Merge #5351

5351: Bring back v1.13.0 changes into main r=irevoire a=Kerollmops

This PR brings back the changes made in v1.13 into the main branch.

Co-authored-by: ManyTheFish <many@meilisearch.com>
Co-authored-by: Kerollmops <clement@meilisearch.com>
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
Co-authored-by: Clémentine <clementine@meilisearch.com>
Co-authored-by: meili-bors[bot] <89034592+meili-bors[bot]@users.noreply.github.com>
Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: Clément Renault <clement@meilisearch.com>

crates/meilisearch/src/routes/features.rs

@@ -50,6 +50,8 @@ pub fn configure(cfg: &mut web::ServiceConfig) {
             logs_route: Some(false),
             edit_documents_by_function: Some(false),
             contains_filter: Some(false),
+            network: Some(false),
+            get_task_documents_route: Some(false),
         })),
         (status = 401, description = "The authorization header is missing", body = ResponseError, content_type = "application/json", example = json!(
             {
@@ -88,6 +90,10 @@ pub struct RuntimeTogglableFeatures {
     pub edit_documents_by_function: Option<bool>,
     #[deserr(default)]
     pub contains_filter: Option<bool>,
+    #[deserr(default)]
+    pub network: Option<bool>,
+    #[deserr(default)]
+    pub get_task_documents_route: Option<bool>,
 }
 
 impl From<meilisearch_types::features::RuntimeTogglableFeatures> for RuntimeTogglableFeatures {
@@ -97,6 +103,8 @@ impl From<meilisearch_types::features::RuntimeTogglableFeatures> for RuntimeTogg
             logs_route,
             edit_documents_by_function,
             contains_filter,
+            network,
+            get_task_documents_route,
         } = value;
 
         Self {
@@ -104,6 +112,8 @@ impl From<meilisearch_types::features::RuntimeTogglableFeatures> for RuntimeTogg
             logs_route: Some(logs_route),
             edit_documents_by_function: Some(edit_documents_by_function),
             contains_filter: Some(contains_filter),
+            network: Some(network),
+            get_task_documents_route: Some(get_task_documents_route),
         }
     }
 }
@@ -114,6 +124,8 @@ pub struct PatchExperimentalFeatureAnalytics {
     logs_route: bool,
     edit_documents_by_function: bool,
     contains_filter: bool,
+    network: bool,
+    get_task_documents_route: bool,
 }
 
 impl Aggregate for PatchExperimentalFeatureAnalytics {
@@ -127,6 +139,8 @@ impl Aggregate for PatchExperimentalFeatureAnalytics {
             logs_route: new.logs_route,
             edit_documents_by_function: new.edit_documents_by_function,
             contains_filter: new.contains_filter,
+            network: new.network,
+            get_task_documents_route: new.get_task_documents_route,
         })
     }
 
@@ -149,6 +163,8 @@ impl Aggregate for PatchExperimentalFeatureAnalytics {
             logs_route: Some(false),
             edit_documents_by_function: Some(false),
             contains_filter: Some(false),
+            network: Some(false),
+            get_task_documents_route: Some(false),
          })),
         (status = 401, description = "The authorization header is missing", body = ResponseError, content_type = "application/json", example = json!(
             {
@@ -181,16 +197,23 @@ async fn patch_features(
             .edit_documents_by_function
             .unwrap_or(old_features.edit_documents_by_function),
         contains_filter: new_features.0.contains_filter.unwrap_or(old_features.contains_filter),
+        network: new_features.0.network.unwrap_or(old_features.network),
+        get_task_documents_route: new_features
+            .0
+            .get_task_documents_route
+            .unwrap_or(old_features.get_task_documents_route),
     };
 
