snowykami/meilisearch
1
0
Fork 0
mirror of https://github.com/meilisearch/meilisearch.git synced 2025-10-30 23:46:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix tests

Browse Source 
# Conflicts:
#	crates/index-scheduler/src/batch.rs
#	crates/index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/after_removing_the_documents.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_bad_primary_key/fifth_task_succeeds.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_bad_primary_key/fourth_task_fails.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_multiple_primary_key/second_task_fails.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_multiple_primary_key/third_task_fails.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_multiple_primary_key_batch_wrong_key/second_and_third_tasks_fails.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_set_and_null_primary_key_inference_works/all_other_tasks_succeeds.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_set_and_null_primary_key_inference_works/second_task_fails.snap
#	crates/index-scheduler/src/snapshots/lib.rs/test_document_addition_with_set_and_null_primary_key_inference_works/third_task_succeeds.snap

# Conflicts:
#	crates/index-scheduler/src/batch.rs
#	crates/meilisearch/src/search/mod.rs
#	crates/meilisearch/tests/vector/mod.rs

# Conflicts:
#	crates/index-scheduler/src/batch.rs
This commit is contained in:
airycanon
2024-11-22 14:19:20 +08:00
parent 95ed079761
commit b75f1f4c17
24 changed files with 378 additions and 281 deletions
Download Patch File Download Diff File Expand all files Collapse all files

