searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-11-25 22:34:28 +01:00
Code Issues Releases Wiki Activity

Merge #4938

Browse Source 
4938: Remove default embedder r=ManyTheFish a=dureuill

# Pull Request

## Related issue
Fixes #4738 

## What does this PR do?

[See public usage](https://meilisearch.notion.site/v1-11-AI-search-changes-0e37727193884a70999f254fa953ce6e#1044b06b651f80edb9d4ef6dc367bad0)

- Remove `hybrid.embedder` boolean from analytics because embedder is now mandatory and so the boolean would always be `true`
- Rework search kind so that a search without query but with vector is a vector search regardless of (non-zero) semantic ratio


Co-authored-by: Louis Dureuil <louis@meilisearch.com>
This commit is contained in:
meili-bors[bot] 2024-09-19 09:17:14 +00:00 committed by GitHub
commit 5f474a640d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
13 changed files with 244 additions and 224 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
meilisearch/src/analytics/segment_analytics.rs
View File

@ -646,8 +646,6 @@ pub struct SearchAggregator {
max_vector_size: usize, max_vector_size: usize,
// Whether the semantic ratio passed to a hybrid search equals the default ratio. // Whether the semantic ratio passed to a hybrid search equals the default ratio.
semantic_ratio: bool, semantic_ratio: bool,
// Whether a non-default embedder was specified
embedder: bool,
hybrid: bool, hybrid: bool,
retrieve_vectors: bool, retrieve_vectors: bool,
@ -795,7 +793,6 @@ impl SearchAggregator {
if let Some(hybrid) = hybrid { if let Some(hybrid) = hybrid {
ret.semantic_ratio = hybrid.semantic_ratio != DEFAULT_SEMANTIC_RATIO(); ret.semantic_ratio = hybrid.semantic_ratio != DEFAULT_SEMANTIC_RATIO();
ret.embedder = hybrid.embedder.is_some();
ret.hybrid = true; ret.hybrid = true;
} }
@ -863,7 +860,6 @@ impl SearchAggregator {
show_ranking_score, show_ranking_score,
show_ranking_score_details, show_ranking_score_details,
semantic_ratio, semantic_ratio,
embedder,
hybrid, hybrid,
total_degraded, total_degraded,
total_used_negative_operator, total_used_negative_operator,
@ -923,7 +919,6 @@ impl SearchAggregator {
self.retrieve_vectors |= retrieve_vectors; self.retrieve_vectors |= retrieve_vectors;
self.semantic_ratio |= semantic_ratio; self.semantic_ratio |= semantic_ratio;
self.hybrid |= hybrid; self.hybrid |= hybrid;
self.embedder |= embedder;
// pagination // pagination
self.max_limit = self.max_limit.max(max_limit); self.max_limit = self.max_limit.max(max_limit);
@ -999,7 +994,6 @@ impl SearchAggregator {
show_ranking_score, show_ranking_score,
show_ranking_score_details, show_ranking_score_details,
semantic_ratio, semantic_ratio,
embedder,
hybrid, hybrid,
total_degraded, total_degraded,
total_used_negative_operator, total_used_negative_operator,
@ -1051,7 +1045,6 @@ impl SearchAggregator {
"hybrid": { "hybrid": {
"enabled": hybrid, "enabled": hybrid,
"semantic_ratio": semantic_ratio, "semantic_ratio": semantic_ratio,
"embedder": embedder,
}, },
"pagination": { "pagination": {
"max_limit": max_limit, "max_limit": max_limit,
@ -1782,7 +1775,6 @@ pub struct SimilarAggregator {
used_syntax: HashMap<String, usize>, used_syntax: HashMap<String, usize>,
// Whether a non-default embedder was specified // Whether a non-default embedder was specified
embedder: bool,
retrieve_vectors: bool, retrieve_vectors: bool,
// pagination // pagination
@ -1803,7 +1795,7 @@ impl SimilarAggregator {
pub fn from_query(query: &SimilarQuery, request: &HttpRequest) -> Self { pub fn from_query(query: &SimilarQuery, request: &HttpRequest) -> Self {
let SimilarQuery { let SimilarQuery {
id: _, id: _,
embedder, embedder: _,
offset, offset,
limit, limit,
attributes_to_retrieve: _, attributes_to_retrieve: _,
@ -1851,7 +1843,6 @@ impl SimilarAggregator {
ret.show_ranking_score_details = *show_ranking_score_details; ret.show_ranking_score_details = *show_ranking_score_details;
ret.ranking_score_threshold = ranking_score_threshold.is_some(); ret.ranking_score_threshold = ranking_score_threshold.is_some();
ret.embedder = embedder.is_some();
ret.retrieve_vectors = *retrieve_vectors; ret.retrieve_vectors = *retrieve_vectors;
ret ret
@ -1883,7 +1874,6 @@ impl SimilarAggregator {
max_attributes_to_retrieve, max_attributes_to_retrieve,
show_ranking_score, show_ranking_score,
show_ranking_score_details, show_ranking_score_details,
embedder,
ranking_score_threshold, ranking_score_threshold,
retrieve_vectors, retrieve_vectors,
} = other; } = other;
@ -1914,7 +1904,6 @@ impl SimilarAggregator {
*used_syntax = used_syntax.saturating_add(value); *used_syntax = used_syntax.saturating_add(value);
} }
self.embedder |= embedder;
self.retrieve_vectors |= retrieve_vectors; self.retrieve_vectors |= retrieve_vectors;
// pagination // pagination
@ -1948,7 +1937,6 @@ impl SimilarAggregator {
max_attributes_to_retrieve, max_attributes_to_retrieve,
show_ranking_score, show_ranking_score,
show_ranking_score_details, show_ranking_score_details,
embedder,
ranking_score_threshold, ranking_score_threshold,
retrieve_vectors, retrieve_vectors,
} = self; } = self;
@ -1980,9 +1968,6 @@ impl SimilarAggregator {
"vector": { "vector": {
"retrieve_vectors": retrieve_vectors, "retrieve_vectors": retrieve_vectors,
}, },
"hybrid": {
"embedder": embedder,
},
"pagination": { "pagination": {
"max_limit": max_limit, "max_limit": max_limit,
"max_offset": max_offset, "max_offset": max_offset,

