Merge pull request #674 from JiaYingZhang/similar-support

Add support for similar docs query

Author: Mubelotix

.github/workflows/tests.yml

@@ -85,34 +85,21 @@ jobs:
           config_file: .yamllint.yml
 
   coverage:
-    # Will not run if the actor is Dependabot (dependabot PRs)
-    # Will not run if the event is a PR to bump-meilisearch-v* (so a pre-release PR)
     if: github.actor != 'dependabot[bot]' && !( github.event_name == 'pull_request' && startsWith(github.base_ref, 'bump-meilisearch-v') )
     runs-on: ubuntu-latest
     needs: integration_tests
     name: Code Coverage
     steps:
       - uses: actions/checkout@v4
-      # Nightly Rust is used for cargo llvm-cov --doc below.
-      - uses: dtolnay/rust-toolchain@nightly
-        with:
-          components: llvm-tools-preview
-      - name: Install cargo-llvm-cov
-        uses: taiki-e/install-action@v2
-        with:
-          tool: cargo-llvm-cov
       - name: Meilisearch (latest version) setup with Docker
         run: docker run -d -p 7700:7700 getmeili/meilisearch:latest meilisearch --no-analytics --master-key=masterKey
-      - name: Collect coverage data
-        # Generate separate reports for tests and doctests, and combine them.
-        run: |
-          set -euo pipefail
-          cargo llvm-cov --no-report --all-features --workspace --exclude 'meilisearch-test-macro'
-          cargo llvm-cov --no-report --doc --all-features --workspace --exclude 'meilisearch-test-macro'
-          cargo llvm-cov report --doctests --codecov --output-path codecov.json
-      - name: Upload coverage reports to Codecov
+      - name: Install cargo-llvm-cov
+        uses: taiki-e/install-action@cargo-llvm-cov
+      - name: Generate code coverage
+        run: cargo llvm-cov --all-features --workspace --lcov --output-path lcov.info
+      - name: Upload coverage to Codecov
         uses: codecov/codecov-action@v5
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
-          files: codecov.json
+          files: lcov.info
           fail_ci_if_error: true

src/documents.rs

@@ -193,7 +193,7 @@ pub struct DocumentsQuery<'a, Http: HttpClient> {
     /// Filters to apply.
     ///
     /// Available since v1.2 of Meilisearch
-    /// Read the [dedicated guide](https://www.meilisearch.com/docs/learn/fine_tuning_results/filtering#filter-basics) to learn the syntax.
+    /// Read the [dedicated guide](https://www.meilisearch.com/docs/learn/filtering_and_sorting) to learn the syntax.
     #[serde(skip_serializing_if = "Option::is_none")]
     pub filter: Option<&'a str>,
 }

src/indexes.rs

@@ -4,6 +4,7 @@ use crate::{
     errors::{Error, MeilisearchCommunicationError, MeilisearchError, MEILISEARCH_VERSION_HINT},
     request::*,
     search::*,
+    similar::*,
     task_info::TaskInfo,
     tasks::*,
     DefaultHttpClient,
@@ -1673,6 +1674,47 @@ impl<Http: HttpClient> Index<Http> {
         }
         Ok(task)
     }
+
+    /// Get similar documents in the index.
+    ///
+    /// # Example
+    ///
+    /// ```no_run
+    /// # use serde::{Serialize, Deserialize};
+    /// # use meilisearch_sdk::{client::*, indexes::*, similar::*};
+    /// #
+    /// # let MEILISEARCH_URL = option_env!("MEILISEARCH_URL").unwrap_or("http://localhost:7700");
+    /// # let MEILISEARCH_API_KEY = option_env!("MEILISEARCH_API_KEY").unwrap_or("masterKey");
+    /// #
+    /// # #[derive(Serialize, Deserialize, Debug)]
+    /// # struct Movie {
+    /// #    name: String,
+    /// #    description: String,
+    /// # }
+    /// # tokio::runtime::Builder::new_current_thread().enable_all().build().unwrap().block_on(async {
+    /// # let client = Client::new(MEILISEARCH_URL, Some(MEILISEARCH_API_KEY)).unwrap();
+    /// # let movies = client.index("similar_query");
+    /// #
+    /// let query = SimilarQuery::new(&movies, "1", "default");
+    /// let results = movies.similar_query::<Movie>(&query).await.unwrap();
+    ///
+    /// assert!(results.hits.len() > 0);
+    /// # movies.delete().await.unwrap().wait_for_completion(&client, None, None).await.unwrap();
+    /// # });
+    /// ```
+    pub async fn similar_query<T: 'static + DeserializeOwned + Send + Sync>(
+        &self,
+        body: &SimilarQuery<'_, Http>,
+    ) -> Result<SimilarResults<T>, Error> {
+        self.client
+            .http_client
+            .request::<(), &SimilarQuery<Http>, SimilarResults<T>>(
+                &format!("{}/indexes/{}/similar", self.client.host, self.uid),
+                Method::Post { body, query: () },
+                200,
+            )
+            .await
+    }
 }
 