210
crates/index-scheduler/src/batch.rs
View File

@@ -29,7 +29,6 @@ use bumpalo::collections::CollectIn;
use bumpalo::Bump; use bumpalo::Bump;
use dump::IndexMetadata; use dump::IndexMetadata;
use meilisearch_types::batches::BatchId; use meilisearch_types::batches::BatchId;
use meilisearch_types::error::Code;
use meilisearch_types::heed::{RoTxn, RwTxn}; use meilisearch_types::heed::{RoTxn, RwTxn};
use meilisearch_types::milli::documents::{obkv_to_object, DocumentsBatchReader, PrimaryKey}; use meilisearch_types::milli::documents::{obkv_to_object, DocumentsBatchReader, PrimaryKey};
use meilisearch_types::milli::heed::CompactionOption; use meilisearch_types::milli::heed::CompactionOption;
@@ -689,7 +688,9 @@ impl IndexScheduler {
let index = self.index_mapper.index(&rtxn, name)?; let index = self.index_mapper.index(&rtxn, name)?;
let dst = temp_snapshot_dir.path().join("indexes").join(uuid.to_string()); let dst = temp_snapshot_dir.path().join("indexes").join(uuid.to_string());
fs::create_dir_all(&dst)?; fs::create_dir_all(&dst)?;
index.copy_to_file(dst.join("data.mdb"), CompactionOption::Enabled)?; index
.copy_to_file(dst.join("data.mdb"), CompactionOption::Enabled)
.map_err(|e| Error::from_milli(e, Some(name.to_string())))?;
} }
drop(rtxn); drop(rtxn);
@@ -791,16 +792,19 @@ impl IndexScheduler {
let content_file = self.file_store.get_update(content_file)?; let content_file = self.file_store.get_update(content_file)?;
let reader = DocumentsBatchReader::from_reader(content_file) let reader = DocumentsBatchReader::from_reader(content_file)
.map_err(milli::Error::from)?; .map_err(|e| Error::from_milli(e.into(), None))?;
let (mut cursor, documents_batch_index) = let (mut cursor, documents_batch_index) =
reader.into_cursor_and_fields_index(); reader.into_cursor_and_fields_index();
while let Some(doc) = while let Some(doc) = cursor
cursor.next_document().map_err(milli::Error::from)? .next_document()
.map_err(|e| Error::from_milli(e.into(), None))?
{ {
dump_content_file dump_content_file.push_document(
.push_document(&obkv_to_object(doc, &documents_batch_index)?)?; &obkv_to_object(doc, &documents_batch_index)
.map_err(|e| Error::from_milli(e, None))?,
)?;
} }
dump_content_file.flush()?; dump_content_file.flush()?;
} }
@@ -814,27 +818,41 @@ impl IndexScheduler {
let metadata = IndexMetadata { let metadata = IndexMetadata {
uid: uid.to_owned(), uid: uid.to_owned(),
primary_key: index.primary_key(&rtxn)?.map(String::from), primary_key: index.primary_key(&rtxn)?.map(String::from),
created_at: index.created_at(&rtxn)?, created_at: index
updated_at: index.updated_at(&rtxn)?, .created_at(&rtxn)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?,
updated_at: index
.updated_at(&rtxn)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?,
}; };
let mut index_dumper = dump.create_index(uid, &metadata)?; let mut index_dumper = dump.create_index(uid, &metadata)?;
let fields_ids_map = index.fields_ids_map(&rtxn)?; let fields_ids_map = index.fields_ids_map(&rtxn)?;
let all_fields: Vec<_> = fields_ids_map.iter().map(|(id, _)| id).collect(); let all_fields: Vec<_> = fields_ids_map.iter().map(|(id, _)| id).collect();
let embedding_configs = index.embedding_configs(&rtxn)?; let embedding_configs = index
.embedding_configs(&rtxn)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?;
let documents = index
.all_documents(&rtxn)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?;
// 3.1. Dump the documents // 3.1. Dump the documents
for ret in index.all_documents(&rtxn)? { for ret in documents {
if self.must_stop_processing.get() { if self.must_stop_processing.get() {
return Err(Error::AbortedTask); return Err(Error::AbortedTask);
} }
let (id, doc) = ret?; let (id, doc) =
ret.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?;
let mut document = milli::obkv_to_json(&all_fields, &fields_ids_map, doc)?; let mut document =
milli::obkv_to_json(&all_fields, &fields_ids_map, doc)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?;
'inject_vectors: { 'inject_vectors: {
let embeddings = index.embeddings(&rtxn, id)?; let embeddings = index
.embeddings(&rtxn, id)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?;
if embeddings.is_empty() { if embeddings.is_empty() {
break 'inject_vectors; break 'inject_vectors;
@@ -845,7 +863,7 @@ impl IndexScheduler {
.or_insert(serde_json::Value::Object(Default::default())); .or_insert(serde_json::Value::Object(Default::default()));
let serde_json::Value::Object(vectors) = vectors else { let serde_json::Value::Object(vectors) = vectors else {
return Err(milli::Error::UserError( let user_err = milli::Error::UserError(
milli::UserError::InvalidVectorsMapType { milli::UserError::InvalidVectorsMapType {
document_id: { document_id: {
if let Ok(Some(Ok(index))) = index if let Ok(Some(Ok(index))) = index
@@ -859,8 +877,9 @@ impl IndexScheduler {
}, },
value: vectors.clone(), value: vectors.clone(),
}, },
) );
.into());
return Err(Error::from_milli(user_err, Some(uid.to_string())));
}; };
for (embedder_name, embeddings) in embeddings { for (embedder_name, embeddings) in embeddings {
@@ -890,7 +909,8 @@ impl IndexScheduler {
index, index,
&rtxn, &rtxn,
meilisearch_types::settings::SecretPolicy::RevealSecrets, meilisearch_types::settings::SecretPolicy::RevealSecrets,
)?; )
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?;
index_dumper.settings(&settings)?; index_dumper.settings(&settings)?;
Ok(()) Ok(())
})?; })?;
@@ -946,7 +966,8 @@ impl IndexScheduler {
// the entire batch. // the entire batch.
let res = || -> Result<()> { let res = || -> Result<()> {
let index_rtxn = index.read_txn()?; let index_rtxn = index.read_txn()?;
let stats = crate::index_mapper::IndexStats::new(&index, &index_rtxn)?; let stats = crate::index_mapper::IndexStats::new(&index, &index_rtxn)
.map_err(|e| Error::from_milli(e, Some(index_uid.to_string())))?;
let mut wtxn = self.env.write_txn()?; let mut wtxn = self.env.write_txn()?;
self.index_mapper.store_stats_of(&mut wtxn, &index_uid, &stats)?; self.index_mapper.store_stats_of(&mut wtxn, &index_uid, &stats)?;
wtxn.commit()?; wtxn.commit()?;
@@ -988,10 +1009,12 @@ impl IndexScheduler {
); );
builder.set_primary_key(primary_key); builder.set_primary_key(primary_key);
let must_stop_processing = self.must_stop_processing.clone(); let must_stop_processing = self.must_stop_processing.clone();
builder.execute( builder
|indexing_step| tracing::debug!(update = ?indexing_step), .execute(
|| must_stop_processing.get(), |indexing_step| tracing::debug!(update = ?indexing_step),
)?; || must_stop_processing.get(),
)
.map_err(|e| Error::from_milli(e, Some(index_uid.to_string())))?;
index_wtxn.commit()?; index_wtxn.commit()?;
} }
@@ -1008,7 +1031,8 @@ impl IndexScheduler {
let res = || -> Result<()> { let res = || -> Result<()> {
let mut wtxn = self.env.write_txn()?; let mut wtxn = self.env.write_txn()?;
let index_rtxn = index.read_txn()?; let index_rtxn = index.read_txn()?;
let stats = crate::index_mapper::IndexStats::new(&index, &index_rtxn)?; let stats = crate::index_mapper::IndexStats::new(&index, &index_rtxn)
.map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
self.index_mapper.store_stats_of(&mut wtxn, &index_uid, &stats)?; self.index_mapper.store_stats_of(&mut wtxn, &index_uid, &stats)?;
wtxn.commit()?; wtxn.commit()?;
Ok(()) Ok(())
@@ -1031,7 +1055,9 @@ impl IndexScheduler {
let number_of_documents = || -> Result<u64> { let number_of_documents = || -> Result<u64> {
let index = self.index_mapper.index(&wtxn, &index_uid)?; let index = self.index_mapper.index(&wtxn, &index_uid)?;
let index_rtxn = index.read_txn()?; let index_rtxn = index.read_txn()?;
Ok(index.number_of_documents(&index_rtxn)?) index
.number_of_documents(&index_rtxn)
.map_err(|e| Error::from_milli(e, Some(index_uid.to_string())))
}() }()
.unwrap_or_default(); .unwrap_or_default();
@@ -1188,8 +1214,10 @@ impl IndexScheduler {
}; };
match operation { match operation {
IndexOperation::DocumentClear { mut tasks, .. } => { IndexOperation::DocumentClear { index_uid, mut tasks } => {
let count = milli::update::ClearDocuments::new(index_wtxn, index).execute()?; let count = milli::update::ClearDocuments::new(index_wtxn, index)
.execute()
.map_err(|e| Error::from_milli(e, Some(index_uid)))?;
let mut first_clear_found = false; let mut first_clear_found = false;
for task in &mut tasks { for task in &mut tasks {
@@ -1209,7 +1237,7 @@ impl IndexScheduler {
Ok(tasks) Ok(tasks)
} }
IndexOperation::DocumentOperation { IndexOperation::DocumentOperation {
index_uid: _, index_uid,
primary_key, primary_key,
method, method,
operations, operations,
@@ -1235,13 +1263,17 @@ impl IndexScheduler {
let mut content_files_iter = content_files.iter(); let mut content_files_iter = content_files.iter();
let mut indexer = indexer::DocumentOperation::new(method); let mut indexer = indexer::DocumentOperation::new(method);
let embedders = index.embedding_configs(index_wtxn)?; let embedders = index
let embedders = self.embedders(embedders)?; .embedding_configs(index_wtxn)
.map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
let embedders = self.embedders(index_uid.clone(), embedders)?;
for operation in operations { for operation in operations {
match operation { match operation {
DocumentOperation::Add(_content_uuid) => { DocumentOperation::Add(_content_uuid) => {
let mmap = content_files_iter.next().unwrap(); let mmap = content_files_iter.next().unwrap();
indexer.add_documents(mmap)?; indexer
.add_documents(mmap)
.map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
} }
DocumentOperation::Delete(document_ids) => { DocumentOperation::Delete(document_ids) => {
let document_ids: bumpalo::collections::vec::Vec<_> = document_ids let document_ids: bumpalo::collections::vec::Vec<_> = document_ids
@@ -1266,15 +1298,17 @@ impl IndexScheduler {
} }
}; };
let (document_changes, operation_stats, primary_key) = indexer.into_changes( let (document_changes, operation_stats, primary_key) = indexer
&indexer_alloc, .into_changes(
index, &indexer_alloc,
&rtxn, index,
primary_key.as_deref(), &rtxn,
&mut new_fields_ids_map, primary_key.as_deref(),
&|| must_stop_processing.get(), &mut new_fields_ids_map,
&send_progress, &|| must_stop_processing.get(),
)?; &send_progress,
)
.map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
let mut addition = 0; let mut addition = 0;
for (stats, task) in operation_stats.into_iter().zip(&mut tasks) { for (stats, task) in operation_stats.into_iter().zip(&mut tasks) {
@@ -1321,14 +1355,15 @@ impl IndexScheduler {
embedders, embedders,
&|| must_stop_processing.get(), &|| must_stop_processing.get(),
&send_progress, &send_progress,
)?; )
.map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
tracing::info!(indexing_result = ?addition, processed_in = ?started_processing_at.elapsed(), "document indexing done"); tracing::info!(indexing_result = ?addition, processed_in = ?started_processing_at.elapsed(), "document indexing done");
} }
Ok(tasks) Ok(tasks)
} }
IndexOperation::DocumentEdition { mut task, .. } => { IndexOperation::DocumentEdition { index_uid, mut task } => {
let (filter, code) = if let KindWithContent::DocumentEdition { let (filter, code) = if let KindWithContent::DocumentEdition {
filter_expr, filter_expr,
context: _, context: _,
@@ -1342,16 +1377,11 @@ impl IndexScheduler {
}; };
let candidates = match filter.as_ref().map(Filter::from_json) { let candidates = match filter.as_ref().map(Filter::from_json) {
Some(Ok(Some(filter))) => { Some(Ok(Some(filter))) => filter
filter.evaluate(index_wtxn, index).map_err(|err| match err { .evaluate(index_wtxn, index)
milli::Error::UserError(milli::UserError::InvalidFilter(_)) => { .map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?,
Error::from(err).with_custom_error_code(Code::InvalidDocumentFilter)
}
e => e.into(),
})?
}
None | Some(Ok(None)) => index.documents_ids(index_wtxn)?, None | Some(Ok(None)) => index.documents_ids(index_wtxn)?,
Some(Err(e)) => return Err(e.into()), Some(Err(e)) => return Err(Error::from_milli(e, Some(index_uid.clone()))),
}; };
let (original_filter, context, function) = if let Some(Details::DocumentEdition { let (original_filter, context, function) = if let Some(Details::DocumentEdition {
@@ -1386,8 +1416,9 @@ impl IndexScheduler {
// candidates not empty => index not empty => a primary key is set // candidates not empty => index not empty => a primary key is set
let primary_key = index.primary_key(&rtxn)?.unwrap(); let primary_key = index.primary_key(&rtxn)?.unwrap();
let primary_key = PrimaryKey::new_or_insert(primary_key, &mut new_fields_ids_map) let primary_key =
.map_err(milli::Error::from)?; PrimaryKey::new_or_insert(primary_key, &mut new_fields_ids_map)
.map_err(|err| Error::from_milli(err.into(), Some(index_uid.clone())))?;
let result_count = Ok((candidates.len(), candidates.len())) as Result<_>; let result_count = Ok((candidates.len(), candidates.len())) as Result<_>;
@@ -1406,11 +1437,17 @@ impl IndexScheduler {
}; };
let indexer = UpdateByFunction::new(candidates, context.clone(), code.clone()); let indexer = UpdateByFunction::new(candidates, context.clone(), code.clone());
let document_changes = let document_changes = pool
pool.install(|| indexer.into_changes(&primary_key)).unwrap()?; .install(|| {
indexer
let embedders = index.embedding_configs(index_wtxn)?; .into_changes(&primary_key)
let embedders = self.embedders(embedders)?; .map_err(|err| Error::from_milli(err, Some(index_uid.clone())))
})
.unwrap()?;
let embedders = index
.embedding_configs(index_wtxn)
.map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?;
let embedders = self.embedders(index_uid.clone(), embedders)?;
indexer::index( indexer::index(
index_wtxn, index_wtxn,
@@ -1424,7 +1461,8 @@ impl IndexScheduler {
embedders, embedders,
&|| must_stop_processing.get(), &|| must_stop_processing.get(),
&send_progress, &send_progress,
)?; )
.map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?;
// tracing::info!(indexing_result = ?addition, processed_in = ?started_processing_at.elapsed(), "document indexing done"); // tracing::info!(indexing_result = ?addition, processed_in = ?started_processing_at.elapsed(), "document indexing done");
} }
@@ -1455,7 +1493,7 @@ impl IndexScheduler {
Ok(vec![task]) Ok(vec![task])
} }
IndexOperation::DocumentDeletion { mut tasks, index_uid: _ } => { IndexOperation::DocumentDeletion { mut tasks, index_uid } => {
let mut to_delete = RoaringBitmap::new(); let mut to_delete = RoaringBitmap::new();
let external_documents_ids = index.external_documents_ids(); let external_documents_ids = index.external_documents_ids();
@@ -1476,35 +1514,23 @@ impl IndexScheduler {
deleted_documents: Some(will_be_removed), deleted_documents: Some(will_be_removed),
}); });
} }
KindWithContent::DocumentDeletionByFilter { index_uid: _, filter_expr } => { KindWithContent::DocumentDeletionByFilter { index_uid, filter_expr } => {
let before = to_delete.len(); let before = to_delete.len();
let filter = match Filter::from_json(filter_expr) { let filter = match Filter::from_json(filter_expr) {
Ok(filter) => filter, Ok(filter) => filter,
Err(err) => { Err(err) => {
// theorically, this should be catched by deserr before reaching the index-scheduler and cannot happens // theorically, this should be catched by deserr before reaching the index-scheduler and cannot happens
task.status = Status::Failed; task.status = Status::Failed;
task.error = match err { task.error = Some(
milli::Error::UserError( Error::from_milli(err, Some(index_uid.clone())).into(),
milli::UserError::InvalidFilterExpression { .. }, );
) => Some(
Error::from(err)
.with_custom_error_code(Code::InvalidDocumentFilter)
.into(),
),
e => Some(e.into()),
};
None None
} }
}; };
if let Some(filter) = filter { if let Some(filter) = filter {
let candidates = let candidates = filter
filter.evaluate(index_wtxn, index).map_err(|err| match err { .evaluate(index_wtxn, index)
milli::Error::UserError( .map_err(|err| Error::from_milli(err, Some(index_uid.clone())));
milli::UserError::InvalidFilter(_),
) => Error::from(err)
.with_custom_error_code(Code::InvalidDocumentFilter),
e => e.into(),
});
match candidates { match candidates {
Ok(candidates) => to_delete |= candidates, Ok(candidates) => to_delete |= candidates,
Err(err) => { Err(err) => {
@@ -1540,8 +1566,9 @@ impl IndexScheduler {
// to_delete not empty => index not empty => primary key set // to_delete not empty => index not empty => primary key set
let primary_key = index.primary_key(&rtxn)?.unwrap(); let primary_key = index.primary_key(&rtxn)?.unwrap();
let primary_key = PrimaryKey::new_or_insert(primary_key, &mut new_fields_ids_map) let primary_key =
.map_err(milli::Error::from)?; PrimaryKey::new_or_insert(primary_key, &mut new_fields_ids_map)
.map_err(|err| Error::from_milli(err.into(), Some(index_uid.clone())))?;
if !tasks.iter().all(|res| res.error.is_some()) { if !tasks.iter().all(|res| res.error.is_some()) {
let local_pool; let local_pool;
@@ -1560,8 +1587,10 @@ impl IndexScheduler {
let mut indexer = indexer::DocumentDeletion::new(); let mut indexer = indexer::DocumentDeletion::new();
indexer.delete_documents_by_docids(to_delete); indexer.delete_documents_by_docids(to_delete);
let document_changes = indexer.into_changes(&indexer_alloc, primary_key); let document_changes = indexer.into_changes(&indexer_alloc, primary_key);
let embedders = index.embedding_configs(index_wtxn)?; let embedders = index
let embedders = self.embedders(embedders)?; .embedding_configs(index_wtxn)
.map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?;
let embedders = self.embedders(index_uid.clone(), embedders)?;
indexer::index( indexer::index(
index_wtxn, index_wtxn,
@@ -1575,14 +1604,15 @@ impl IndexScheduler {
embedders, embedders,
&|| must_stop_processing.get(), &|| must_stop_processing.get(),
&send_progress, &send_progress,
)?; )
.map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?;
// tracing::info!(indexing_result = ?addition, processed_in = ?started_processing_at.elapsed(), "document indexing done"); // tracing::info!(indexing_result = ?addition, processed_in = ?started_processing_at.elapsed(), "document indexing done");
} }
Ok(tasks) Ok(tasks)
} }
IndexOperation::Settings { index_uid: _, settings, mut tasks } => { IndexOperation::Settings { index_uid, settings, mut tasks } => {
let indexer_config = self.index_mapper.indexer_config(); let indexer_config = self.index_mapper.indexer_config();
let mut builder = milli::update::Settings::new(index_wtxn, index, indexer_config); let mut builder = milli::update::Settings::new(index_wtxn, index, indexer_config);
@@ -1596,10 +1626,12 @@ impl IndexScheduler {
task.status = Status::Succeeded; task.status = Status::Succeeded;
} }
builder.execute( builder
|indexing_step| tracing::debug!(update = ?indexing_step), .execute(
|| must_stop_processing.get(), |indexing_step| tracing::debug!(update = ?indexing_step),
)?; || must_stop_processing.get(),
)
.map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?;
Ok(tasks) Ok(tasks)
} }