2
meilisearch/src/error.rs
View File

@ -72,7 +72,7 @@ pub enum MeilisearchHttpError {
DocumentFormat(#[from] DocumentFormatError), DocumentFormat(#[from] DocumentFormatError),
#[error(transparent)] #[error(transparent)]
Join(#[from] JoinError), Join(#[from] JoinError),
#[error("Invalid request: missing `hybrid` parameter when both `q` and `vector` are present.")] #[error("Invalid request: missing `hybrid` parameter when `vector` is present.")]
MissingSearchHybrid, MissingSearchHybrid,
} }

91
meilisearch/src/routes/indexes/search.rs
View File

@ -128,8 +128,10 @@ impl std::ops::Deref for SemanticRatioGet {
} }
} }
impl From<SearchQueryGet> for SearchQuery { impl TryFrom<SearchQueryGet> for SearchQuery {
fn from(other: SearchQueryGet) -> Self { type Error = ResponseError;
fn try_from(other: SearchQueryGet) -> Result<Self, Self::Error> {
let filter = match other.filter { let filter = match other.filter {
Some(f) => match serde_json::from_str(&f) { Some(f) => match serde_json::from_str(&f) {
Ok(v) => Some(v), Ok(v) => Some(v),
@ -140,19 +142,28 @@ impl From<SearchQueryGet> for SearchQuery {
let hybrid = match (other.hybrid_embedder, other.hybrid_semantic_ratio) { let hybrid = match (other.hybrid_embedder, other.hybrid_semantic_ratio) {
(None, None) => None, (None, None) => None,
(None, Some(semantic_ratio)) => { (None, Some(_)) => {
Some(HybridQuery { semantic_ratio: *semantic_ratio, embedder: None }) return Err(ResponseError::from_msg(
"`hybridEmbedder` is mandatory when `hybridSemanticRatio` is present".into(),
meilisearch_types::error::Code::InvalidHybridQuery,
));
}
(Some(embedder), None) => {
Some(HybridQuery { semantic_ratio: DEFAULT_SEMANTIC_RATIO(), embedder })
} }
(Some(embedder), None) => Some(HybridQuery {
semantic_ratio: DEFAULT_SEMANTIC_RATIO(),
embedder: Some(embedder),
}),
(Some(embedder), Some(semantic_ratio)) => { (Some(embedder), Some(semantic_ratio)) => {
Some(HybridQuery { semantic_ratio: *semantic_ratio, embedder: Some(embedder) }) Some(HybridQuery { semantic_ratio: *semantic_ratio, embedder })
} }
}; };
Self { if other.vector.is_some() && hybrid.is_none() {
return Err(ResponseError::from_msg(
"`hybridEmbedder` is mandatory when `vector` is present".into(),
meilisearch_types::error::Code::MissingSearchHybrid,
));
}
Ok(Self {
q: other.q, q: other.q,
vector: other.vector.map(CS::into_inner), vector: other.vector.map(CS::into_inner),
offset: other.offset.0, offset: other.offset.0,
@ -179,7 +190,7 @@ impl From<SearchQueryGet> for SearchQuery {
hybrid, hybrid,
ranking_score_threshold: other.ranking_score_threshold.map(|o| o.0), ranking_score_threshold: other.ranking_score_threshold.map(|o| o.0),
locales: other.locales.map(|o| o.into_iter().collect()), locales: other.locales.map(|o| o.into_iter().collect()),
} })
} }
} }
@ -219,7 +230,7 @@ pub async fn search_with_url_query(
debug!(parameters = ?params, "Search get"); debug!(parameters = ?params, "Search get");
let index_uid = IndexUid::try_from(index_uid.into_inner())?; let index_uid = IndexUid::try_from(index_uid.into_inner())?;
let mut query: SearchQuery = params.into_inner().into(); let mut query: SearchQuery = params.into_inner().try_into()?;
// Tenant token search_rules. // Tenant token search_rules.
if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) { if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) {
@ -312,44 +323,36 @@ pub fn search_kind(
features.check_vector("Passing `hybrid` as a parameter")?; features.check_vector("Passing `hybrid` as a parameter")?;
} }
// regardless of anything, always do a keyword search when we don't have a vector and the query is whitespace or missing // handle with care, the order of cases matters, the semantics is subtle
if query.vector.is_none() { match (query.q.as_deref(), &query.hybrid, query.vector.as_deref()) {
match &query.q { // empty query, no vector => placeholder search
Some(q) if q.trim().is_empty() => return Ok(SearchKind::KeywordOnly), (Some(q), _, None) if q.trim().is_empty() => Ok(SearchKind::KeywordOnly),
None => return Ok(SearchKind::KeywordOnly), // no query, no vector => placeholder search
_ => {} (None, _, None) => Ok(SearchKind::KeywordOnly),
// hybrid.semantic_ratio == 1.0 => vector
(_, Some(HybridQuery { semantic_ratio, embedder }), v) if **semantic_ratio == 1.0 => {
SearchKind::semantic(index_scheduler, index, embedder, v.map(|v| v.len()))
} }
} // hybrid.semantic_ratio == 0.0 => keyword
(_, Some(HybridQuery { semantic_ratio, embedder: _ }), _) if **semantic_ratio == 0.0 => {
match &query.hybrid {
Some(HybridQuery { semantic_ratio, embedder }) if **semantic_ratio == 1.0 => {
Ok(SearchKind::semantic(
index_scheduler,
index,
embedder.as_deref(),
query.vector.as_ref().map(Vec::len),
)?)
}
Some(HybridQuery { semantic_ratio, embedder: _ }) if **semantic_ratio == 0.0 => {
Ok(SearchKind::KeywordOnly) Ok(SearchKind::KeywordOnly)
} }
Some(HybridQuery { semantic_ratio, embedder }) => Ok(SearchKind::hybrid( // no query, hybrid, vector => semantic
(None, Some(HybridQuery { semantic_ratio: _, embedder }), Some(v)) => {
SearchKind::semantic(index_scheduler, index, embedder, Some(v.len()))
}
// query, no hybrid, no vector => keyword
(Some(_), None, None) => Ok(SearchKind::KeywordOnly),
// query, hybrid, maybe vector => hybrid
(Some(_), Some(HybridQuery { semantic_ratio, embedder }), v) => SearchKind::hybrid(
index_scheduler, index_scheduler,
index, index,
embedder.as_deref(), embedder,
**semantic_ratio, **semantic_ratio,
query.vector.as_ref().map(Vec::len), v.map(|v| v.len()),
)?), ),
None => match (query.q.as_deref(), query.vector.as_deref()) {
(_query, None) => Ok(SearchKind::KeywordOnly), (_, None, Some(_)) => Err(MeilisearchHttpError::MissingSearchHybrid.into()),
(None, Some(_vector)) => Ok(SearchKind::semantic(
index_scheduler,
index,
None,
query.vector.as_ref().map(Vec::len),
)?),
(Some(_), Some(_)) => Err(MeilisearchHttpError::MissingSearchHybrid.into()),
},
} }
} }

