Merge #196

196: Implements the synonyms in transplant r=MarinPostma a=irevoire

closes #18 

Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: marin postma <postma.marin@protonmail.com>

meilisearch-http/src/index/mod.rs

@@ -84,6 +84,19 @@ impl Index {
             .unwrap_or_else(BTreeSet::new);
         let distinct_field = self.distinct_field(&txn)?.map(String::from);
 
+        // in milli each word in the synonyms map were split on their separator. Since we lost
+        // this information we are going to put space between words.
+        let synonyms = self
+            .synonyms(&txn)?
+            .iter()
+            .map(|(key, values)| {
+                (
+                    key.join(" "),
+                    values.iter().map(|value| value.join(" ")).collect(),
+                )
+            })
+            .collect();
+
         Ok(Settings {
             displayed_attributes: Some(displayed_attributes),
             searchable_attributes: Some(searchable_attributes),
@@ -91,6 +104,7 @@ impl Index {
             ranking_rules: Some(Some(criteria)),
             stop_words: Some(Some(stop_words)),
             distinct_attribute: Some(distinct_field),
+            synonyms: Some(Some(synonyms)),
             _kind: PhantomData,
         })
     }

meilisearch-http/src/index/updates.rs

@@ -1,4 +1,4 @@
-use std::collections::{BTreeSet, HashSet};
+use std::collections::{BTreeSet, BTreeMap, HashSet};
 use std::io;
 use std::marker::PhantomData;
 use std::num::NonZeroUsize;
@@ -70,6 +70,12 @@ pub struct Settings<T> {
         deserialize_with = "deserialize_some",
         skip_serializing_if = "Option::is_none"
     )]
+    pub synonyms: Option<Option<BTreeMap<String, Vec<String>>>>,
+    #[serde(
+        default,
+        deserialize_with = "deserialize_some",
+        skip_serializing_if = "Option::is_none"
+    )]
     pub distinct_attribute: Option<Option<String>>,
 
     #[serde(skip)]
@@ -84,6 +90,7 @@ impl Settings<Checked> {
             filterable_attributes: Some(None),
             ranking_rules: Some(None),
             stop_words: Some(None),
+            synonyms: Some(None),
             distinct_attribute: Some(None),
             _kind: PhantomData,
         }
@@ -96,6 +103,7 @@ impl Settings<Checked> {
             filterable_attributes,
             ranking_rules,
             stop_words,
+            synonyms,
             distinct_attribute,
             ..
         } = self;
@@ -106,6 +114,7 @@ impl Settings<Checked> {
             filterable_attributes,
             ranking_rules,
             stop_words,
+            synonyms,
             distinct_attribute,
             _kind: PhantomData,
         }
@@ -142,6 +151,7 @@ impl Settings<Unchecked> {
             filterable_attributes: self.filterable_attributes,
             ranking_rules: self.ranking_rules,
             stop_words: self.stop_words,
+            synonyms: self.synonyms,
             distinct_attribute: self.distinct_attribute,
             _kind: PhantomData,
         }
@@ -267,7 +277,14 @@ impl Index {
         if let Some(ref stop_words) = settings.stop_words {
             match stop_words {
                 Some(stop_words) => builder.set_stop_words(stop_words.clone()),
-                _ => builder.reset_stop_words(),
+                None => builder.reset_stop_words(),
+            }
+        }
+
+        if let Some(ref synonyms) = settings.synonyms {
+            match synonyms {
+                Some(synonyms) => builder.set_synonyms(synonyms.clone().into_iter().collect()),
+                None => builder.reset_synonyms(),
             }
         }
 
@@ -332,6 +349,7 @@ mod test {
             filterable_attributes: None,
             ranking_rules: None,
             stop_words: None,
+            synonyms: None,
             distinct_attribute: None,
             _kind: PhantomData::<Unchecked>,
         };
@@ -351,6 +369,7 @@ mod test {
             filterable_attributes: None,
             ranking_rules: None,
             stop_words: None,
+            synonyms: None,
             distinct_attribute: None,
             _kind: PhantomData::<Unchecked>,
         };

meilisearch-http/src/index_controller/dump_actor/loaders/v1.rs

@@ -133,9 +133,6 @@ fn load_index(
 /// we need to **always** be able to convert the old settings to the settings currently being used
 impl From<Settings> for index_controller::Settings<Unchecked> {
     fn from(settings: Settings) -> Self {
-        if settings.synonyms.flatten().is_some() {
-            error!("`synonyms` are not yet implemented and thus will be ignored");
-        }
         Self {
             distinct_attribute: settings.distinct_attribute,
             // we need to convert the old `Vec<String>` into a `BTreeSet<String>`
@@ -167,6 +164,8 @@ impl From<Settings> for index_controller::Settings<Unchecked> {
                 }).collect())),
             // we need to convert the old `Vec<String>` into a `BTreeSet<String>`
             stop_words: settings.stop_words.map(|o| o.map(|vec| vec.into_iter().collect())),
+            // we need to convert the old `Vec<String>` into a `BTreeMap<String>`
+            synonyms: settings.synonyms.map(|o| o.map(|vec| vec.into_iter().collect())),
             _kind: PhantomData,
         }
     }