23
crates/index-scheduler/src/error.rs
View File

@@ -1,13 +1,12 @@
use std::fmt::Display; use std::fmt::Display;
use crate::TaskId;
use meilisearch_types::batches::BatchId; use meilisearch_types::batches::BatchId;
use meilisearch_types::error::{Code, ErrorCode}; use meilisearch_types::error::{Code, ErrorCode};
use meilisearch_types::tasks::{Kind, Status}; use meilisearch_types::tasks::{Kind, Status};
use meilisearch_types::{heed, milli}; use meilisearch_types::{heed, milli};
use thiserror::Error; use thiserror::Error;
use crate::TaskId;
#[derive(Copy, Clone, Debug, PartialEq, Eq)] #[derive(Copy, Clone, Debug, PartialEq, Eq)]
pub enum DateField { pub enum DateField {
BeforeEnqueuedAt, BeforeEnqueuedAt,
@@ -122,11 +121,11 @@ pub enum Error {
Dump(#[from] dump::Error), Dump(#[from] dump::Error),
#[error(transparent)] #[error(transparent)]
Heed(#[from] heed::Error), Heed(#[from] heed::Error),
#[error("{}", match .index_name { #[error("{}", match .index_uid {
Some(name) if !name.is_empty() => format!("Index `{}`: {error}", name), Some(uid) if !uid.is_empty() => format!("Index `{}`: {error}", uid),
_ => format!("{error}") _ => format!("{error}")
})] })]
Milli { error: milli::Error, index_name: Option<String> }, Milli { error: milli::Error, index_uid: Option<String> },
#[error("An unexpected crash occurred when processing the task.")] #[error("An unexpected crash occurred when processing the task.")]
ProcessBatchPanicked, ProcessBatchPanicked,
#[error(transparent)] #[error(transparent)]
@@ -213,8 +212,18 @@ impl Error {
Self::WithCustomErrorCode(code, Box::new(self)) Self::WithCustomErrorCode(code, Box::new(self))
} }
pub fn from_milli(error: milli::Error, index_name: Option<String>) -> Self { pub fn from_milli(err: milli::Error, index_uid: Option<String>) -> Self {
Self::Milli { error, index_name } match err {
milli::Error::UserError(milli::UserError::InvalidFilter(_)) => {
Self::Milli { error: err, index_uid }
.with_custom_error_code(Code::InvalidDocumentFilter)
}
milli::Error::UserError(milli::UserError::InvalidFilterExpression { .. }) => {
Self::Milli { error: err, index_uid }
.with_custom_error_code(Code::InvalidDocumentFilter)
}
_ => Self::Milli { error: err, index_uid },
}
} }
} }

3
crates/index-scheduler/src/index_mapper/index_map.rs
View File

@@ -8,9 +8,8 @@ use time::OffsetDateTime;
use uuid::Uuid; use uuid::Uuid;
use super::IndexStatus::{self, Available, BeingDeleted, Closing, Missing}; use super::IndexStatus::{self, Available, BeingDeleted, Closing, Missing};
use crate::clamp_to_page_size;
use crate::lru::{InsertionOutcome, LruMap}; use crate::lru::{InsertionOutcome, LruMap};
use crate::{clamp_to_page_size};
/// Keep an internally consistent view of the open indexes in memory. /// Keep an internally consistent view of the open indexes in memory.
/// ///
/// This view is made of an LRU cache that will evict the least frequently used indexes when new indexes are opened. /// This view is made of an LRU cache that will evict the least frequently used indexes when new indexes are opened.

51
crates/index-scheduler/src/index_mapper/mod.rs
View File

@@ -3,19 +3,19 @@ use std::sync::{Arc, RwLock};
use std::time::Duration; use std::time::Duration;
use std::{fs, thread}; use std::{fs, thread};
use self::index_map::IndexMap;
use self::IndexStatus::{Available, BeingDeleted, Closing, Missing};
use crate::uuid_codec::UuidCodec;
use crate::{Error, Result};
use meilisearch_types::heed::types::{SerdeJson, Str}; use meilisearch_types::heed::types::{SerdeJson, Str};
use meilisearch_types::heed::{Database, Env, RoTxn, RwTxn}; use meilisearch_types::heed::{Database, Env, RoTxn, RwTxn};
use meilisearch_types::milli;
use meilisearch_types::milli::update::IndexerConfig; use meilisearch_types::milli::update::IndexerConfig;
use meilisearch_types::milli::{FieldDistribution, Index}; use meilisearch_types::milli::{FieldDistribution, Index};
use serde::{Deserialize, Serialize}; use serde::{Deserialize, Serialize};
use time::OffsetDateTime; use time::OffsetDateTime;
use tracing::error; use tracing::error;
use uuid::Uuid; use uuid::Uuid;
use meilisearch_types::milli;
use self::index_map::IndexMap;
use self::IndexStatus::{Available, BeingDeleted, Closing, Missing};
use crate::uuid_codec::UuidCodec;
use crate::{Error, Result};
mod index_map; mod index_map;
@@ -183,13 +183,18 @@ impl IndexMapper {
// Error if the UUIDv4 somehow already exists in the map, since it should be fresh. // Error if the UUIDv4 somehow already exists in the map, since it should be fresh.
// This is very unlikely to happen in practice. // This is very unlikely to happen in practice.
// TODO: it would be better to lazily create the index. But we need an Index::open function for milli. // TODO: it would be better to lazily create the index. But we need an Index::open function for milli.
let index = self.index_map.write().unwrap().create( let index = self
&uuid, .index_map
&index_path, .write()
date, .unwrap()
self.enable_mdb_writemap, .create(
self.index_base_map_size, &uuid,
).map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?; &index_path,
date,
self.enable_mdb_writemap,
self.index_base_map_size,
)
.map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?;
wtxn.commit()?; wtxn.commit()?;
@@ -357,7 +362,8 @@ impl IndexMapper {
}; };
let index_path = self.base_path.join(uuid.to_string()); let index_path = self.base_path.join(uuid.to_string());
// take the lock to reopen the environment. // take the lock to reopen the environment.
reopen.reopen(&mut self.index_map.write().unwrap(), &index_path) reopen
.reopen(&mut self.index_map.write().unwrap(), &index_path)
.map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?; .map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?;
continue; continue;
} }
@@ -373,13 +379,15 @@ impl IndexMapper {
Missing => { Missing => {
let index_path = self.base_path.join(uuid.to_string()); let index_path = self.base_path.join(uuid.to_string());
break index_map.create( break index_map
&uuid, .create(
&index_path, &uuid,
None, &index_path,
self.enable_mdb_writemap, None,
self.index_base_map_size, self.enable_mdb_writemap,
).map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?; self.index_base_map_size,
)
.map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?;
} }
Available(index) => break index, Available(index) => break index,
Closing(_) => { Closing(_) => {
@@ -460,7 +468,8 @@ impl IndexMapper {
None => { None => {
let index = self.index(rtxn, index_uid)?; let index = self.index(rtxn, index_uid)?;
let index_rtxn = index.read_txn()?; let index_rtxn = index.read_txn()?;
IndexStats::new(&index, &index_rtxn).map_err(|e| Error::from_milli(e, Some(uuid.to_string()))) IndexStats::new(&index, &index_rtxn)
.map_err(|e| Error::from_milli(e, Some(uuid.to_string())))
} }
} }
} }