     // explicitly destructure for analytics rather than using the `Serialize` implementation, because
-    // the it renames to camelCase, which we don't want for analytics.
+    // it renames to camelCase, which we don't want for analytics.
     // **Do not** ignore fields with `..` or `_` here, because we want to add them in the future.
     let meilisearch_types::features::RuntimeTogglableFeatures {
         metrics,
         logs_route,
         edit_documents_by_function,
         contains_filter,
+        network,
+        get_task_documents_route,
     } = new_features;
 
     analytics.publish(
@@ -199,6 +222,8 @@ async fn patch_features(
             logs_route,
             edit_documents_by_function,
             contains_filter,
+            network,
+            get_task_documents_route,
         },
         &req,
     );

crates/meilisearch/src/routes/indexes/mod.rs

@@ -496,6 +496,12 @@ pub struct IndexStats {
     pub number_of_documents: u64,
     /// Whether or not the index is currently ingesting document
     pub is_indexing: bool,
+    /// Number of embeddings in the index
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub number_of_embeddings: Option<u64>,
+    /// Number of embedded documents in the index
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub number_of_embedded_documents: Option<u64>,
     /// Association of every field name with the number of times it occurs in the documents.
     #[schema(value_type = HashMap<String, u64>)]
     pub field_distribution: FieldDistribution,
@@ -506,6 +512,8 @@ impl From<index_scheduler::IndexStats> for IndexStats {
         IndexStats {
             number_of_documents: stats.inner_stats.number_of_documents,
             is_indexing: stats.is_indexing,
+            number_of_embeddings: stats.inner_stats.number_of_embeddings,
+            number_of_embedded_documents: stats.inner_stats.number_of_embedded_documents,
             field_distribution: stats.inner_stats.field_distribution,
         }
     }
@@ -524,6 +532,8 @@ impl From<index_scheduler::IndexStats> for IndexStats {
         (status = OK, description = "The stats of the index", body = IndexStats, content_type = "application/json", example = json!(
             {
                 "numberOfDocuments": 10,
+                "numberOfEmbeddings": 10,
+                "numberOfEmbeddedDocuments": 10,
                 "isIndexing": true,
                 "fieldDistribution": {
                     "genre": 10,

crates/meilisearch/src/routes/mod.rs

@@ -34,6 +34,7 @@ use crate::routes::features::RuntimeTogglableFeatures;
 use crate::routes::indexes::documents::{DocumentDeletionByFilter, DocumentEditionByFunction};
 use crate::routes::indexes::IndexView;
 use crate::routes::multi_search::SearchResults;
+use crate::routes::network::{Network, Remote};
 use crate::routes::swap_indexes::SwapIndexesPayload;
 use crate::search::{
     FederatedSearch, FederatedSearchResult, Federation, FederationOptions, MergeFacets,
@@ -54,6 +55,7 @@ mod logs;
 mod metrics;
 mod multi_search;
 mod multi_search_analytics;
+pub mod network;
 mod open_api_utils;
 mod snapshot;
 mod swap_indexes;
@@ -75,6 +77,7 @@ pub mod tasks;
         (path = "/multi-search", api = multi_search::MultiSearchApi),
         (path = "/swap-indexes", api = swap_indexes::SwapIndexesApi),
         (path = "/experimental-features", api = features::ExperimentalFeaturesApi),
+        (path = "/network", api = network::NetworkApi),
     ),
     paths(get_health, get_version, get_stats),
     tags(
@@ -85,7 +88,7 @@ pub mod tasks;
         url = "/",
         description = "Local server",
     )),
-    components(schemas(PaginationView<KeyView>, PaginationView<IndexView>, IndexView, DocumentDeletionByFilter, AllBatches, BatchStats, ProgressStepView, ProgressView, BatchView, RuntimeTogglableFeatures, SwapIndexesPayload, DocumentEditionByFunction, MergeFacets, FederationOptions, SearchQueryWithIndex, Federation, FederatedSearch, FederatedSearchResult, SearchResults, SearchResultWithIndex, SimilarQuery, SimilarResult, PaginationView<serde_json::Value>, BrowseQuery, UpdateIndexRequest, IndexUid, IndexCreateRequest, KeyView, Action, CreateApiKey, UpdateStderrLogs, LogMode, GetLogs, IndexStats, Stats, HealthStatus, HealthResponse, VersionResponse, Code, ErrorType, AllTasks, TaskView, Status, DetailsView, ResponseError, Settings<Unchecked>, Settings<Checked>, TypoSettings, MinWordSizeTyposSetting, FacetingSettings, PaginationSettings, SummarizedTaskView, Kind))
+    components(schemas(PaginationView<KeyView>, PaginationView<IndexView>, IndexView, DocumentDeletionByFilter, AllBatches, BatchStats, ProgressStepView, ProgressView, BatchView, RuntimeTogglableFeatures, SwapIndexesPayload, DocumentEditionByFunction, MergeFacets, FederationOptions, SearchQueryWithIndex, Federation, FederatedSearch, FederatedSearchResult, SearchResults, SearchResultWithIndex, SimilarQuery, SimilarResult, PaginationView<serde_json::Value>, BrowseQuery, UpdateIndexRequest, IndexUid, IndexCreateRequest, KeyView, Action, CreateApiKey, UpdateStderrLogs, LogMode, GetLogs, IndexStats, Stats, HealthStatus, HealthResponse, VersionResponse, Code, ErrorType, AllTasks, TaskView, Status, DetailsView, ResponseError, Settings<Unchecked>, Settings<Checked>, TypoSettings, MinWordSizeTyposSetting, FacetingSettings, PaginationSettings, SummarizedTaskView, Kind, Network, Remote))
 )]
 pub struct MeilisearchApi;
 
