pub fn execute<FP, FA>(mut self, progress_callback: FP, should_abort: FA) -> Result<()>
where
F: Fn(UpdateIndexingStep) + Sync,
FP: Fn(UpdateIndexingStep) + Sync,
FA: Fn() -> bool + Sync,
{
self.index.set_updated_at(self.wtxn, &OffsetDateTime::now_utc())?;
@ -695,7 +703,7 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
|| searchable_updated
|| exact_attributes_updated
{
self.reindex(&progress_callback, old_fields_ids_map)?;
self.reindex(&progress_callback, &should_abort, old_fields_ids_map)?;
}
Ok(())

2
milli/tests/search/distinct.rs
View File

@ -19,7 +19,7 @@ macro_rules! test_distinct {
let config = milli::update::IndexerConfig::default();
let mut builder = Settings::new(&mut wtxn, &index, &config);
builder.set_distinct_field(S(stringify!($distinct)));
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
wtxn.commit().unwrap();
let rtxn = index.read_txn().unwrap();

5
milli/tests/search/facet_distribution.rs
View File

@ -23,13 +23,14 @@ fn test_facet_distribution_with_no_facet_values() {
S("genres"),
S("tags"),
});
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
// index documents
let config = IndexerConfig { max_memory: Some(10 * 1024 * 1024), ..Default::default() };
let indexing_config = IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let builder = IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ()).unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| (), || false).unwrap();
let mut documents_builder = DocumentsBatchBuilder::new(Vec::new());
let reader = Cursor::new(
r#"{

5
milli/tests/search/mod.rs
View File

@ -57,13 +57,14 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
S("america") => vec![S("the united states")],
});
builder.set_searchable_fields(vec![S("title"), S("description")]);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
// index documents
let config = IndexerConfig { max_memory: Some(10 * 1024 * 1024), ..Default::default() };
let indexing_config = IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let builder = IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ()).unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| (), || false).unwrap();
let mut documents_builder = DocumentsBatchBuilder::new(Vec::new());
let reader = Cursor::new(CONTENT.as_bytes());

9
milli/tests/search/query_criteria.rs
View File

@ -345,7 +345,7 @@ fn criteria_mixup() {
let mut wtxn = index.write_txn().unwrap();
let mut builder = Settings::new(&mut wtxn, &index, &config);
builder.set_criteria(criteria.iter().map(ToString::to_string).collect());
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
wtxn.commit().unwrap();
let mut rtxn = index.read_txn().unwrap();
@ -385,12 +385,13 @@ fn criteria_ascdesc() {
S("name"),
S("age"),
});
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
// index documents
let config = IndexerConfig { max_memory: Some(10 * 1024 * 1024), ..Default::default() };
let indexing_config = IndexDocumentsConfig { autogenerate_docids: true, ..Default::default() };
let builder = IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| ()).unwrap();
let builder =
IndexDocuments::new(&mut wtxn, &index, &config, indexing_config, |_| (), || false).unwrap();
let mut batch_builder = DocumentsBatchBuilder::new(Vec::new());
@ -436,7 +437,7 @@ fn criteria_ascdesc() {
let mut wtxn = index.write_txn().unwrap();
let mut builder = Settings::new(&mut wtxn, &index, &config);
builder.set_criteria(vec![criterion.to_string()]);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
wtxn.commit().unwrap();
let mut rtxn = index.read_txn().unwrap();

11
milli/tests/search/typo_tolerance.rs
View File

@ -40,7 +40,7 @@ fn test_typo_tolerance_one_typo() {
let config = IndexerConfig::default();
let mut builder = Settings::new(&mut txn, &index, &config);
builder.set_min_word_len_one_typo(4);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
// typo is now supported for 4 letters words
let mut search = Search::new(&txn, &index);
@ -86,7 +86,7 @@ fn test_typo_tolerance_two_typo() {
let config = IndexerConfig::default();
let mut builder = Settings::new(&mut txn, &index, &config);
builder.set_min_word_len_two_typos(7);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
// typo is now supported for 4 letters words
let mut search = Search::new(&txn, &index);
@ -127,7 +127,8 @@ fn test_typo_disabled_on_word() {
let mut txn = index.write_txn().unwrap();
let config = IndexerConfig::default();
let indexing_config = IndexDocumentsConfig::default();
let builder = IndexDocuments::new(&mut txn, &index, &config, indexing_config, |_| ()).unwrap();
let builder =
IndexDocuments::new(&mut txn, &index, &config, indexing_config, |_| (), || false).unwrap();
let (builder, user_error) = builder.add_documents(documents).unwrap();
user_error.unwrap();
@ -156,7 +157,7 @@ fn test_typo_disabled_on_word() {
// `zealand` doesn't allow typos anymore
exact_words.insert("zealand".to_string());
builder.set_exact_words(exact_words);
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
let mut search = Search::new(&txn, &index);
search.query("zealand");
@ -194,7 +195,7 @@ fn test_disable_typo_on_attribute() {
let mut builder = Settings::new(&mut txn, &index, &config);
// disable typos on `description`
builder.set_exact_attributes(vec!["description".to_string()].into_iter().collect());
builder.execute(|_| ()).unwrap();
builder.execute(|_| (), || false).unwrap();
let mut search = Search::new(&txn, &index);
search.query("antebelum");