24
crates/index-scheduler/src/lib.rs
View File

@@ -1678,8 +1678,9 @@ impl IndexScheduler {
tracing::info!("A batch of tasks was successfully completed with {success} successful tasks and {failure} failed tasks."); tracing::info!("A batch of tasks was successfully completed with {success} successful tasks and {failure} failed tasks.");
} }
// If we have an abortion error we must stop the tick here and re-schedule tasks. // If we have an abortion error we must stop the tick here and re-schedule tasks.
Err(Error::Milli{ Err(Error::Milli {
error: milli::Error::InternalError(milli::InternalError::AbortedIndexation), .. error: milli::Error::InternalError(milli::InternalError::AbortedIndexation),
..
}) })
| Err(Error::AbortedTask) => { | Err(Error::AbortedTask) => {
#[cfg(test)] #[cfg(test)]
@@ -1700,7 +1701,8 @@ impl IndexScheduler {
// 3. resize it // 3. resize it
// 4. re-schedule tasks // 4. re-schedule tasks
Err(Error::Milli { Err(Error::Milli {
error: milli::Error::UserError(milli::UserError::MaxDatabaseSizeReached), .. error: milli::Error::UserError(milli::UserError::MaxDatabaseSizeReached),
..
}) if index_uid.is_some() => { }) if index_uid.is_some() => {
// fixme: add index_uid to match to avoid the unwrap // fixme: add index_uid to match to avoid the unwrap
let index_uid = index_uid.unwrap(); let index_uid = index_uid.unwrap();
@@ -1954,11 +1956,12 @@ impl IndexScheduler {
config: milli::vector::EmbeddingConfig { embedder_options, prompt, quantized }, config: milli::vector::EmbeddingConfig { embedder_options, prompt, quantized },
.. ..
}| { }| {
let prompt = let prompt = Arc::new(
Arc::new(prompt.try_into() prompt
.try_into()
.map_err(meilisearch_types::milli::Error::from) .map_err(meilisearch_types::milli::Error::from)
.map_err(|e| Error::from_milli(e, Some(index_uid.clone())))? .map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?,
); );
// optimistically return existing embedder // optimistically return existing embedder
{ {
let embedders = self.embedders.read().unwrap(); let embedders = self.embedders.read().unwrap();
@@ -1974,8 +1977,9 @@ impl IndexScheduler {
let embedder = Arc::new( let embedder = Arc::new(
Embedder::new(embedder_options.clone()) Embedder::new(embedder_options.clone())
.map_err(meilisearch_types::milli::vector::Error::from) .map_err(meilisearch_types::milli::vector::Error::from)
.map_err(meilisearch_types::milli::Error::from) .map_err(|err| {
.map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?, Error::from_milli(err.into(), Some(index_uid.clone()))
})?,
); );
{ {
let mut embedders = self.embedders.write().unwrap(); let mut embedders = self.embedders.write().unwrap();
@@ -6176,7 +6180,7 @@ mod tests {
insta::assert_json_snapshot!(simple_hf_config.embedder_options); insta::assert_json_snapshot!(simple_hf_config.embedder_options);
let simple_hf_name = name.clone(); let simple_hf_name = name.clone();
let configs = index_scheduler.embedders(configs).unwrap(); let configs = index_scheduler.embedders("doggos".to_string(), configs).unwrap();
let (hf_embedder, _, _) = configs.get(&simple_hf_name).unwrap(); let (hf_embedder, _, _) = configs.get(&simple_hf_name).unwrap();
let beagle_embed = let beagle_embed =
hf_embedder.embed_one(S("Intel the beagle best doggo"), None).unwrap(); hf_embedder.embed_one(S("Intel the beagle best doggo"), None).unwrap();

4
crates/index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/after_removing_the_documents.snap
View File

@@ -9,8 +9,8 @@ source: crates/index-scheduler/src/lib.rs
0 {uid: 0, batch_uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, facet_search: NotSet, prefix_search: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, facet_search: NotSet, prefix_search: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }} 0 {uid: 0, batch_uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, facet_search: NotSet, prefix_search: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, facet_search: NotSet, prefix_search: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
1 {uid: 1, batch_uid: 1, status: succeeded, details: { received_documents: 3, indexed_documents: Some(3) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }} 1 {uid: 1, batch_uid: 1, status: succeeded, details: { received_documents: 3, indexed_documents: Some(3) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
2 {uid: 2, batch_uid: 2, status: succeeded, details: { received_document_ids: 1, deleted_documents: Some(1) }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1"] }} 2 {uid: 2, batch_uid: 2, status: succeeded, details: { received_document_ids: 1, deleted_documents: Some(1) }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1"] }}
3 {uid: 3, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Invalid type for filter subexpression: expected: String, Array, found: true.", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: true, deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: Bool(true) }} 3 {uid: 3, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Index `doggos`: Invalid type for filter subexpression: expected: String, Array, found: true.", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: true, deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: Bool(true) }}
4 {uid: 4, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Attribute `id` is not filterable. Available filterable attributes are: `catto`.\n1:3 id = 2", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: "id = 2", deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("id = 2") }} 4 {uid: 4, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Index `doggos`: Attribute `id` is not filterable. Available filterable attributes are: `catto`.\n1:3 id = 2", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: "id = 2", deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("id = 2") }}
5 {uid: 5, batch_uid: 2, status: succeeded, details: { original_filter: "catto EXISTS", deleted_documents: Some(1) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("catto EXISTS") }} 5 {uid: 5, batch_uid: 2, status: succeeded, details: { original_filter: "catto EXISTS", deleted_documents: Some(1) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("catto EXISTS") }}
---------------------------------------------------------------------- ----------------------------------------------------------------------
### Status: ### Status:

6
crates/meilisearch/src/error.rs
View File

@@ -4,10 +4,10 @@ use byte_unit::{Byte, UnitType};
use meilisearch_types::document_formats::{DocumentFormatError, PayloadType}; use meilisearch_types::document_formats::{DocumentFormatError, PayloadType};
use meilisearch_types::error::{Code, ErrorCode, ResponseError}; use meilisearch_types::error::{Code, ErrorCode, ResponseError};
use meilisearch_types::index_uid::{IndexUid, IndexUidFormatError}; use meilisearch_types::index_uid::{IndexUid, IndexUidFormatError};
use meilisearch_types::milli;
use meilisearch_types::milli::OrderBy; use meilisearch_types::milli::OrderBy;
use serde_json::Value; use serde_json::Value;
use tokio::task::JoinError; use tokio::task::JoinError;
use meilisearch_types::milli;
#[derive(Debug, thiserror::Error)] #[derive(Debug, thiserror::Error)]
pub enum MeilisearchHttpError { pub enum MeilisearchHttpError {
@@ -67,7 +67,7 @@ pub enum MeilisearchHttpError {
Some(name) if !name.is_empty() => format!("Index `{}`: {error}", name), Some(name) if !name.is_empty() => format!("Index `{}`: {error}", name),
_ => format!("{error}") _ => format!("{error}")
})] })]
Milli { error: meilisearch_types::milli::Error, index_name: Option<String> }, Milli { error: milli::Error, index_name: Option<String> },
#[error(transparent)] #[error(transparent)]
Payload(#[from] PayloadError), Payload(#[from] PayloadError),
#[error(transparent)] #[error(transparent)]
@@ -105,7 +105,7 @@ impl ErrorCode for MeilisearchHttpError {
MeilisearchHttpError::SerdeJson(_) => Code::Internal, MeilisearchHttpError::SerdeJson(_) => Code::Internal,
MeilisearchHttpError::HeedError(_) => Code::Internal, MeilisearchHttpError::HeedError(_) => Code::Internal,
MeilisearchHttpError::IndexScheduler(e) => e.error_code(), MeilisearchHttpError::IndexScheduler(e) => e.error_code(),
MeilisearchHttpError::Milli{error, ..} => error.error_code(), MeilisearchHttpError::Milli { error, .. } => error.error_code(),
MeilisearchHttpError::Payload(e) => e.error_code(), MeilisearchHttpError::Payload(e) => e.error_code(),
MeilisearchHttpError::FileStore(_) => Code::Internal, MeilisearchHttpError::FileStore(_) => Code::Internal,
MeilisearchHttpError::DocumentFormat(e) => e.error_code(), MeilisearchHttpError::DocumentFormat(e) => e.error_code(),

3
crates/meilisearch/src/routes/indexes/facet_search.rs
View File

@@ -185,7 +185,8 @@ pub async fn search(
let index = index_scheduler.index(&index_uid)?; let index = index_scheduler.index(&index_uid)?;
let features = index_scheduler.features(); let features = index_scheduler.features();
let search_kind = search_kind(&search_query, &index_scheduler, index_uid.to_string(), &index, features)?; let search_kind =
search_kind(&search_query, &index_scheduler, index_uid.to_string(), &index, features)?;
let permit = search_queue.try_get_search_permit().await?; let permit = search_queue.try_get_search_permit().await?;
let search_result = tokio::task::spawn_blocking(move || { let search_result = tokio::task::spawn_blocking(move || {
perform_facet_search( perform_facet_search(

5
crates/meilisearch/src/routes/indexes/mod.rs
View File

@@ -107,7 +107,10 @@ pub async fn list_indexes(
if !filters.is_index_authorized(uid) { if !filters.is_index_authorized(uid) {
return Ok(None); return Ok(None);
} }
Ok(Some(IndexView::new(uid.to_string(), index).map_err(|e| Error::from_milli(e, Some(uid.to_string())))?)) Ok(Some(
IndexView::new(uid.to_string(), index)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?,
))
})?; })?;
// Won't cause to open all indexes because IndexView doesn't keep the `Index` opened. // Won't cause to open all indexes because IndexView doesn't keep the `Index` opened.
let indexes: Vec<IndexView> = indexes.into_iter().flatten().collect(); let indexes: Vec<IndexView> = indexes.into_iter().flatten().collect();