@@ -103,7 +106,8 @@ pub fn configure(cfg: &mut web::ServiceConfig) {
         .service(web::scope("/multi-search").configure(multi_search::configure))
         .service(web::scope("/swap-indexes").configure(swap_indexes::configure))
         .service(web::scope("/metrics").configure(metrics::configure))
-        .service(web::scope("/experimental-features").configure(features::configure));
+        .service(web::scope("/experimental-features").configure(features::configure))
+        .service(web::scope("/network").configure(network::configure));
 
     #[cfg(feature = "swagger")]
     {
@@ -359,9 +363,9 @@ pub async fn running() -> HttpResponse {
 #[derive(Serialize, Debug, ToSchema)]
 #[serde(rename_all = "camelCase")]
 pub struct Stats {
-    /// The size of the database, in bytes.
+    /// The disk space used by the database, in bytes.
     pub database_size: u64,
-    #[serde(skip)]
+    /// The size of the database, in bytes.
     pub used_database_size: u64,
     /// The date of the last update in the RFC 3339 formats. Can be `null` if no update has ever been processed.
     #[serde(serialize_with = "time::serde::rfc3339::option::serialize")]
@@ -383,6 +387,7 @@ pub struct Stats {
         (status = 200, description = "The stats of the instance", body = Stats, content_type = "application/json", example = json!(
             {
                 "databaseSize": 567,
+                "usedDatabaseSize": 456,
                 "lastUpdate": "2019-11-20T09:40:33.711324Z",
                 "indexes": {
                     "movies": {

crates/meilisearch/src/routes/multi_search.rs

@@ -20,6 +20,7 @@ use crate::routes::indexes::search::search_kind;
 use crate::search::{
     add_search_rules, perform_federated_search, perform_search, FederatedSearch,
     FederatedSearchResult, RetrieveVectors, SearchQueryWithIndex, SearchResultWithIndex,
+    PROXY_SEARCH_HEADER, PROXY_SEARCH_HEADER_VALUE,
 };
 use crate::search_queue::SearchQueue;
 
@@ -48,6 +49,7 @@ pub struct SearchResults {
 /// Bundle multiple search queries in a single API request. Use this endpoint to search through multiple indexes at once.
 #[utoipa::path(
     post,
+    request_body = FederatedSearch,
     path = "",
     tag = "Multi-search",
     security(("Bearer" = ["search", "*"])),
@@ -186,18 +188,22 @@ pub async fn multi_search_with_post(
 
     let response = match federation {
         Some(federation) => {
-            let search_result = tokio::task::spawn_blocking(move || {
-                perform_federated_search(&index_scheduler, queries, federation, features)
-            })
-            .await;
+            // check remote header
+            let is_proxy = req
+                .headers()
+                .get(PROXY_SEARCH_HEADER)
+                .is_some_and(|value| value.as_bytes() == PROXY_SEARCH_HEADER_VALUE.as_bytes());
+            let search_result =
+                perform_federated_search(&index_scheduler, queries, federation, features, is_proxy)
+                    .await;
             permit.drop().await;
 