meilisearch-http/src/index_controller/updates.rs

@@ -14,6 +14,7 @@ pub enum UpdateResult {
     Other,
 }
 
+#[allow(clippy::large_enum_variant)]
 #[derive(Debug, Clone, Serialize, Deserialize)]
 #[serde(tag = "type")]
 pub enum UpdateMeta {

meilisearch-http/src/lib.rs

@@ -43,7 +43,6 @@ macro_rules! create_app {
             .configure(index::services)
             .configure(search::services)
             .configure(settings::services)
-            .configure(synonym::services)
             .configure(health::services)
             .configure(stats::services)
             .configure(key::services)

meilisearch-http/src/routes/mod.rs

@@ -15,9 +15,9 @@ pub mod key;
 pub mod search;
 pub mod settings;
 pub mod stats;
-pub mod synonym;
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
+#[allow(clippy::clippy::large_enum_variant)]
 #[serde(tag = "name")]
 pub enum UpdateType {
     ClearAll,

meilisearch-http/src/routes/settings/mod.rs

@@ -97,6 +97,12 @@ make_setting_route!(
     stop_words
 );
 
+make_setting_route!(
+    "/indexes/{index_uid}/settings/synonyms",
+    std::collections::BTreeMap<String, Vec<String>>,
+    synonyms
+);
+
 make_setting_route!(
     "/indexes/{index_uid}/settings/distinct-attribute",
     String,
@@ -131,6 +137,7 @@ create_services!(
     searchable_attributes,
     distinct_attribute,
     stop_words,
+    synonyms,
     ranking_rules
 );
 

meilisearch-http/src/routes/synonym.rs

@@ -1,47 +0,0 @@
-use std::collections::BTreeMap;
-
-use actix_web::{delete, get, post};
-use actix_web::{web, HttpResponse};
-
-use crate::error::ResponseError;
-use crate::helpers::Authentication;
-use crate::routes::IndexParam;
-use crate::Data;
-
-pub fn services(cfg: &mut web::ServiceConfig) {
-    cfg.service(get).service(update).service(delete);
-}
-
-#[get(
-    "/indexes/{index_uid}/settings/synonyms",
-    wrap = "Authentication::Private"
-)]
-async fn get(
-    _data: web::Data<Data>,
-    _path: web::Path<IndexParam>,
-) -> Result<HttpResponse, ResponseError> {
-    todo!()
-}
-
-#[post(
-    "/indexes/{index_uid}/settings/synonyms",
-    wrap = "Authentication::Private"
-)]
-async fn update(
-    _data: web::Data<Data>,
-    _path: web::Path<IndexParam>,
-    _body: web::Json<BTreeMap<String, Vec<String>>>,
-) -> Result<HttpResponse, ResponseError> {
-    todo!()
-}
-
-#[delete(
-    "/indexes/{index_uid}/settings/synonyms",
-    wrap = "Authentication::Private"
-)]
-async fn delete(
-    _data: web::Data<Data>,
-    _path: web::Path<IndexParam>,
-) -> Result<HttpResponse, ResponseError> {
-    todo!()
-}

meilisearch-http/tests/settings/get_settings.rs

@@ -16,7 +16,7 @@ async fn get_settings() {
     let (response, code) = index.settings().await;
     assert_eq!(code, 200);
     let settings = response.as_object().unwrap();
-    assert_eq!(settings.keys().len(), 6);
+    assert_eq!(settings.keys().len(), 7);
     assert_eq!(settings["displayedAttributes"], json!(["*"]));
     assert_eq!(settings["searchableAttributes"], json!(["*"]));
     assert_eq!(settings["filterableAttributes"], json!([]));
@@ -87,7 +87,7 @@ async fn reset_all_settings() {
     index.wait_update_id(0).await;
 
     index
-        .update_settings(json!({"displayedAttributes": ["name", "age"], "searchableAttributes": ["name"], "stopWords": ["the"], "filterableAttributes": ["age"] }))
+        .update_settings(json!({"displayedAttributes": ["name", "age"], "searchableAttributes": ["name"], "stopWords": ["the"], "filterableAttributes": ["age"], "synonyms": {"puppy": ["dog", "doggo", "potat"] }}))
         .await;
     index.wait_update_id(1).await;
     let (response, code) = index.settings().await;
@@ -95,6 +95,10 @@ async fn reset_all_settings() {
     assert_eq!(response["displayedAttributes"], json!(["name", "age"]));
     assert_eq!(response["searchableAttributes"], json!(["name"]));
     assert_eq!(response["stopWords"], json!(["the"]));
+    assert_eq!(
+        response["synonyms"],
+        json!({"puppy": ["dog", "doggo", "potat"] })
+    );
     assert_eq!(response["filterableAttributes"], json!(["age"]));
 
     index.delete_settings().await;
@@ -106,6 +110,7 @@ async fn reset_all_settings() {
     assert_eq!(response["searchableAttributes"], json!(["*"]));
     assert_eq!(response["stopWords"], json!([]));
     assert_eq!(response["filterableAttributes"], json!([]));
+    assert_eq!(response["synonyms"], json!({}));
 
     let (response, code) = index.get_document(1, None).await;
     assert_eq!(code, 200);
@@ -184,5 +189,6 @@ test_setting_routes!(
     filterable_attributes,
     displayed_attributes,
     searchable_attributes,
-    stop_words
+    stop_words,
+    synonyms
 );