24
crates/meilisearch/src/routes/indexes/search.rs
View File

@@ -243,11 +243,19 @@ pub async fn search_with_url_query(
let index = index_scheduler.index(&index_uid)?; let index = index_scheduler.index(&index_uid)?;
let features = index_scheduler.features(); let features = index_scheduler.features();
let search_kind = search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index, features)?; let search_kind =
search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index, features)?;
let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features)?; let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features)?;
let permit = search_queue.try_get_search_permit().await?; let permit = search_queue.try_get_search_permit().await?;
let search_result = tokio::task::spawn_blocking(move || { let search_result = tokio::task::spawn_blocking(move || {
perform_search(index_uid.to_string(), &index, query, search_kind, retrieve_vector, index_scheduler.features()) perform_search(
index_uid.to_string(),
&index,
query,
search_kind,
retrieve_vector,
index_scheduler.features(),
)
}) })
.await; .await;
permit.drop().await; permit.drop().await;
@@ -287,12 +295,20 @@ pub async fn search_with_post(
let features = index_scheduler.features(); let features = index_scheduler.features();
let search_kind = search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index, features)?; let search_kind =
search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index, features)?;
let retrieve_vectors = RetrieveVectors::new(query.retrieve_vectors, features)?; let retrieve_vectors = RetrieveVectors::new(query.retrieve_vectors, features)?;
let permit = search_queue.try_get_search_permit().await?; let permit = search_queue.try_get_search_permit().await?;
let search_result = tokio::task::spawn_blocking(move || { let search_result = tokio::task::spawn_blocking(move || {
perform_search(index_uid.to_string(), &index, query, search_kind, retrieve_vectors, index_scheduler.features()) perform_search(
index_uid.to_string(),
&index,
query,
search_kind,
retrieve_vectors,
index_scheduler.features(),
)
}) })
.await; .await;
permit.drop().await; permit.drop().await;

9
crates/meilisearch/src/routes/indexes/similar.rs
View File

@@ -103,8 +103,13 @@ async fn similar(
let index = index_scheduler.index(&index_uid)?; let index = index_scheduler.index(&index_uid)?;
let (embedder_name, embedder, quantized) = let (embedder_name, embedder, quantized) = SearchKind::embedder(
SearchKind::embedder(&index_scheduler, index_uid.to_string(), &index, &query.embedder, None)?; &index_scheduler,
index_uid.to_string(),
&index,
&query.embedder,
None,
)?;
tokio::task::spawn_blocking(move || { tokio::task::spawn_blocking(move || {
perform_similar( perform_similar(

20
crates/meilisearch/src/routes/multi_search.rs
View File

@@ -127,14 +127,26 @@ pub async fn multi_search_with_post(
let index_uid_str = index_uid.to_string(); let index_uid_str = index_uid.to_string();
let search_kind = let search_kind = search_kind(
search_kind(&query, index_scheduler.get_ref(), index_uid_str.clone(), &index, features) &query,
.with_index(query_index)?; index_scheduler.get_ref(),
index_uid_str.clone(),
&index,
features,
)
.with_index(query_index)?;
let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features) let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features)
.with_index(query_index)?; .with_index(query_index)?;
let search_result = tokio::task::spawn_blocking(move || { let search_result = tokio::task::spawn_blocking(move || {
perform_search(index_uid_str.clone(), &index, query, search_kind, retrieve_vector, features) perform_search(
index_uid_str.clone(),
&index,
query,
search_kind,
retrieve_vector,
features,
)
}) })
.await .await
.with_index(query_index)?; .with_index(query_index)?;

12
crates/meilisearch/src/search/federated.rs
View File

@@ -560,7 +560,8 @@ pub fn perform_federated_search(
// use an immediately invoked lambda to capture the result without returning from the function // use an immediately invoked lambda to capture the result without returning from the function
let res: Result<(), ResponseError> = (|| { let res: Result<(), ResponseError> = (|| {
let search_kind = search_kind(&query, index_scheduler, index_uid.to_string(), &index, features)?; let search_kind =
search_kind(&query, index_scheduler, index_uid.to_string(), &index, features)?;
let canonicalization_kind = match (&search_kind, &query.q) { let canonicalization_kind = match (&search_kind, &query.q) {
(SearchKind::SemanticOnly { .. }, _) => { (SearchKind::SemanticOnly { .. }, _) => {
@@ -636,7 +637,8 @@ pub fn perform_federated_search(
search.offset(0); search.offset(0);
search.limit(required_hit_count); search.limit(required_hit_count);
let (result, _semantic_hit_count) = super::search_from_kind(index_uid.to_string(), search_kind, search)?; let (result, _semantic_hit_count) =
super::search_from_kind(index_uid.to_string(), search_kind, search)?;
let format = AttributesFormat { let format = AttributesFormat {
attributes_to_retrieve: query.attributes_to_retrieve, attributes_to_retrieve: query.attributes_to_retrieve,
retrieve_vectors, retrieve_vectors,
@@ -670,8 +672,10 @@ pub fn perform_federated_search(
let formatter_builder = HitMaker::formatter_builder(matching_words, tokenizer); let formatter_builder = HitMaker::formatter_builder(matching_words, tokenizer);
let hit_maker = HitMaker::new(&index, &rtxn, format, formatter_builder) let hit_maker =
.map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid.to_string())))?; HitMaker::new(&index, &rtxn, format, formatter_builder).map_err(|e| {
MeilisearchHttpError::from_milli(e, Some(index_uid.to_string()))
})?;
results_by_query.push(SearchResultByQuery { results_by_query.push(SearchResultByQuery {
federation_options, federation_options,