-            if let Ok(Ok(_)) = search_result {
+            if search_result.is_ok() {
                 multi_aggregate.succeed();
             }
 
             analytics.publish(multi_aggregate, &req);
-            HttpResponse::Ok().json(search_result??)
+            HttpResponse::Ok().json(search_result?)
         }
         None => {
             // Explicitly expect a `(ResponseError, usize)` for the error type rather than `ResponseError` only,

crates/meilisearch/src/routes/multi_search_analytics.rs

@@ -13,6 +13,8 @@ pub struct MultiSearchAggregator {
 
     // sum of the number of distinct indexes in each single request, use with total_received to compute an avg
     total_distinct_index_count: usize,
+    // sum of the number of distinct remotes in each single request, use with total_received to compute an avg
+    total_distinct_remote_count: usize,
     // number of queries with a single index, use with total_received to compute a proportion
     total_single_index: usize,
 
@@ -31,46 +33,49 @@ impl MultiSearchAggregator {
     pub fn from_federated_search(federated_search: &FederatedSearch) -> Self {
         let use_federation = federated_search.federation.is_some();
 
-        let distinct_indexes: HashSet<_> = federated_search
-            .queries
-            .iter()
-            .map(|query| {
-                let query = &query;
-                // make sure we get a compilation error if a field gets added to / removed from SearchQueryWithIndex
-                let SearchQueryWithIndex {
-                    index_uid,
-                    federation_options: _,
-                    q: _,
-                    vector: _,
-                    offset: _,
-                    limit: _,
-                    page: _,
-                    hits_per_page: _,
-                    attributes_to_retrieve: _,
-                    retrieve_vectors: _,
-                    attributes_to_crop: _,
-                    crop_length: _,
-                    attributes_to_highlight: _,
-                    show_ranking_score: _,
-                    show_ranking_score_details: _,
-                    show_matches_position: _,
-                    filter: _,
-                    sort: _,
-                    distinct: _,
-                    facets: _,
-                    highlight_pre_tag: _,
-                    highlight_post_tag: _,
-                    crop_marker: _,
-                    matching_strategy: _,
-                    attributes_to_search_on: _,
-                    hybrid: _,
-                    ranking_score_threshold: _,
-                    locales: _,
-                } = query;
+        let mut distinct_indexes = HashSet::with_capacity(federated_search.queries.len());
+        let mut distinct_remotes = HashSet::with_capacity(federated_search.queries.len());
 
-                index_uid.as_str()
-            })
-            .collect();
+        // make sure we get a compilation error if a field gets added to / removed from SearchQueryWithIndex
+        for SearchQueryWithIndex {
+            index_uid,
+            federation_options,
+            q: _,
+            vector: _,
+            offset: _,
+            limit: _,
+            page: _,
+            hits_per_page: _,
+            attributes_to_retrieve: _,
+            retrieve_vectors: _,
+            attributes_to_crop: _,
+            crop_length: _,
+            attributes_to_highlight: _,
+            show_ranking_score: _,
+            show_ranking_score_details: _,
+            show_matches_position: _,
+            filter: _,
+            sort: _,
+            distinct: _,
+            facets: _,
+            highlight_pre_tag: _,
+            highlight_post_tag: _,
+            crop_marker: _,
+            matching_strategy: _,
+            attributes_to_search_on: _,
+            hybrid: _,
+            ranking_score_threshold: _,
+            locales: _,
+        } in &federated_search.queries
+        {
+            if let Some(federation_options) = federation_options {
+                if let Some(remote) = &federation_options.remote {
+                    distinct_remotes.insert(remote.as_str());
+                }
+            }
+
+            distinct_indexes.insert(index_uid.as_str());
+        }
 