6
meilisearch/src/routes/indexes/similar.rs
View File

@ -103,7 +103,7 @@ async fn similar(
let index = index_scheduler.index(&index_uid)?; let index = index_scheduler.index(&index_uid)?;
let (embedder_name, embedder) = let (embedder_name, embedder) =
SearchKind::embedder(&index_scheduler, &index, query.embedder.as_deref(), None)?; SearchKind::embedder(&index_scheduler, &index, &query.embedder, None)?;
tokio::task::spawn_blocking(move || { tokio::task::spawn_blocking(move || {
perform_similar( perform_similar(
@ -139,8 +139,8 @@ pub struct SimilarQueryGet {
show_ranking_score_details: Param<bool>, show_ranking_score_details: Param<bool>,
#[deserr(default, error = DeserrQueryParamError<InvalidSimilarRankingScoreThreshold>, default)] #[deserr(default, error = DeserrQueryParamError<InvalidSimilarRankingScoreThreshold>, default)]
pub ranking_score_threshold: Option<RankingScoreThresholdGet>, pub ranking_score_threshold: Option<RankingScoreThresholdGet>,
#[deserr(default, error = DeserrQueryParamError<InvalidEmbedder>)] #[deserr(error = DeserrQueryParamError<InvalidEmbedder>)]
pub embedder: Option<String>, pub embedder: String,
} }
#[derive(Debug, Clone, Copy, PartialEq, deserr::Deserr)] #[derive(Debug, Clone, Copy, PartialEq, deserr::Deserr)]

16
meilisearch/src/search/mod.rs
View File

@ -267,8 +267,8 @@ impl fmt::Debug for SearchQuery {
pub struct HybridQuery { pub struct HybridQuery {
#[deserr(default, error = DeserrJsonError<InvalidSearchSemanticRatio>, default)] #[deserr(default, error = DeserrJsonError<InvalidSearchSemanticRatio>, default)]
pub semantic_ratio: SemanticRatio, pub semantic_ratio: SemanticRatio,
#[deserr(default, error = DeserrJsonError<InvalidEmbedder>, default)] #[deserr(error = DeserrJsonError<InvalidEmbedder>)]
pub embedder: Option<String>, pub embedder: String,
} }
#[derive(Clone)] #[derive(Clone)]
@ -282,7 +282,7 @@ impl SearchKind {
pub(crate) fn semantic( pub(crate) fn semantic(
index_scheduler: &index_scheduler::IndexScheduler, index_scheduler: &index_scheduler::IndexScheduler,
index: &Index, index: &Index,
embedder_name: Option<&str>, embedder_name: &str,
vector_len: Option<usize>, vector_len: Option<usize>,
) -> Result<Self, ResponseError> { ) -> Result<Self, ResponseError> {
let (embedder_name, embedder) = let (embedder_name, embedder) =
@ -293,7 +293,7 @@ impl SearchKind {
pub(crate) fn hybrid( pub(crate) fn hybrid(
index_scheduler: &index_scheduler::IndexScheduler, index_scheduler: &index_scheduler::IndexScheduler,
index: &Index, index: &Index,
embedder_name: Option<&str>, embedder_name: &str,
semantic_ratio: f32, semantic_ratio: f32,
vector_len: Option<usize>, vector_len: Option<usize>,
) -> Result<Self, ResponseError> { ) -> Result<Self, ResponseError> {
@ -305,14 +305,12 @@ impl SearchKind {
pub(crate) fn embedder( pub(crate) fn embedder(
index_scheduler: &index_scheduler::IndexScheduler, index_scheduler: &index_scheduler::IndexScheduler,
index: &Index, index: &Index,
embedder_name: Option<&str>, embedder_name: &str,
vector_len: Option<usize>, vector_len: Option<usize>,
) -> Result<(String, Arc<Embedder>), ResponseError> { ) -> Result<(String, Arc<Embedder>), 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(embedder_configs)?;
let embedder_name = embedder_name.unwrap_or_else(|| embedders.get_default_embedder_name());
let embedder = embedders.get(embedder_name); let embedder = embedders.get(embedder_name);
let embedder = embedder let embedder = embedder
@ -538,8 +536,8 @@ pub struct SimilarQuery {
pub limit: usize, pub limit: usize,
#[deserr(default, error = DeserrJsonError<InvalidSimilarFilter>)] #[deserr(default, error = DeserrJsonError<InvalidSimilarFilter>)]
pub filter: Option<Value>, pub filter: Option<Value>,
#[deserr(default, error = DeserrJsonError<InvalidEmbedder>, default)] #[deserr(error = DeserrJsonError<InvalidEmbedder>)]
pub embedder: Option<String>, pub embedder: String,
#[deserr(default, error = DeserrJsonError<InvalidSimilarAttributesToRetrieve>)] #[deserr(default, error = DeserrJsonError<InvalidSimilarAttributesToRetrieve>)]
pub attributes_to_retrieve: Option<BTreeSet<String>>, pub attributes_to_retrieve: Option<BTreeSet<String>>,
#[deserr(default, error = DeserrJsonError<InvalidSimilarRetrieveVectors>)] #[deserr(default, error = DeserrJsonError<InvalidSimilarRetrieveVectors>)]