55
crates/meilisearch/src/search/mod.rs
View File

@@ -19,7 +19,9 @@ use meilisearch_types::locales::Locale;
use meilisearch_types::milli::score_details::{ScoreDetails, ScoringStrategy}; use meilisearch_types::milli::score_details::{ScoreDetails, ScoringStrategy};
use meilisearch_types::milli::vector::parsed_vectors::ExplicitVectors; use meilisearch_types::milli::vector::parsed_vectors::ExplicitVectors;
use meilisearch_types::milli::vector::Embedder; use meilisearch_types::milli::vector::Embedder;
use meilisearch_types::milli::{FacetValueHit, OrderBy, SearchForFacetValues, TimeBudget}; use meilisearch_types::milli::{
FacetValueHit, InternalError, OrderBy, SearchForFacetValues, TimeBudget,
};
use meilisearch_types::settings::DEFAULT_PAGINATION_MAX_TOTAL_HITS; use meilisearch_types::settings::DEFAULT_PAGINATION_MAX_TOTAL_HITS;
use meilisearch_types::{milli, Document}; use meilisearch_types::{milli, Document};
use milli::tokenizer::{Language, TokenizerBuilder}; use milli::tokenizer::{Language, TokenizerBuilder};
@@ -281,35 +283,38 @@ pub enum SearchKind {
impl SearchKind { impl SearchKind {
pub(crate) fn semantic( pub(crate) fn semantic(
index_scheduler: &index_scheduler::IndexScheduler, index_scheduler: &index_scheduler::IndexScheduler,
index_uid: String,
index: &Index, index: &Index,
embedder_name: &str, embedder_name: &str,
vector_len: Option<usize>, vector_len: Option<usize>,
) -> Result<Self, ResponseError> { ) -> Result<Self, ResponseError> {
let (embedder_name, embedder, quantized) = let (embedder_name, embedder, quantized) =
Self::embedder(index_scheduler, index, embedder_name, vector_len)?; Self::embedder(index_scheduler, index_uid, index, embedder_name, vector_len)?;
Ok(Self::SemanticOnly { embedder_name, embedder, quantized }) Ok(Self::SemanticOnly { embedder_name, embedder, quantized })
} }
pub(crate) fn hybrid( pub(crate) fn hybrid(
index_scheduler: &index_scheduler::IndexScheduler, index_scheduler: &index_scheduler::IndexScheduler,
index_uid: String,
index: &Index, index: &Index,
embedder_name: &str, embedder_name: &str,
semantic_ratio: f32, semantic_ratio: f32,
vector_len: Option<usize>, vector_len: Option<usize>,
) -> Result<Self, ResponseError> { ) -> Result<Self, ResponseError> {
let (embedder_name, embedder, quantized) = let (embedder_name, embedder, quantized) =
Self::embedder(index_scheduler, index, embedder_name, vector_len)?; Self::embedder(index_scheduler, index_uid, index, embedder_name, vector_len)?;
Ok(Self::Hybrid { embedder_name, embedder, quantized, semantic_ratio }) Ok(Self::Hybrid { embedder_name, embedder, quantized, semantic_ratio })
} }
pub(crate) fn embedder( pub(crate) fn embedder(
index_scheduler: &index_scheduler::IndexScheduler, index_scheduler: &index_scheduler::IndexScheduler,
index_uid: String,
index: &Index, index: &Index,
embedder_name: &str, embedder_name: &str,
vector_len: Option<usize>, vector_len: Option<usize>,
) -> Result<(String, Arc<Embedder>, bool), ResponseError> { ) -> Result<(String, Arc<Embedder>, bool), ResponseError> {
let embedder_configs = index.embedding_configs(&index.read_txn()?)?; let embedder_configs = index.embedding_configs(&index.read_txn()?)?;
let embedders = index_scheduler.embedders(embedder_configs)?; let embedders = index_scheduler.embedders(index_uid, embedder_configs)?;
let (embedder, _, quantized) = embedders let (embedder, _, quantized) = embedders
.get(embedder_name) .get(embedder_name)
@@ -890,6 +895,7 @@ fn prepare_search<'t>(
} }
pub fn perform_search( pub fn perform_search(
index_uid: String,
index: &Index, index: &Index,
query: SearchQuery, query: SearchQuery,
search_kind: SearchKind, search_kind: SearchKind,
@@ -916,7 +922,7 @@ pub fn perform_search(
used_negative_operator, used_negative_operator,
}, },
semantic_hit_count, semantic_hit_count,
) = search_from_kind(search_kind, search)?; ) = search_from_kind(index_uid, search_kind, search)?;
let SearchQuery { let SearchQuery {
q, q,
@@ -1069,17 +1075,27 @@ fn compute_facet_distribution_stats<S: AsRef<str>>(
} }
pub fn search_from_kind( pub fn search_from_kind(
index_uid: String,
search_kind: SearchKind, search_kind: SearchKind,
search: milli::Search<'_>, search: milli::Search<'_>,
) -> Result<(milli::SearchResult, Option<u32>), MeilisearchHttpError> { ) -> Result<(milli::SearchResult, Option<u32>), MeilisearchHttpError> {
let (milli_result, semantic_hit_count) = match &search_kind { let (milli_result, semantic_hit_count) = match &search_kind {
SearchKind::KeywordOnly => (search.execute()?, None), SearchKind::KeywordOnly => {
let results = search
.execute()
.map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid.to_string())))?;
(results, None)
}
SearchKind::SemanticOnly { .. } => { SearchKind::SemanticOnly { .. } => {
let results = search.execute()?; let results = search
.execute()
.map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid.to_string())))?;
let semantic_hit_count = results.document_scores.len() as u32; let semantic_hit_count = results.document_scores.len() as u32;
(results, Some(semantic_hit_count)) (results, Some(semantic_hit_count))
} }
SearchKind::Hybrid { semantic_ratio, .. } => search.execute_hybrid(*semantic_ratio)?, SearchKind::Hybrid { semantic_ratio, .. } => search
.execute_hybrid(*semantic_ratio)
.map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid)))?,
}; };
Ok((milli_result, semantic_hit_count)) Ok((milli_result, semantic_hit_count))
} }
@@ -1181,7 +1197,7 @@ impl<'a> HitMaker<'a> {
rtxn: &'a RoTxn<'a>, rtxn: &'a RoTxn<'a>,
format: AttributesFormat, format: AttributesFormat,
mut formatter_builder: MatcherBuilder<'a>, mut formatter_builder: MatcherBuilder<'a>,
) -> Result<Self, MeilisearchHttpError> { ) -> milli::Result<Self> {
formatter_builder.crop_marker(format.crop_marker); formatter_builder.crop_marker(format.crop_marker);
formatter_builder.highlight_prefix(format.highlight_pre_tag); formatter_builder.highlight_prefix(format.highlight_pre_tag);
formatter_builder.highlight_suffix(format.highlight_post_tag); formatter_builder.highlight_suffix(format.highlight_post_tag);
@@ -1276,11 +1292,7 @@ impl<'a> HitMaker<'a> {
}) })
} }
pub fn make_hit( pub fn make_hit(&self, id: u32, score: &[ScoreDetails]) -> milli::Result<SearchHit> {
&self,
id: u32,
score: &[ScoreDetails],
) -> Result<SearchHit, MeilisearchHttpError> {
let (_, obkv) = let (_, obkv) =
self.index.iter_documents(self.rtxn, std::iter::once(id))?.next().unwrap()?; self.index.iter_documents(self.rtxn, std::iter::once(id))?.next().unwrap()?;
@@ -1323,7 +1335,10 @@ impl<'a> HitMaker<'a> {
.is_some_and(|conf| conf.user_provided.contains(id)); .is_some_and(|conf| conf.user_provided.contains(id));
let embeddings = let embeddings =
ExplicitVectors { embeddings: Some(vector.into()), regenerate: !user_provided }; ExplicitVectors { embeddings: Some(vector.into()), regenerate: !user_provided };
vectors.insert(name, serde_json::to_value(embeddings)?); vectors.insert(
name,
serde_json::to_value(embeddings).map_err(InternalError::SerdeJson)?,
);
} }
document.insert("_vectors".into(), vectors.into()); document.insert("_vectors".into(), vectors.into());
} }
@@ -1369,7 +1384,7 @@ fn make_hits<'a>(
format: AttributesFormat, format: AttributesFormat,
matching_words: milli::MatchingWords, matching_words: milli::MatchingWords,
documents_ids_scores: impl Iterator<Item = (u32, &'a Vec<ScoreDetails>)> + 'a, documents_ids_scores: impl Iterator<Item = (u32, &'a Vec<ScoreDetails>)> + 'a,
) -> Result<Vec<SearchHit>, MeilisearchHttpError> { ) -> milli::Result<Vec<SearchHit>> {
let mut documents = Vec::new(); let mut documents = Vec::new();
let dictionary = index.dictionary(rtxn)?; let dictionary = index.dictionary(rtxn)?;
@@ -1697,12 +1712,12 @@ fn make_document(
displayed_attributes: &BTreeSet<FieldId>, displayed_attributes: &BTreeSet<FieldId>,
field_ids_map: &FieldsIdsMap, field_ids_map: &FieldsIdsMap,
obkv: &obkv::KvReaderU16, obkv: &obkv::KvReaderU16,
) -> Result<Document, MeilisearchHttpError> { ) -> milli::Result<Document> {
let mut document = serde_json::Map::new(); let mut document = serde_json::Map::new();
// recreate the original json // recreate the original json
for (key, value) in obkv.iter() { for (key, value) in obkv.iter() {
let value = serde_json::from_slice(value)?; let value = serde_json::from_slice(value).map_err(InternalError::SerdeJson)?;
let key = field_ids_map.name(key).expect("Missing field name").to_string(); let key = field_ids_map.name(key).expect("Missing field name").to_string();
document.insert(key, value); document.insert(key, value);
@@ -1727,7 +1742,7 @@ fn format_fields(
displayable_ids: &BTreeSet<FieldId>, displayable_ids: &BTreeSet<FieldId>,
locales: Option<&[Language]>, locales: Option<&[Language]>,
localized_attributes: &[LocalizedAttributesRule], localized_attributes: &[LocalizedAttributesRule],
) -> Result<(Option<MatchesPosition>, Document), MeilisearchHttpError> { ) -> milli::Result<(Option<MatchesPosition>, Document)> {
let mut matches_position = compute_matches.then(BTreeMap::new); let mut matches_position = compute_matches.then(BTreeMap::new);
let mut document = document.clone(); let mut document = document.clone();
@@ -1905,7 +1920,7 @@ fn parse_filter_array(arr: &[Value]) -> Result<Option<Filter>, MeilisearchHttpEr
} }
} }
Ok(Filter::from_array(ands)?) Filter::from_array(ands).map_err(|e| MeilisearchHttpError::from_milli(e, None))
} }
#[cfg(test)] #[cfg(test)]

32
crates/meilisearch/tests/documents/add_documents.rs
View File

@@ -1681,7 +1681,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "The `_geo` field in the document with the id: `\"11\"` is not an object. Was expecting an object with the `_geo.lat` and `_geo.lng` fields but instead got `\"foobar\"`.", "message": "Index `test`: The `_geo` field in the document with the id: `\"11\"` is not an object. Was expecting an object with the `_geo.lat` and `_geo.lng` fields but instead got `\"foobar\"`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1719,7 +1719,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.", "message": "Index `test`: Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1757,7 +1757,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.", "message": "Index `test`: Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1795,7 +1795,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.", "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1833,7 +1833,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.", "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1871,7 +1871,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.", "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1909,7 +1909,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.", "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1947,7 +1947,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `false` and `true`.", "message": "Index `test`: Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `false` and `true`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1985,7 +1985,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.", "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2023,7 +2023,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.", "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2061,7 +2061,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `\"doggo\"` and `\"doggo\"`.", "message": "Index `test`: Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `\"doggo\"` and `\"doggo\"`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2099,7 +2099,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "The `_geo` field in the document with the id: `\"11\"` contains the following unexpected fields: `{\"doggo\":\"are the best\"}`.", "message": "Index `test`: The `_geo` field in the document with the id: `\"11\"` contains the following unexpected fields: `{\"doggo\":\"are the best\"}`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2138,7 +2138,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not parse longitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.", "message": "Index `test`: Could not parse longitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2175,7 +2175,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not parse latitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.", "message": "Index `test`: Could not parse latitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2212,7 +2212,7 @@ async fn add_documents_invalid_geo_field() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Could not parse latitude nor longitude in the document with the id: `\"13\"`. Was expecting finite numbers but instead got `null` and `null`.", "message": "Index `test`: Could not parse latitude nor longitude in the document with the id: `\"13\"`. Was expecting finite numbers but instead got `null` and `null`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2279,7 +2279,7 @@ async fn add_invalid_geo_and_then_settings() {
] ]
}, },
"error": { "error": {
"message": "Could not parse latitude in the document with the id: `\"11\"`. Was expecting a finite number but instead got `null`.", "message": "Index `test`: Could not parse latitude in the document with the id: `\"11\"`. Was expecting a finite number but instead got `null`.",
"code": "invalid_document_geo_field", "code": "invalid_document_geo_field",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_geo_field" "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"

4
crates/meilisearch/tests/documents/errors.rs
View File

@@ -604,7 +604,7 @@ async fn delete_document_by_filter() {
"originalFilter": "\"doggo = bernese\"" "originalFilter": "\"doggo = bernese\""
}, },
"error": { "error": {
"message": "Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese", "message": "Index `EMPTY_INDEX`: Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
"code": "invalid_document_filter", "code": "invalid_document_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_filter" "link": "https://docs.meilisearch.com/errors#invalid_document_filter"
@@ -636,7 +636,7 @@ async fn delete_document_by_filter() {
"originalFilter": "\"catto = jorts\"" "originalFilter": "\"catto = jorts\""
}, },
"error": { "error": {
"message": "Attribute `catto` is not filterable. Available filterable attributes are: `id`, `title`.\n1:6 catto = jorts", "message": "Index `SHARED_DOCUMENTS`: Attribute `catto` is not filterable. Available filterable attributes are: `id`, `title`.\n1:6 catto = jorts",
"code": "invalid_document_filter", "code": "invalid_document_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_filter" "link": "https://docs.meilisearch.com/errors#invalid_document_filter"