         let show_ranking_score =
             federated_search.queries.iter().any(|query| query.show_ranking_score);
@@ -81,6 +86,7 @@ impl MultiSearchAggregator {
             total_received: 1,
             total_succeeded: 0,
             total_distinct_index_count: distinct_indexes.len(),
+            total_distinct_remote_count: distinct_remotes.len(),
             total_single_index: if distinct_indexes.len() == 1 { 1 } else { 0 },
             total_search_count: federated_search.queries.len(),
             show_ranking_score,
@@ -110,6 +116,8 @@ impl Aggregate for MultiSearchAggregator {
         let total_succeeded = this.total_succeeded.saturating_add(new.total_succeeded);
         let total_distinct_index_count =
             this.total_distinct_index_count.saturating_add(new.total_distinct_index_count);
+        let total_distinct_remote_count =
+            this.total_distinct_remote_count.saturating_add(new.total_distinct_remote_count);
         let total_single_index = this.total_single_index.saturating_add(new.total_single_index);
         let total_search_count = this.total_search_count.saturating_add(new.total_search_count);
         let show_ranking_score = this.show_ranking_score || new.show_ranking_score;
@@ -121,6 +129,7 @@ impl Aggregate for MultiSearchAggregator {
             total_received,
             total_succeeded,
             total_distinct_index_count,
+            total_distinct_remote_count,
             total_single_index,
             total_search_count,
             show_ranking_score,
@@ -134,6 +143,7 @@ impl Aggregate for MultiSearchAggregator {
             total_received,
             total_succeeded,
             total_distinct_index_count,
+            total_distinct_remote_count,
             total_single_index,
             total_search_count,
             show_ranking_score,
@@ -152,6 +162,10 @@ impl Aggregate for MultiSearchAggregator {
                 "total_distinct_index_count": total_distinct_index_count,
                 "avg_distinct_index_count": (total_distinct_index_count as f64) / (total_received as f64), // not 0 else returned early
             },
+            "remotes": {
+                "total_distinct_remote_count": total_distinct_remote_count,
+                "avg_distinct_remote_count": (total_distinct_remote_count as f64) / (total_received as f64), // not 0 else returned early
+            },
             "searches": {
                 "total_search_count": total_search_count,
                 "avg_search_count": (total_search_count as f64) / (total_received as f64),