65
meilisearch/tests/search/hybrid.rs
View File

@ -128,7 +128,7 @@ async fn simple_search() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.2}, "retrieveVectors": true}), json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.2, "embedder": "default"}, "retrieveVectors": true}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -137,7 +137,7 @@ async fn simple_search() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.5}, "showRankingScore": true, "retrieveVectors": true}), json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.5, "embedder": "default"}, "showRankingScore": true, "retrieveVectors": true}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -146,7 +146,7 @@ async fn simple_search() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.8}, "showRankingScore": true, "retrieveVectors": true}), json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.8, "embedder": "default"}, "showRankingScore": true, "retrieveVectors": true}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -161,7 +161,7 @@ async fn limit_offset() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.2}, "retrieveVectors": true, "offset": 1, "limit": 1}), json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.2, "embedder": "default"}, "retrieveVectors": true, "offset": 1, "limit": 1}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -174,7 +174,7 @@ async fn limit_offset() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.9}, "retrieveVectors": true, "offset": 1, "limit": 1}), json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 0.9, "embedder": "default"}, "retrieveVectors": true, "offset": 1, "limit": 1}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -188,8 +188,11 @@ async fn simple_search_hf() {
let server = Server::new().await; let server = Server::new().await;
let index = index_with_documents_hf(&server, &SIMPLE_SEARCH_DOCUMENTS).await; let index = index_with_documents_hf(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
let (response, code) = let (response, code) = index
index.search_post(json!({"q": "Captain", "hybrid": {"semanticRatio": 0.2}})).await; .search_post(
json!({"q": "Captain", "hybrid": {"semanticRatio": 0.2, "embedder": "default"}}),
)
.await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2"},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3"},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1"}]"###); snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2"},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3"},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1"}]"###);
snapshot!(response["semanticHitCount"], @"0"); snapshot!(response["semanticHitCount"], @"0");
@ -197,7 +200,7 @@ async fn simple_search_hf() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
// disable ranking score as the vectors between architectures are not equal // disable ranking score as the vectors between architectures are not equal
json!({"q": "Captain", "hybrid": {"semanticRatio": 0.55}, "showRankingScore": false}), json!({"q": "Captain", "hybrid": {"embedder": "default", "semanticRatio": 0.55}, "showRankingScore": false}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -206,7 +209,7 @@ async fn simple_search_hf() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "hybrid": {"semanticRatio": 0.8}, "showRankingScore": false}), json!({"q": "Captain", "hybrid": {"embedder": "default", "semanticRatio": 0.8}, "showRankingScore": false}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -215,7 +218,7 @@ async fn simple_search_hf() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Movie World", "hybrid": {"semanticRatio": 0.2}, "showRankingScore": false}), json!({"q": "Movie World", "hybrid": {"embedder": "default", "semanticRatio": 0.2}, "showRankingScore": false}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -224,7 +227,7 @@ async fn simple_search_hf() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Wonder replacement", "hybrid": {"semanticRatio": 0.2}, "showRankingScore": false}), json!({"q": "Wonder replacement", "hybrid": {"embedder": "default", "semanticRatio": 0.2}, "showRankingScore": false}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -237,7 +240,7 @@ async fn distribution_shift() {
let server = Server::new().await; let server = Server::new().await;
let index = index_with_documents_user_provided(&server, &SIMPLE_SEARCH_DOCUMENTS_VEC).await; let index = index_with_documents_user_provided(&server, &SIMPLE_SEARCH_DOCUMENTS_VEC).await;
let search = json!({"q": "Captain", "vector": [1.0, 1.0], "showRankingScore": true, "hybrid": {"semanticRatio": 1.0}, "retrieveVectors": true}); let search = json!({"q": "Captain", "vector": [1.0, 1.0], "showRankingScore": true, "hybrid": {"embedder": "default", "semanticRatio": 1.0}, "retrieveVectors": true});
let (response, code) = index.search_post(search.clone()).await; let (response, code) = index.search_post(search.clone()).await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
snapshot!(response["hits"], @r###"[{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3","_vectors":{"default":{"embeddings":[[2.0,3.0]],"regenerate":false}},"_rankingScore":0.990290343761444},{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":{"embeddings":[[1.0,2.0]],"regenerate":false}},"_rankingScore":0.974341630935669},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1","_vectors":{"default":{"embeddings":[[1.0,3.0]],"regenerate":false}},"_rankingScore":0.9472135901451112}]"###); snapshot!(response["hits"], @r###"[{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3","_vectors":{"default":{"embeddings":[[2.0,3.0]],"regenerate":false}},"_rankingScore":0.990290343761444},{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":{"embeddings":[[1.0,2.0]],"regenerate":false}},"_rankingScore":0.974341630935669},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1","_vectors":{"default":{"embeddings":[[1.0,3.0]],"regenerate":false}},"_rankingScore":0.9472135901451112}]"###);
@ -271,7 +274,7 @@ async fn highlighter() {
let (response, code) = index let (response, code) = index
.search_post(json!({"q": "Captain Marvel", "vector": [1.0, 1.0], .search_post(json!({"q": "Captain Marvel", "vector": [1.0, 1.0],
"hybrid": {"semanticRatio": 0.2}, "hybrid": {"embedder": "default", "semanticRatio": 0.2},
"retrieveVectors": true, "retrieveVectors": true,
"attributesToHighlight": [ "attributesToHighlight": [
"desc", "desc",
@ -287,7 +290,7 @@ async fn highlighter() {
let (response, code) = index let (response, code) = index
.search_post(json!({"q": "Captain Marvel", "vector": [1.0, 1.0], .search_post(json!({"q": "Captain Marvel", "vector": [1.0, 1.0],
"hybrid": {"semanticRatio": 0.8}, "hybrid": {"embedder": "default", "semanticRatio": 0.8},
"retrieveVectors": true, "retrieveVectors": true,
"showRankingScore": true, "showRankingScore": true,
"attributesToHighlight": [ "attributesToHighlight": [
@ -304,7 +307,7 @@ async fn highlighter() {
// no highlighting on full semantic // no highlighting on full semantic
let (response, code) = index let (response, code) = index
.search_post(json!({"q": "Captain Marvel", "vector": [1.0, 1.0], .search_post(json!({"q": "Captain Marvel", "vector": [1.0, 1.0],