2
crates/meilisearch/tests/index/update_index.rs
View File

@@ -95,7 +95,7 @@ async fn error_update_existing_primary_key() {
let response = index.wait_task(2).await; let response = index.wait_task(2).await;
let expected_response = json!({ let expected_response = json!({
"message": "Index already has a primary key: `id`.", "message": "Index `test`: Index already has a primary key: `id`.",
"code": "index_primary_key_already_exists", "code": "index_primary_key_already_exists",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#index_primary_key_already_exists" "link": "https://docs.meilisearch.com/errors#index_primary_key_already_exists"

104
crates/meilisearch/tests/search/errors.rs
View File

@@ -711,7 +711,7 @@ async fn filter_invalid_attribute_array() {
index.wait_task(task.uid()).await; index.wait_task(task.uid()).await;
let expected_response = json!({ let expected_response = json!({
"message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", "message": format!("Index `{}`: Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", index.uid),
"code": "invalid_search_filter", "code": "invalid_search_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_filter" "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -733,7 +733,7 @@ async fn filter_invalid_attribute_string() {
index.wait_task(task.uid()).await; index.wait_task(task.uid()).await;
let expected_response = json!({ let expected_response = json!({
"message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", "message": format!("Index `{}`: Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", index.uid),
"code": "invalid_search_filter", "code": "invalid_search_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_filter" "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -940,7 +940,7 @@ async fn sort_unsortable_attribute() {
index.wait_task(response.uid()).await.succeeded(); index.wait_task(response.uid()).await.succeeded();
let expected_response = json!({ let expected_response = json!({
"message": "Attribute `title` is not sortable. Available sortable attributes are: `id`.", "message": format!("Index `{}`: Attribute `title` is not sortable. Available sortable attributes are: `id`.", index.uid),
"code": "invalid_search_sort", "code": "invalid_search_sort",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_sort" "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -998,7 +998,7 @@ async fn sort_unset_ranking_rule() {
index.wait_task(response.uid()).await.succeeded(); index.wait_task(response.uid()).await.succeeded();
let expected_response = json!({ let expected_response = json!({
"message": "You must specify where `sort` is listed in the rankingRules setting to use the sort parameter at search time.", "message": format!("Index `{}`: You must specify where `sort` is listed in the rankingRules setting to use the sort parameter at search time.", index.uid),
"code": "invalid_search_sort", "code": "invalid_search_sort",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_sort" "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -1024,19 +1024,18 @@ async fn search_on_unknown_field() {
index.update_settings_searchable_attributes(json!(["id", "title"])).await; index.update_settings_searchable_attributes(json!(["id", "title"])).await;
index.wait_task(response.uid()).await.succeeded(); index.wait_task(response.uid()).await.succeeded();
let expected_response = json!({
"message": format!("Index `{}`: Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.", index.uid),
"code": "invalid_search_attributes_to_search_on",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
});
index index
.search( .search(
json!({"q": "Captain Marvel", "attributesToSearchOn": ["unknown"]}), json!({"q": "Captain Marvel", "attributesToSearchOn": ["unknown"]}),
|response, code| { |response, code| {
snapshot!(code, @"400 Bad Request"); assert_eq!(response, expected_response);
snapshot!(json_string!(response), @r###" assert_eq!(code, 400);
{
"message": "Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.",
"code": "invalid_search_attributes_to_search_on",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
}
"###);
}, },
) )
.await; .await;
@@ -1050,19 +1049,18 @@ async fn search_on_unknown_field_plus_joker() {
index.update_settings_searchable_attributes(json!(["id", "title"])).await; index.update_settings_searchable_attributes(json!(["id", "title"])).await;
index.wait_task(response.uid()).await.succeeded(); index.wait_task(response.uid()).await.succeeded();
let expected_response = json!({
"message": format!("Index `{}`: Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.", index.uid),
"code": "invalid_search_attributes_to_search_on",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
});
index index
.search( .search(
json!({"q": "Captain Marvel", "attributesToSearchOn": ["*", "unknown"]}), json!({"q": "Captain Marvel", "attributesToSearchOn": ["*", "unknown"]}),
|response, code| { |response, code| {
snapshot!(code, @"400 Bad Request"); assert_eq!(response, expected_response);
snapshot!(json_string!(response), @r###" assert_eq!(code, 400);
{
"message": "Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.",
"code": "invalid_search_attributes_to_search_on",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
}
"###);
}, },
) )
.await; .await;
@@ -1071,15 +1069,8 @@ async fn search_on_unknown_field_plus_joker() {
.search( .search(
json!({"q": "Captain Marvel", "attributesToSearchOn": ["unknown", "*"]}), json!({"q": "Captain Marvel", "attributesToSearchOn": ["unknown", "*"]}),
|response, code| { |response, code| {
snapshot!(code, @"400 Bad Request"); assert_eq!(response, expected_response);
snapshot!(json_string!(response), @r###" assert_eq!(code, 400);
{
"message": "Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.",
"code": "invalid_search_attributes_to_search_on",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
}
"###);
}, },
) )
.await; .await;
@@ -1092,47 +1083,44 @@ async fn distinct_at_search_time() {
let (task, _) = index.create(None).await; let (task, _) = index.create(None).await;
index.wait_task(task.uid()).await.succeeded(); index.wait_task(task.uid()).await.succeeded();
let expected_response = json!({
"message": format!("Index `{}`: Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. This index does not have configured filterable attributes.", index.uid),
"code": "invalid_search_distinct",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
});
let (response, code) = let (response, code) =
index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await; index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await;
snapshot!(code, @"400 Bad Request"); assert_eq!(response, expected_response);
snapshot!(response, @r###" assert_eq!(code, 400);
{
"message": "Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. This index does not have configured filterable attributes.",
"code": "invalid_search_distinct",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
}
"###);
let (task, _) = index.update_settings_filterable_attributes(json!(["color", "machin"])).await; let (task, _) = index.update_settings_filterable_attributes(json!(["color", "machin"])).await;
index.wait_task(task.uid()).await; index.wait_task(task.uid()).await;
let expected_response = json!({
"message": format!("Index `{}`: Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, machin`.", index.uid),
"code": "invalid_search_distinct",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
});
let (response, code) = let (response, code) =
index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await; index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await;
snapshot!(code, @"400 Bad Request"); assert_eq!(response, expected_response);
snapshot!(response, @r###" assert_eq!(code, 400);
{
"message": "Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, machin`.",
"code": "invalid_search_distinct",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
}
"###);
let (task, _) = index.update_settings_displayed_attributes(json!(["color"])).await; let (task, _) = index.update_settings_displayed_attributes(json!(["color"])).await;
index.wait_task(task.uid()).await; index.wait_task(task.uid()).await;
let expected_response = json!({
"message": format!("Index `{}`: Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, <..hidden-attributes>`.", index.uid),
"code": "invalid_search_distinct",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
});
let (response, code) = let (response, code) =
index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await; index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await;
snapshot!(code, @"400 Bad Request"); assert_eq!(response, expected_response);
snapshot!(response, @r###" assert_eq!(code, 400);
{
"message": "Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, <..hidden-attributes>`.",
"code": "invalid_search_distinct",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
}
"###);
let (response, code) = let (response, code) =
index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": true})).await; index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": true})).await;

12
crates/meilisearch/tests/search/multi.rs
View File

@@ -1070,7 +1070,7 @@ async fn federation_one_query_error() {
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
"message": "Inside `.queries[1]`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto", "message": "Inside `.queries[1]`: Index `nested`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
"code": "invalid_search_filter", "code": "invalid_search_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_filter" "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -1102,7 +1102,7 @@ async fn federation_one_query_sort_error() {
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
"message": "Inside `.queries[1]`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.", "message": "Inside `.queries[1]`: Index `nested`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
"code": "invalid_search_sort", "code": "invalid_search_sort",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_sort" "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -1166,7 +1166,7 @@ async fn federation_multiple_query_errors() {
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
"message": "Inside `.queries[0]`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto", "message": "Inside `.queries[0]`: Index `test`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
"code": "invalid_search_filter", "code": "invalid_search_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_filter" "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -1198,7 +1198,7 @@ async fn federation_multiple_query_sort_errors() {
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
"message": "Inside `.queries[0]`: Attribute `title` is not sortable. This index does not have configured sortable attributes.", "message": "Inside `.queries[0]`: Index `test`: Attribute `title` is not sortable. This index does not have configured sortable attributes.",
"code": "invalid_search_sort", "code": "invalid_search_sort",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_sort" "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -1231,7 +1231,7 @@ async fn federation_multiple_query_errors_interleaved() {
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
"message": "Inside `.queries[1]`: Attribute `doggos` is not filterable. This index does not have configured filterable attributes.\n1:7 doggos IN [intel, kefir]", "message": "Inside `.queries[1]`: Index `nested`: Attribute `doggos` is not filterable. This index does not have configured filterable attributes.\n1:7 doggos IN [intel, kefir]",
"code": "invalid_search_filter", "code": "invalid_search_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_filter" "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -1264,7 +1264,7 @@ async fn federation_multiple_query_sort_errors_interleaved() {
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
"message": "Inside `.queries[1]`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.", "message": "Inside `.queries[1]`: Index `nested`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
"code": "invalid_search_sort", "code": "invalid_search_sort",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_search_sort" "link": "https://docs.meilisearch.com/errors#invalid_search_sort"

2
crates/meilisearch/tests/tasks/mod.rs
View File

@@ -448,7 +448,7 @@ async fn test_summarized_delete_documents_by_filter() {
"originalFilter": "\"doggo = bernese\"" "originalFilter": "\"doggo = bernese\""
}, },
"error": { "error": {
"message": "Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese", "message": "Index `test`: Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
"code": "invalid_document_filter", "code": "invalid_document_filter",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_filter" "link": "https://docs.meilisearch.com/errors#invalid_document_filter"