crates/meilisearch/src/routes/network.rs

@@ -0,0 +1,261 @@
+use std::collections::BTreeMap;
+
+use actix_web::web::{self, Data};
+use actix_web::{HttpRequest, HttpResponse};
+use deserr::actix_web::AwebJson;
+use deserr::Deserr;
+use index_scheduler::IndexScheduler;
+use itertools::{EitherOrBoth, Itertools};
+use meilisearch_types::deserr::DeserrJsonError;
+use meilisearch_types::error::deserr_codes::{
+    InvalidNetworkRemotes, InvalidNetworkSearchApiKey, InvalidNetworkSelf, InvalidNetworkUrl,
+};
+use meilisearch_types::error::ResponseError;
+use meilisearch_types::features::{Network as DbNetwork, Remote as DbRemote};
+use meilisearch_types::keys::actions;
+use meilisearch_types::milli::update::Setting;
+use serde::Serialize;
+use tracing::debug;
+use utoipa::{OpenApi, ToSchema};
+
+use crate::analytics::{Aggregate, Analytics};
+use crate::extractors::authentication::policies::ActionPolicy;
+use crate::extractors::authentication::GuardedData;
+use crate::extractors::sequential_extractor::SeqHandler;
+
+#[derive(OpenApi)]
+#[openapi(
+    paths(get_network, patch_network),
+    tags((
+        name = "Network",
+        description = "The `/network` route allows you to describe the topology of a network of Meilisearch instances.
+
+This route is **synchronous**. This means that no task object will be returned, and any change to the network will be made available immediately.",
+        external_docs(url = "https://www.meilisearch.com/docs/reference/api/network"),
+    )),
+)]
+pub struct NetworkApi;
+
+pub fn configure(cfg: &mut web::ServiceConfig) {
+    cfg.service(
+        web::resource("")
+            .route(web::get().to(get_network))
+            .route(web::patch().to(SeqHandler(patch_network))),
+    );
+}
+
+/// Get network topology
+///
+/// Get a list of all Meilisearch instances currently known to this instance.
+#[utoipa::path(
+    get,
+    path = "",
+    tag = "Network",
+    security(("Bearer" = ["network.get", "network.*", "*"])),
+    responses(
+        (status = OK, description = "Known nodes are returned", body = Network, content_type = "application/json", example = json!(
+            {
+            "self": "ms-0",
+            "remotes": {
+            "ms-0": Remote { url: Setting::Set("http://localhost:7700".into()), search_api_key: Setting::Reset },
+            "ms-1": Remote { url: Setting::Set("http://localhost:7701".into()), search_api_key: Setting::Set("foo".into()) },
+            "ms-2": Remote { url: Setting::Set("http://localhost:7702".into()), search_api_key: Setting::Set("bar".into()) },
+        }
+    })),
+        (status = 401, description = "The authorization header is missing", body = ResponseError, content_type = "application/json", example = json!(
+            {
+                "message": "The Authorization header is missing. It must use the bearer authorization method.",
+                "code": "missing_authorization_header",
+                "type": "auth",
+                "link": "https://docs.meilisearch.com/errors#missing_authorization_header"
+            }
+        )),
+    )
+)]
+async fn get_network(
+    index_scheduler: GuardedData<ActionPolicy<{ actions::NETWORK_GET }>, Data<IndexScheduler>>,
+) -> Result<HttpResponse, ResponseError> {
+    index_scheduler.features().check_network("Using the /network route")?;
+
+    let network = index_scheduler.network();
+    debug!(returns = ?network, "Get network");
+    Ok(HttpResponse::Ok().json(network))
+}
+
+#[derive(Debug, Deserr, ToSchema, Serialize)]
+#[deserr(error = DeserrJsonError<InvalidNetworkRemotes>, rename_all = camelCase, deny_unknown_fields)]
+#[serde(rename_all = "camelCase")]
+#[schema(rename_all = "camelCase")]
+pub struct Remote {
+    #[schema(value_type = Option<String>, example = json!({
+        "ms-0": Remote { url: Setting::Set("http://localhost:7700".into()), search_api_key: Setting::Reset },
+        "ms-1": Remote { url: Setting::Set("http://localhost:7701".into()), search_api_key: Setting::Set("foo".into()) },
+        "ms-2": Remote { url: Setting::Set("http://localhost:7702".into()), search_api_key: Setting::Set("bar".into()) },
+    }))]
+    #[deserr(default, error = DeserrJsonError<InvalidNetworkUrl>)]
+    #[serde(default)]
+    pub url: Setting<String>,
+    #[schema(value_type = Option<String>, example = json!("XWnBI8QHUc-4IlqbKPLUDuhftNq19mQtjc6JvmivzJU"))]
+    #[deserr(default, error = DeserrJsonError<InvalidNetworkSearchApiKey>)]
+    #[serde(default)]
+    pub search_api_key: Setting<String>,
+}
+
+#[derive(Debug, Deserr, ToSchema, Serialize)]
+#[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
+#[serde(rename_all = "camelCase")]
+#[schema(rename_all = "camelCase")]
+pub struct Network {
+    #[schema(value_type = Option<BTreeMap<String, Remote>>, example = json!("http://localhost:7700"))]
+    #[deserr(default, error = DeserrJsonError<InvalidNetworkRemotes>)]