"hybrid": {"semanticRatio": 1.0}, "hybrid": {"embedder": "default", "semanticRatio": 1.0},
"retrieveVectors": true, "retrieveVectors": true,
"showRankingScore": true, "showRankingScore": true,
"attributesToHighlight": [ "attributesToHighlight": [
@ -326,7 +329,7 @@ async fn invalid_semantic_ratio() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": 1.2}}), json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"embedder": "default", "semanticRatio": 1.2}}),
) )
.await; .await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
@ -341,7 +344,7 @@ async fn invalid_semantic_ratio() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"semanticRatio": -0.8}}), json!({"q": "Captain", "vector": [1.0, 1.0], "hybrid": {"embedder": "default", "semanticRatio": -0.8}}),
) )
.await; .await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
@ -357,7 +360,7 @@ async fn invalid_semantic_ratio() {
let (response, code) = index let (response, code) = index
.search_get( .search_get(
&yaup::to_string( &yaup::to_string(
&json!({"q": "Captain", "vector": [1.0, 1.0], "hybridSemanticRatio": 1.2}), &json!({"q": "Captain", "vector": [1.0, 1.0], "hybridEmbedder": "default", "hybridSemanticRatio": 1.2}),
) )
.unwrap(), .unwrap(),
) )
@ -375,7 +378,7 @@ async fn invalid_semantic_ratio() {
let (response, code) = index let (response, code) = index
.search_get( .search_get(
&yaup::to_string( &yaup::to_string(
&json!({"q": "Captain", "vector": [1.0, 1.0], "hybridSemanticRatio": -0.2}), &json!({"q": "Captain", "vector": [1.0, 1.0], "hybridEmbedder": "default", "hybridSemanticRatio": -0.2}),
) )
.unwrap(), .unwrap(),
) )
@ -398,7 +401,7 @@ async fn single_document() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"vector": [1.0, 3.0], "hybrid": {"semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true}), json!({"vector": [1.0, 3.0], "hybrid": {"semanticRatio": 1.0, "embedder": "default"}, "showRankingScore": true, "retrieveVectors": true}),
) )
.await; .await;
@ -414,7 +417,7 @@ async fn query_combination() {
// search without query and vector, but with hybrid => still placeholder // search without query and vector, but with hybrid => still placeholder
let (response, code) = index let (response, code) = index
.search_post(json!({"hybrid": {"semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true})) .search_post(json!({"hybrid": {"embedder": "default", "semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true}))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -423,7 +426,7 @@ async fn query_combination() {
// same with a different semantic ratio // same with a different semantic ratio
let (response, code) = index let (response, code) = index
.search_post(json!({"hybrid": {"semanticRatio": 0.76}, "showRankingScore": true, "retrieveVectors": true})) .search_post(json!({"hybrid": {"embedder": "default", "semanticRatio": 0.76}, "showRankingScore": true, "retrieveVectors": true}))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -432,7 +435,7 @@ async fn query_combination() {
// wrong vector dimensions // wrong vector dimensions
let (response, code) = index let (response, code) = index
.search_post(json!({"vector": [1.0, 0.0, 1.0], "hybrid": {"semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true})) .search_post(json!({"vector": [1.0, 0.0, 1.0], "hybrid": {"embedder": "default", "semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true}))
.await; .await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
@ -447,7 +450,7 @@ async fn query_combination() {
// full vector // full vector
let (response, code) = index let (response, code) = index
.search_post(json!({"vector": [1.0, 0.0], "hybrid": {"semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true})) .search_post(json!({"vector": [1.0, 0.0], "hybrid": {"embedder": "default", "semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true}))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -456,7 +459,7 @@ async fn query_combination() {
// full keyword, without a query // full keyword, without a query
let (response, code) = index let (response, code) = index
.search_post(json!({"vector": [1.0, 0.0], "hybrid": {"semanticRatio": 0.0}, "showRankingScore": true, "retrieveVectors": true})) .search_post(json!({"vector": [1.0, 0.0], "hybrid": {"embedder": "default", "semanticRatio": 0.0}, "showRankingScore": true, "retrieveVectors": true}))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -465,7 +468,7 @@ async fn query_combination() {
// query + vector, full keyword => keyword // query + vector, full keyword => keyword
let (response, code) = index let (response, code) = index
.search_post(json!({"q": "Captain", "vector": [1.0, 0.0], "hybrid": {"semanticRatio": 0.0}, "showRankingScore": true, "retrieveVectors": true})) .search_post(json!({"q": "Captain", "vector": [1.0, 0.0], "hybrid": {"embedder": "default", "semanticRatio": 0.0}, "showRankingScore": true, "retrieveVectors": true}))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -480,7 +483,7 @@ async fn query_combination() {
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(response, @r###" snapshot!(response, @r###"
{ {
"message": "Invalid request: missing `hybrid` parameter when both `q` and `vector` are present.", "message": "Invalid request: missing `hybrid` parameter when `vector` is present.",
"code": "missing_search_hybrid", "code": "missing_search_hybrid",
"type": "invalid_request", "type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#missing_search_hybrid" "link": "https://docs.meilisearch.com/errors#missing_search_hybrid"
@ -490,7 +493,7 @@ async fn query_combination() {
// full vector, without a vector => error // full vector, without a vector => error
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "hybrid": {"semanticRatio": 1.0}, "showRankingScore": true, "retrieveVectors": true}), json!({"q": "Captain", "hybrid": {"semanticRatio": 1.0, "embedder": "default"}, "showRankingScore": true, "retrieveVectors": true}),
) )
.await; .await;
@ -507,7 +510,7 @@ async fn query_combination() {
// hybrid without a vector => full keyword // hybrid without a vector => full keyword
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Planet", "hybrid": {"semanticRatio": 0.99}, "showRankingScore": true, "retrieveVectors": true}), json!({"q": "Planet", "hybrid": {"semanticRatio": 0.99, "embedder": "default"}, "showRankingScore": true, "retrieveVectors": true}),
) )
.await; .await;
@ -523,7 +526,7 @@ async fn retrieve_vectors() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "hybrid": {"semanticRatio": 0.2}, "retrieveVectors": true}), json!({"q": "Captain", "hybrid": {"embedder": "default", "semanticRatio": 0.2}, "retrieveVectors": true}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -573,7 +576,7 @@ async fn retrieve_vectors() {
let (response, code) = index let (response, code) = index
.search_post( .search_post(
json!({"q": "Captain", "hybrid": {"semanticRatio": 0.2}, "retrieveVectors": true}), json!({"q": "Captain", "hybrid": {"embedder": "default", "semanticRatio": 0.2}, "retrieveVectors": true}),
) )
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");