 impl<Http: HttpClient> AsRef<str> for Index<Http> {

src/lib.rs

@@ -249,6 +249,8 @@ pub mod request;
 pub mod search;
 /// Module containing [`Settings`](settings::Settings).
 pub mod settings;
+/// Module related to [similar queries](https://www.meilisearch.com/docs/learn/ai_powered_search/retrieve_related_search_results#return-similar-documents).
+pub mod similar;
 /// Module containing the [snapshots](snapshots::create_snapshot)-feature.
 pub mod snapshots;
 /// Module representing the [`TaskInfo`](task_info::TaskInfo)s.

src/search.rs

@@ -123,17 +123,6 @@ pub struct SearchResults<T> {
     pub index_uid: Option<String>,
 }
 
-fn serialize_with_wildcard<S: Serializer, T: Serialize>(
-    data: &Option<Selectors<T>>,
-    s: S,
-) -> Result<S::Ok, S::Error> {
-    match data {
-        Some(Selectors::All) => ["*"].serialize(s),
-        Some(Selectors::Some(data)) => data.serialize(s),
-        None => s.serialize_none(),
-    }
-}
-
 fn serialize_attributes_to_crop_with_wildcard<S: Serializer>(
     data: &Option<Selectors<&[AttributeToCrop]>>,
     s: S,
@@ -169,6 +158,15 @@ pub enum Selectors<T> {
     All,
 }
 
+impl<T: Serialize> Serialize for Selectors<T> {
+    fn serialize<S: Serializer>(&self, s: S) -> Result<S::Ok, S::Error> {
+        match self {
+            Selectors::Some(data) => data.serialize(s),
+            Selectors::All => ["*"].serialize(s),
+        }
+    }
+}
+
 /// Configures Meilisearch to return search results based on a query’s meaning and context
 #[derive(Debug, Serialize, Clone)]
 #[serde(rename_all = "camelCase")]
@@ -282,7 +280,7 @@ pub struct SearchQuery<'a, Http: HttpClient> {
     pub hits_per_page: Option<usize>,
     /// Filter applied to documents.
     ///
-    /// Read the [dedicated guide](https://www.meilisearch.com/docs/learn/advanced/filtering) to learn the syntax.
+    /// Read the [dedicated guide](https://www.meilisearch.com/docs/learn/filtering_and_sorting) to learn the syntax.
     #[serde(skip_serializing_if = "Option::is_none")]
     pub filter: Option<Filter<'a>>,
     /// Facets for which to retrieve the matching count.
@@ -291,7 +289,6 @@ pub struct SearchQuery<'a, Http: HttpClient> {
     ///
     /// **Default: all attributes found in the documents.**
     #[serde(skip_serializing_if = "Option::is_none")]
-    #[serde(serialize_with = "serialize_with_wildcard")]
     pub facets: Option<Selectors<&'a [&'a str]>>,
     /// Attributes to sort.
     #[serde(skip_serializing_if = "Option::is_none")]
@@ -309,7 +306,6 @@ pub struct SearchQuery<'a, Http: HttpClient> {
     ///
     /// **Default: all attributes found in the documents.**
     #[serde(skip_serializing_if = "Option::is_none")]
-    #[serde(serialize_with = "serialize_with_wildcard")]
     pub attributes_to_retrieve: Option<Selectors<&'a [&'a str]>>,
     /// Attributes whose values have to be cropped.
     ///
@@ -337,7 +333,6 @@ pub struct SearchQuery<'a, Http: HttpClient> {
     ///
     /// Can be set to a [wildcard value](enum.Selectors.html#variant.All) that will select all existing attributes.
     #[serde(skip_serializing_if = "Option::is_none")]
-    #[serde(serialize_with = "serialize_with_wildcard")]
     pub attributes_to_highlight: Option<Selectors<&'a [&'a str]>>,
     /// Tag in front of a highlighted term.
     ///
@@ -971,7 +966,7 @@ pub struct FacetSearchQuery<'a, Http: HttpClient = DefaultHttpClient> {
     pub search_query: Option<&'a str>,
     /// Filter applied to documents.
     ///
-    /// Read the [dedicated guide](https://www.meilisearch.com/docs/learn/advanced/filtering) to learn the syntax.
+    /// Read the [dedicated guide](https://www.meilisearch.com/docs/learn/filtering_and_sorting) to learn the syntax.
     #[serde(skip_serializing_if = "Option::is_none")]
     pub filter: Option<Filter<'a>>,
     /// Defines the strategy on how to handle search queries containing multiple words.
@@ -1078,7 +1073,7 @@ pub struct FacetSearchResponse {
 }
 