2
crates/meilisearch/tests/vector/binary_quantized.rs
View File

@@ -318,7 +318,7 @@ async fn try_to_disable_binary_quantization() {
} }
}, },
"error": { "error": {
"message": "`.embedders.manual.binaryQuantized`: Cannot disable the binary quantization.\n - Note: Binary quantization is a lossy operation that cannot be reverted.\n - Hint: Add a new embedder that is non-quantized and regenerate the vectors.", "message": "Index `doggo`: `.embedders.manual.binaryQuantized`: Cannot disable the binary quantization.\n - Note: Binary quantization is a lossy operation that cannot be reverted.\n - Hint: Add a new embedder that is non-quantized and regenerate the vectors.",
"code": "invalid_settings_embedders", "code": "invalid_settings_embedders",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_settings_embedders" "link": "https://docs.meilisearch.com/errors#invalid_settings_embedders"

24
crates/meilisearch/tests/vector/mod.rs
View File

@@ -250,7 +250,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -280,7 +280,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -311,7 +311,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Could not parse `._vectors.manual.regenerate`: invalid type: string \"yes please\", expected a boolean at line 1 column 26", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Could not parse `._vectors.manual.regenerate`: invalid type: string \"yes please\", expected a boolean at line 1 column 26",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -340,7 +340,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings`: expected null or an array, but found a boolean: `true`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings`: expected null or an array, but found a boolean: `true`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -369,7 +369,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0]`: expected a number or an array, but found a boolean: `true`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0]`: expected a number or an array, but found a boolean: `true`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -398,7 +398,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][0]`: expected a number, but found a boolean: `true`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][0]`: expected a number, but found a boolean: `true`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -440,7 +440,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected a number, but found an array: `[0.2,0.3]`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected a number, but found an array: `[0.2,0.3]`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -469,7 +469,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected an array, but found a number: `0.3`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected an array, but found a number: `0.3`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -498,7 +498,7 @@ async fn user_provided_embeddings_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][1]`: expected a number, but found a boolean: `true`", "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][1]`: expected a number, but found a boolean: `true`",
"code": "invalid_vectors_type", "code": "invalid_vectors_type",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_vectors_type" "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -539,7 +539,7 @@ async fn user_provided_vectors_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "While embedding documents for embedder `manual`: no vectors provided for document `40` and at least 4 other document(s)\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: opt-out for a document with `_vectors.manual: null`", "message": "Index `doggo`: While embedding documents for embedder `manual`: no vectors provided for document `40` and at least 4 other document(s)\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: opt-out for a document with `_vectors.manual: null`",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -569,7 +569,7 @@ async fn user_provided_vectors_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vector` by `_vectors` in 1 document(s).", "message": "Index `doggo`: While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vector` by `_vectors` in 1 document(s).",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -599,7 +599,7 @@ async fn user_provided_vectors_error() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vectors.manaul` by `_vectors.manual` in 1 document(s).", "message": "Index `doggo`: While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vectors.manaul` by `_vectors.manual` in 1 document(s).",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"

4
crates/meilisearch/tests/vector/openai.rs
View File

@@ -713,7 +713,7 @@ async fn bad_api_key() {
} }
}, },
"error": { "error": {
"message": "While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n - server replied with `{\"error\":{\"message\":\"Incorrect API key provided: Bearer doggo. You can find your API key at https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":\"invalid_api_key\"}}`\n - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables", "message": "Index `doggo`: While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n - server replied with `{\"error\":{\"message\":\"Incorrect API key provided: Bearer doggo. You can find your API key at https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":\"invalid_api_key\"}}`\n - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -757,7 +757,7 @@ async fn bad_api_key() {
} }
}, },
"error": { "error": {
"message": "While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n - server replied with `{\"error\":{\"message\":\"You didn't provide an API key. You need to provide your API key in an Authorization header using Bearer auth (i.e. Authorization: Bearer YOUR_KEY), or as the password field (with blank username) if you're accessing the API from your browser and are prompted for a username and password. You can obtain an API key from https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":null}}`\n - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables", "message": "Index `doggo`: While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n - server replied with `{\"error\":{\"message\":\"You didn't provide an API key. You need to provide your API key in an Authorization header using Bearer auth (i.e. Authorization: Bearer YOUR_KEY), or as the password field (with blank username) if you're accessing the API from your browser and are prompted for a username and password. You can obtain an API key from https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":null}}`\n - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"

24
crates/meilisearch/tests/vector/rest.rs
View File

@@ -985,7 +985,7 @@ async fn bad_settings() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response`, while extracting a single \"{{embedding}}\", expected `response` to be an array of numbers, but failed to parse server response:\n - invalid type: map, expected a sequence", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response`, while extracting a single \"{{embedding}}\", expected `response` to be an array of numbers, but failed to parse server response:\n - invalid type: map, expected a sequence",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1025,7 +1025,7 @@ async fn bad_settings() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "While embedding documents for embedder `rest`: runtime error: was expecting embeddings of dimension `2`, got embeddings of dimensions `3`", "message": "Index `doggo`: While embedding documents for embedder `rest`: runtime error: was expecting embeddings of dimension `2`, got embeddings of dimensions `3`",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1178,7 +1178,7 @@ async fn server_returns_bad_request() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with user error: sent a bad request to embedding server\n - Hint: check that the `request` in the embedder configuration matches the remote server's API\n - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"test\\\", expected struct MultipleRequest at line 1 column 6\"}`", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with user error: sent a bad request to embedding server\n - Hint: check that the `request` in the embedder configuration matches the remote server's API\n - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"test\\\", expected struct MultipleRequest at line 1 column 6\"}`",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1247,7 +1247,7 @@ async fn server_returns_bad_request() {
"indexedDocuments": 0 "indexedDocuments": 0
}, },
"error": { "error": {
"message": "While embedding documents for embedder `rest`: user error: sent a bad request to embedding server\n - Hint: check that the `request` in the embedder configuration matches the remote server's API\n - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"name: kefir\\\\n\\\", expected struct MultipleRequest at line 1 column 15\"}`", "message": "Index `doggo`: While embedding documents for embedder `rest`: user error: sent a bad request to embedding server\n - Hint: check that the `request` in the embedder configuration matches the remote server's API\n - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"name: kefir\\\\n\\\", expected struct MultipleRequest at line 1 column 15\"}`",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1306,7 +1306,7 @@ async fn server_returns_bad_response() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response`, while extracting the array of \"{{embedding}}\"s, configuration expects `response` to be an array with at least 1 item(s) but server sent an object with 1 field(s)", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response`, while extracting the array of \"{{embedding}}\"s, configuration expects `response` to be an array with at least 1 item(s) but server sent an object with 1 field(s)",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1362,7 +1362,7 @@ async fn server_returns_bad_response() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response`, while extracting item #0 from the array of \"{{embedding}}\"s, expected `response` to be an array of numbers, but failed to parse server response:\n - invalid type: map, expected a sequence", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response`, while extracting item #0 from the array of \"{{embedding}}\"s, expected `response` to be an array of numbers, but failed to parse server response:\n - invalid type: map, expected a sequence",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1414,7 +1414,7 @@ async fn server_returns_bad_response() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response.output`, while extracting a single \"{{embedding}}\", expected `output` to be an array of numbers, but failed to parse server response:\n - invalid type: map, expected f32", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response.output`, while extracting a single \"{{embedding}}\", expected `output` to be an array of numbers, but failed to parse server response:\n - invalid type: map, expected f32",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1478,7 +1478,7 @@ async fn server_returns_bad_response() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response.embedding`, while extracting item #0 from the array of \"{{embedding}}\"s, configuration expects `embedding` to be an object with key `data` but server sent an array of size 3", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response.embedding`, while extracting item #0 from the array of \"{{embedding}}\"s, configuration expects `embedding` to be an object with key `data` but server sent an array of size 3",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1542,7 +1542,7 @@ async fn server_returns_bad_response() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response.output[0]`, while extracting a single \"{{embedding}}\", configuration expects key \"embeddings\", which is missing in response\n - Hint: item #0 inside `output` has key `embedding`, did you mean `response.output[0].embedding` in embedder configuration?", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with runtime error: error extracting embeddings from the response:\n - in `response.output[0]`, while extracting a single \"{{embedding}}\", configuration expects key \"embeddings\", which is missing in response\n - Hint: item #0 inside `output` has key `embedding`, did you mean `response.output[0].embedding` in embedder configuration?",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1908,7 +1908,7 @@ async fn server_custom_header() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with user error: could not authenticate against embedding server\n - server replied with `{\"error\":\"missing header 'my-nonstandard-auth'\"}`\n - Hint: Check the `apiKey` parameter in the embedder configuration", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with user error: could not authenticate against embedding server\n - server replied with `{\"error\":\"missing header 'my-nonstandard-auth'\"}`\n - Hint: Check the `apiKey` parameter in the embedder configuration",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1951,7 +1951,7 @@ async fn server_custom_header() {
} }
}, },
"error": { "error": {
"message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with user error: could not authenticate against embedding server\n - server replied with `{\"error\":\"thou shall not pass, Balrog\"}`\n - Hint: Check the `apiKey` parameter in the embedder configuration", "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n - test embedding failed with user error: could not authenticate against embedding server\n - server replied with `{\"error\":\"thou shall not pass, Balrog\"}`\n - Hint: Check the `apiKey` parameter in the embedder configuration",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -2099,7 +2099,7 @@ async fn searchable_reindex() {
] ]
}, },
"error": { "error": {
"message": "While embedding documents for embedder `rest`: error: received unexpected HTTP 404 from embedding server\n - server replied with `{\"error\":\"text not found\",\"text\":\"breed: patou\\n\"}`", "message": "Index `doggo`: While embedding documents for embedder `rest`: error: received unexpected HTTP 404 from embedding server\n - server replied with `{\"error\":\"text not found\",\"text\":\"breed: patou\\n\"}`",
"code": "vector_embedding_error", "code": "vector_embedding_error",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#vector_embedding_error" "link": "https://docs.meilisearch.com/errors#vector_embedding_error"