35
meilisearch/tests/search/mod.rs
View File

@ -1099,22 +1099,28 @@ async fn experimental_feature_vector_store() {
index.add_documents(json!(documents), None).await; index.add_documents(json!(documents), None).await;
index.wait_task(0).await; index.wait_task(0).await;
index let (response, code) = index
.search(json!({ .search_post(json!({
"vector": [1.0, 2.0, 3.0], "vector": [1.0, 2.0, 3.0],
"hybrid": {
"embedder": "manual",
},
"showRankingScore": true "showRankingScore": true
}), |response, code|{ }))
meili_snap::snapshot!(code, @"400 Bad Request");
meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
{
"message": "Passing `vector` as a parameter requires enabling the `vector store` experimental feature. See https://github.com/meilisearch/product/discussions/677",
"code": "feature_not_enabled",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#feature_not_enabled"
}
"###);
})
.await; .await;
{
meili_snap::snapshot!(code, @"400 Bad Request");
meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
{
"message": "Passing `vector` as a parameter requires enabling the `vector store` experimental feature. See https://github.com/meilisearch/product/discussions/677",
"code": "feature_not_enabled",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#feature_not_enabled"
}
"###);
}
index index
.search(json!({ .search(json!({
"retrieveVectors": true, "retrieveVectors": true,
@ -1162,6 +1168,9 @@ async fn experimental_feature_vector_store() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"vector": [1.0, 2.0, 3.0], "vector": [1.0, 2.0, 3.0],
"hybrid": {
"embedder": "manual",
},
"showRankingScore": true, "showRankingScore": true,
"retrieveVectors": true, "retrieveVectors": true,
})) }))

120
meilisearch/tests/similar/errors.rs
View File

@ -18,7 +18,7 @@ async fn similar_unexisting_index() {
}); });
index index
.similar(json!({"id": 287947}), |response, code| { .similar(json!({"id": 287947, "embedder": "manual"}), |response, code| {
assert_eq!(code, 404); assert_eq!(code, 404);
assert_eq!(response, expected_response); assert_eq!(response, expected_response);
}) })
@ -44,7 +44,7 @@ async fn similar_feature_not_enabled() {
let server = Server::new().await; let server = Server::new().await;
let index = server.index("test"); let index = server.index("test");
let (response, code) = index.similar_post(json!({"id": 287947})).await; let (response, code) = index.similar_post(json!({"id": 287947, "embedder": "manual"})).await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -199,7 +199,8 @@ async fn similar_not_found_id() {
snapshot!(code, @"202 Accepted"); snapshot!(code, @"202 Accepted");
server.wait_task(response.uid()).await; server.wait_task(response.uid()).await;
let (response, code) = index.similar_post(json!({"id": "definitely-doesnt-exist"})).await; let (response, code) =
index.similar_post(json!({"id": "definitely-doesnt-exist", "embedder": "manual"})).await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -230,7 +231,8 @@ async fn similar_bad_offset() {
snapshot!(code, @"202 Accepted"); snapshot!(code, @"202 Accepted");
server.wait_task(response.uid()).await; server.wait_task(response.uid()).await;
let (response, code) = index.similar_post(json!({"id": 287947, "offset": "doggo"})).await; let (response, code) =
index.similar_post(json!({"id": 287947, "offset": "doggo", "embedder": "manual"})).await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -241,7 +243,7 @@ async fn similar_bad_offset() {
} }
"###); "###);
let (response, code) = index.similar_get("?id=287947&offset=doggo").await; let (response, code) = index.similar_get("?id=287947&offset=doggo&embedder=manual").await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -272,7 +274,8 @@ async fn similar_bad_limit() {
snapshot!(code, @"202 Accepted"); snapshot!(code, @"202 Accepted");
server.wait_task(response.uid()).await; server.wait_task(response.uid()).await;
let (response, code) = index.similar_post(json!({"id": 287947, "limit": "doggo"})).await; let (response, code) =
index.similar_post(json!({"id": 287947, "limit": "doggo", "embedder": "manual"})).await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -283,7 +286,7 @@ async fn similar_bad_limit() {
} }
"###); "###);
let (response, code) = index.similar_get("?id=287946&limit=doggo").await; let (response, code) = index.similar_get("?id=287946&limit=doggo&embedder=manual").await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -323,7 +326,8 @@ async fn similar_bad_filter() {
snapshot!(code, @"202 Accepted"); snapshot!(code, @"202 Accepted");
index.wait_task(value.uid()).await; index.wait_task(value.uid()).await;
let (response, code) = index.similar_post(json!({ "id": 287947, "filter": true })).await; let (response, code) =
index.similar_post(json!({ "id": 287947, "filter": true, "embedder": "manual" })).await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -361,7 +365,7 @@ async fn filter_invalid_syntax_object() {
index.wait_task(value.uid()).await; index.wait_task(value.uid()).await;
index index
.similar(json!({"id": 287947, "filter": "title & Glass"}), |response, code| { .similar(json!({"id": 287947, "filter": "title & Glass", "embedder": "manual"}), |response, code| {
snapshot!(response, @r###" snapshot!(response, @r###"
{ {
"message": "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `title & Glass`.\n1:14 title & Glass", "message": "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `title & Glass`.\n1:14 title & Glass",
@ -400,7 +404,7 @@ async fn filter_invalid_syntax_array() {
index.wait_task(value.uid()).await; index.wait_task(value.uid()).await;
index index
.similar(json!({"id": 287947, "filter": ["title & Glass"]}), |response, code| { .similar(json!({"id": 287947, "filter": ["title & Glass"], "embedder": "manual"}), |response, code| {
snapshot!(response, @r###" snapshot!(response, @r###"
{ {
"message": "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `title & Glass`.\n1:14 title & Glass", "message": "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `CONTAINS`, `NOT CONTAINS`, `STARTS WITH`, `NOT STARTS WITH`, `_geoRadius`, or `_geoBoundingBox` at `title & Glass`.\n1:14 title & Glass",
@ -446,7 +450,7 @@ async fn filter_invalid_syntax_string() {
}); });
index index
.similar( .similar(
json!({"id": 287947, "filter": "title = Glass XOR title = Glass"}), json!({"id": 287947, "filter": "title = Glass XOR title = Glass", "embedder": "manual"}),
|response, code| { |response, code| {
assert_eq!(response, expected_response); assert_eq!(response, expected_response);
assert_eq!(code, 400); assert_eq!(code, 400);
@ -486,10 +490,13 @@ async fn filter_invalid_attribute_array() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": ["many = Glass"]}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": ["many = Glass"], "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -524,10 +531,13 @@ async fn filter_invalid_attribute_string() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": "many = Glass"}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": "many = Glass", "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -562,10 +572,13 @@ async fn filter_reserved_geo_attribute_array() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": ["_geo = Glass"]}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": ["_geo = Glass"], "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -600,10 +613,13 @@ async fn filter_reserved_geo_attribute_string() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": "_geo = Glass"}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": "_geo = Glass", "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -638,10 +654,13 @@ async fn filter_reserved_attribute_array() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": ["_geoDistance = Glass"]}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": ["_geoDistance = Glass"], "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -676,10 +695,13 @@ async fn filter_reserved_attribute_string() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": "_geoDistance = Glass"}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": "_geoDistance = Glass", "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -714,10 +736,13 @@ async fn filter_reserved_geo_point_array() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": ["_geoPoint = Glass"]}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": ["_geoPoint = Glass"], "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -752,10 +777,13 @@ async fn filter_reserved_geo_point_string() {
"link": "https://docs.meilisearch.com/errors#invalid_similar_filter" "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
}); });
index index
.similar(json!({"id": 287947, "filter": "_geoPoint = Glass"}), |response, code| { .similar(
assert_eq!(response, expected_response); json!({"id": 287947, "filter": "_geoPoint = Glass", "embedder": "manual"}),
assert_eq!(code, 400); |response, code| {
}) assert_eq!(response, expected_response);
assert_eq!(code, 400);
},
)
.await; .await;
} }
@ -765,7 +793,8 @@ async fn similar_bad_retrieve_vectors() {
server.set_features(json!({"vectorStore": true})).await; server.set_features(json!({"vectorStore": true})).await;
let index = server.index("test"); let index = server.index("test");
let (response, code) = index.similar_post(json!({"retrieveVectors": "doggo"})).await; let (response, code) =
index.similar_post(json!({"retrieveVectors": "doggo", "embedder": "manual"})).await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {
@ -776,7 +805,8 @@ async fn similar_bad_retrieve_vectors() {
} }
"###); "###);
let (response, code) = index.similar_post(json!({"retrieveVectors": [true]})).await; let (response, code) =
index.similar_post(json!({"retrieveVectors": [true], "embedder": "manual"})).await;
snapshot!(code, @"400 Bad Request"); snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###" snapshot!(json_string!(response), @r###"
{ {