 #[cfg(test)]
-mod tests {
+pub(crate) mod tests {
     use crate::{
         client::*,
         key::{Action, KeyBuilder},
@@ -1091,19 +1086,19 @@ mod tests {
     use serde_json::{json, Map, Value};
 
     #[derive(Debug, Serialize, Deserialize, PartialEq)]
-    struct Nested {
+    pub struct Nested {
         child: String,
     }
 
     #[derive(Debug, Serialize, Deserialize, PartialEq)]
-    struct Document {
-        id: usize,
-        value: String,
-        kind: String,
-        number: i32,
-        nested: Nested,
+    pub struct Document {
+        pub id: usize,
+        pub value: String,
+        pub kind: String,
+        pub number: i32,
+        pub nested: Nested,
         #[serde(skip_serializing_if = "Option::is_none", default)]
-        _vectors: Option<Vectors>,
+        pub _vectors: Option<Vectors>,
     }
 
     #[derive(Debug, Serialize, Deserialize, PartialEq)]
@@ -1120,7 +1115,7 @@ mod tests {
     }
 
     #[derive(Debug, Serialize, Deserialize, PartialEq)]
-    struct Vectors(HashMap<String, Vector>);
+    pub struct Vectors(HashMap<String, Vector>);
 
     impl<T: Into<Vec<f32>>> From<T> for Vectors {
         fn from(value: T) -> Self {
@@ -1151,7 +1146,7 @@ mod tests {
         vector
     }
 
-    async fn setup_test_index(client: &Client, index: &Index) -> Result<(), Error> {
+    pub(crate) async fn setup_test_index(client: &Client, index: &Index) -> Result<(), Error> {
         let t0 = index.add_documents(&[
             Document { id: 0, kind: "text".into(), number: 0, value: S("Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum."), nested: Nested { child: S("first") }, _vectors: Some(Vectors::from(vectorize(false, 0))) },
             Document { id: 1, kind: "text".into(), number: 10, value: S("dolor sit amet, consectetur adipiscing elit"), nested: Nested { child: S("second") }, _vectors: Some(Vectors::from(vectorize(false, 1))) },
@@ -1225,7 +1220,7 @@ mod tests {
         Ok(())
     }
 
-    async fn setup_hybrid_searching(client: &Client, index: &Index) -> Result<(), Error> {
+    pub(crate) async fn setup_embedder(client: &Client, index: &Index) -> Result<(), Error> {
         use crate::settings::Embedder;
         let embedder_setting = Embedder {
             source: EmbedderSource::UserProvided,
@@ -1998,7 +1993,7 @@ mod tests {
 
     #[meilisearch_test]
     async fn test_with_vectors(client: Client, index: Index) -> Result<(), Error> {
-        setup_hybrid_searching(&client, &index).await?;
+        setup_embedder(&client, &index).await?;
         setup_test_index(&client, &index).await?;
 
         let results: SearchResults<Document> = index
@@ -2024,7 +2019,7 @@ mod tests {
 
     #[meilisearch_test]
     async fn test_hybrid(client: Client, index: Index) -> Result<(), Error> {
-        setup_hybrid_searching(&client, &index).await?;
+        setup_embedder(&client, &index).await?;
         setup_test_index(&client, &index).await?;
 
         // Search for an Harry Potter but with lorem ipsum's id