49
meilisearch/tests/similar/mod.rs
View File

@ -80,9 +80,11 @@ async fn basic() {
index.wait_task(value.uid()).await; index.wait_task(value.uid()).await;
index index
.similar(json!({"id": 143, "retrieveVectors": true}), |response, code| { .similar(
snapshot!(code, @"200 OK"); json!({"id": 143, "retrieveVectors": true, "embedder": "manual"}),
snapshot!(json_string!(response["hits"]), @r###" |response, code| {
snapshot!(code, @"200 OK");
snapshot!(json_string!(response["hits"]), @r###"
[ [
{ {
"title": "Escape Room", "title": "Escape Room",
@ -154,13 +156,16 @@ async fn basic() {
} }
] ]
"###); "###);
}) },
)
.await; .await;
index index
.similar(json!({"id": "299537", "retrieveVectors": true}), |response, code| { .similar(
snapshot!(code, @"200 OK"); json!({"id": "299537", "retrieveVectors": true, "embedder": "manual"}),
snapshot!(json_string!(response["hits"]), @r###" |response, code| {
snapshot!(code, @"200 OK");
snapshot!(json_string!(response["hits"]), @r###"
[ [
{ {
"title": "How to Train Your Dragon: The Hidden World", "title": "How to Train Your Dragon: The Hidden World",
@ -232,7 +237,8 @@ async fn basic() {
} }
] ]
"###); "###);
}) },
)
.await; .await;
} }
@ -272,7 +278,7 @@ async fn ranking_score_threshold() {
index index
.similar( .similar(
json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0, "retrieveVectors": true}), json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0, "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"4"); meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"4");
@ -358,7 +364,7 @@ async fn ranking_score_threshold() {
index index
.similar( .similar(
json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.2, "retrieveVectors": true}), json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.2, "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"3"); meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"3");
@ -426,7 +432,7 @@ async fn ranking_score_threshold() {
index index
.similar( .similar(
json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.3, "retrieveVectors": true}), json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.3, "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"2"); meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"2");
@ -476,7 +482,7 @@ async fn ranking_score_threshold() {
index index
.similar( .similar(
json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.6, "retrieveVectors": true}), json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.6, "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"1"); meili_snap::snapshot!(meili_snap::json_string!(response["estimatedTotalHits"]), @"1");
@ -508,7 +514,7 @@ async fn ranking_score_threshold() {
index index
.similar( .similar(
json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.9, "retrieveVectors": true}), json!({"id": 143, "showRankingScore": true, "rankingScoreThreshold": 0.9, "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
snapshot!(json_string!(response["hits"]), @"[]"); snapshot!(json_string!(response["hits"]), @"[]");
@ -553,7 +559,7 @@ async fn filter() {
index index
.similar( .similar(
json!({"id": 522681, "filter": "release_year = 2019", "retrieveVectors": true}), json!({"id": 522681, "filter": "release_year = 2019", "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
snapshot!(json_string!(response["hits"]), @r###" snapshot!(json_string!(response["hits"]), @r###"
@ -617,7 +623,7 @@ async fn filter() {
index index
.similar( .similar(
json!({"id": 522681, "filter": "release_year < 2000", "retrieveVectors": true}), json!({"id": 522681, "filter": "release_year < 2000", "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
snapshot!(json_string!(response["hits"]), @r###" snapshot!(json_string!(response["hits"]), @r###"
@ -681,9 +687,11 @@ async fn limit_and_offset() {
index.wait_task(value.uid()).await; index.wait_task(value.uid()).await;
index index
.similar(json!({"id": 143, "limit": 1, "retrieveVectors": true}), |response, code| { .similar(
snapshot!(code, @"200 OK"); json!({"id": 143, "limit": 1, "retrieveVectors": true, "embedder": "manual"}),
snapshot!(json_string!(response["hits"]), @r###" |response, code| {
snapshot!(code, @"200 OK");
snapshot!(json_string!(response["hits"]), @r###"
[ [
{ {
"title": "Escape Room", "title": "Escape Room",
@ -704,12 +712,13 @@ async fn limit_and_offset() {
} }
] ]
"###); "###);
}) },
)
.await; .await;
index index
.similar( .similar(
json!({"id": 143, "limit": 1, "offset": 1, "retrieveVectors": true}), json!({"id": 143, "limit": 1, "offset": 1, "retrieveVectors": true, "embedder": "manual"}),
|response, code| { |response, code| {
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
snapshot!(json_string!(response["hits"]), @r###" snapshot!(json_string!(response["hits"]), @r###"

9
meilisearch/tests/vector/mod.rs
View File

@ -624,7 +624,8 @@ async fn clear_documents() {
"###); "###);
// Make sure the arroy DB has been cleared // Make sure the arroy DB has been cleared
let (documents, _code) = index.search_post(json!({ "vector": [1, 1, 1] })).await; let (documents, _code) =
index.search_post(json!({ "vector": [1, 1, 1], "hybrid": {"embedder": "manual"} })).await;
snapshot!(documents, @r###" snapshot!(documents, @r###"
{ {
"hits": [], "hits": [],
@ -685,7 +686,11 @@ async fn add_remove_one_vector_4588() {
let task = index.wait_task(value.uid()).await; let task = index.wait_task(value.uid()).await;
snapshot!(task, name: "document-deleted"); snapshot!(task, name: "document-deleted");
let (documents, _code) = index.search_post(json!({"vector": [1, 1, 1] })).await; let (documents, _code) = index
.search_post(
json!({"vector": [1, 1, 1], "hybrid": {"semanticRatio": 1.0, "embedder": "manual"} }),
)
.await;
snapshot!(documents, @r###" snapshot!(documents, @r###"
{ {
"hits": [ "hits": [

32
meilisearch/tests/vector/openai.rs
View File

@ -449,7 +449,7 @@ async fn it_works() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "chien de chasse", "q": "chien de chasse",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"},
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -489,7 +489,7 @@ async fn it_works() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "petit chien", "q": "petit chien",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -529,7 +529,7 @@ async fn it_works() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "grand chien de berger des montagnes", "q": "grand chien de berger des montagnes",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -616,7 +616,7 @@ async fn tokenize_long_text() {
"q": "grand chien de berger des montagnes", "q": "grand chien de berger des montagnes",
"showRankingScore": true, "showRankingScore": true,
"attributesToRetrieve": ["id"], "attributesToRetrieve": ["id"],
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1064,7 +1064,7 @@ async fn smaller_dimensions() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "chien de chasse", "q": "chien de chasse",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1104,7 +1104,7 @@ async fn smaller_dimensions() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "petit chien", "q": "petit chien",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1144,7 +1144,7 @@ async fn smaller_dimensions() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "grand chien de berger des montagnes", "q": "grand chien de berger des montagnes",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1295,7 +1295,7 @@ async fn small_embedding_model() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "chien de chasse", "q": "chien de chasse",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1335,7 +1335,7 @@ async fn small_embedding_model() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "petit chien", "q": "petit chien",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1375,7 +1375,7 @@ async fn small_embedding_model() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "grand chien de berger des montagnes", "q": "grand chien de berger des montagnes",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1525,7 +1525,7 @@ async fn legacy_embedding_model() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "chien de chasse", "q": "chien de chasse",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1565,7 +1565,7 @@ async fn legacy_embedding_model() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "petit chien", "q": "petit chien",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1605,7 +1605,7 @@ async fn legacy_embedding_model() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "grand chien de berger des montagnes", "q": "grand chien de berger des montagnes",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1756,7 +1756,7 @@ async fn it_still_works() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "chien de chasse", "q": "chien de chasse",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1796,7 +1796,7 @@ async fn it_still_works() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "petit chien", "q": "petit chien",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");
@ -1836,7 +1836,7 @@ async fn it_still_works() {
let (response, code) = index let (response, code) = index
.search_post(json!({ .search_post(json!({
"q": "grand chien de berger des montagnes", "q": "grand chien de berger des montagnes",
"hybrid": {"semanticRatio": 1.0} "hybrid": {"semanticRatio": 1.0, "embedder": "default"}
})) }))
.await; .await;
snapshot!(code, @"200 OK"); snapshot!(code, @"200 OK");

3
meilisearch/tests/vector/settings.rs
View File

@ -218,7 +218,8 @@ async fn reset_embedder_documents() {
"###); "###);
// Make sure the arroy DB has been cleared // Make sure the arroy DB has been cleared
let (documents, _code) = index.search_post(json!({ "vector": [1, 1, 1] })).await; let (documents, _code) =
index.search_post(json!({ "vector": [1, 1, 1], "hybrid": {"embedder": "default"} })).await;
snapshot!(json_string!(documents), @r###" snapshot!(json_string!(documents), @r###"
{ {
"message": "Cannot find embedder with name `default`.", "message": "Cannot find embedder with name `default`.",

23
milli/src/vector/mod.rs
View File

@ -144,11 +144,6 @@ impl EmbeddingConfigs {
self.0.get(name).cloned() self.0.get(name).cloned()
} }
/// Get the default embedder configuration, if any.
pub fn get_default(&self) -> Option<(Arc<Embedder>, Arc<Prompt>)> {
self.get(self.get_default_embedder_name())
}
pub fn inner_as_ref(&self) -> &HashMap<String, (Arc<Embedder>, Arc<Prompt>)> { pub fn inner_as_ref(&self) -> &HashMap<String, (Arc<Embedder>, Arc<Prompt>)> {
&self.0 &self.0
} }
@ -156,24 +151,6 @@ impl EmbeddingConfigs {
pub fn into_inner(self) -> HashMap<String, (Arc<Embedder>, Arc<Prompt>)> { pub fn into_inner(self) -> HashMap<String, (Arc<Embedder>, Arc<Prompt>)> {
self.0 self.0
} }
/// Get the name of the default embedder configuration.
///
/// The default embedder is determined as follows:
///
/// - If there is only one embedder, it is always the default.
/// - If there are multiple embedders and one of them is called `default`, then that one is the default embedder.
/// - In all other cases, there is no default embedder.
pub fn get_default_embedder_name(&self) -> &str {
let mut it = self.0.keys();
let first_name = it.next();
let second_name = it.next();
match (first_name, second_name) {
(None, _) => "default",
(Some(first), None) => first,
(Some(_), Some(_)) => "default",
}
}
} }
impl IntoIterator for EmbeddingConfigs { impl IntoIterator for EmbeddingConfigs {