+    #[serde(default)]
+    pub remotes: Setting<BTreeMap<String, Option<Remote>>>,
+    #[schema(value_type = Option<String>, example = json!("ms-00"), rename = "self")]
+    #[serde(default, rename = "self")]
+    #[deserr(default, rename = "self", error = DeserrJsonError<InvalidNetworkSelf>)]
+    pub local: Setting<String>,
+}
+
+impl Remote {
+    pub fn try_into_db_node(self, name: &str) -> Result<DbRemote, ResponseError> {
+        Ok(DbRemote {
+            url: self.url.set().ok_or(ResponseError::from_msg(
+                format!("Missing field `.remotes.{name}.url`"),
+                meilisearch_types::error::Code::MissingNetworkUrl,
+            ))?,
+            search_api_key: self.search_api_key.set(),
+        })
+    }
+}
+
+#[derive(Serialize)]
+pub struct PatchNetworkAnalytics {
+    network_size: usize,
+    network_has_self: bool,
+}
+
+impl Aggregate for PatchNetworkAnalytics {
+    fn event_name(&self) -> &'static str {
+        "Network Updated"
+    }
+
+    fn aggregate(self: Box<Self>, new: Box<Self>) -> Box<Self> {
+        Box::new(Self { network_size: new.network_size, network_has_self: new.network_has_self })
+    }
+
+    fn into_event(self: Box<Self>) -> serde_json::Value {
+        serde_json::to_value(*self).unwrap_or_default()
+    }
+}
+
+/// Configure Network
+///
+/// Add or remove nodes from network.
+#[utoipa::path(
+    patch,
+    path = "",
+    tag = "Network",
+    request_body = Network,
+    security(("Bearer" = ["network.update", "network.*", "*"])),
+    responses(
+        (status = OK, description = "New network state is returned",  body = Network, content_type = "application/json", example = json!(
+            {
+                "self": "ms-0",
+                "remotes": {
+                "ms-0": Remote { url: Setting::Set("http://localhost:7700".into()), search_api_key: Setting::Reset },
+                "ms-1": Remote { url: Setting::Set("http://localhost:7701".into()), search_api_key: Setting::Set("foo".into()) },
+                "ms-2": Remote { url: Setting::Set("http://localhost:7702".into()), search_api_key: Setting::Set("bar".into()) },
+            }
+        })),
+        (status = 401, description = "The authorization header is missing", body = ResponseError, content_type = "application/json", example = json!(
+            {
+                "message": "The Authorization header is missing. It must use the bearer authorization method.",
+                "code": "missing_authorization_header",
+                "type": "auth",
+                "link": "https://docs.meilisearch.com/errors#missing_authorization_header"
+            }
+        )),
+    )
+)]
+async fn patch_network(
+    index_scheduler: GuardedData<ActionPolicy<{ actions::NETWORK_UPDATE }>, Data<IndexScheduler>>,
+    new_network: AwebJson<Network, DeserrJsonError>,
+    req: HttpRequest,
+    analytics: Data<Analytics>,
+) -> Result<HttpResponse, ResponseError> {
+    index_scheduler.features().check_network("Using the /network route")?;
+
+    let new_network = new_network.0;
+    let old_network = index_scheduler.network();
+    debug!(parameters = ?new_network, "Patch network");
+
+    let merged_self = match new_network.local {
+        Setting::Set(new_self) => Some(new_self),
+        Setting::Reset => None,
+        Setting::NotSet => old_network.local,
+    };
+
+    let merged_remotes = match new_network.remotes {
+        Setting::Set(new_remotes) => {
+            let mut merged_remotes = BTreeMap::new();
+            for either_or_both in old_network
+                .remotes
+                .into_iter()
+                .merge_join_by(new_remotes.into_iter(), |left, right| left.0.cmp(&right.0))
+            {
+                match either_or_both {
+                    EitherOrBoth::Both((key, old), (_, Some(new))) => {
+                        let DbRemote { url: old_url, search_api_key: old_search_api_key } = old;
+
+                        let Remote { url: new_url, search_api_key: new_search_api_key } = new;
+
+                        let merged = DbRemote {
+                            url: match new_url {
+                                Setting::Set(new_url) => new_url,
+                                Setting::Reset => {
+                                    return Err(ResponseError::from_msg(
+                                        format!(
+                                            "Field `.remotes.{key}.url` cannot be set to `null`"
+                                        ),
+                                        meilisearch_types::error::Code::InvalidNetworkUrl,
+                                    ))
+                                }
+                                Setting::NotSet => old_url,
+                            },
+                            search_api_key: match new_search_api_key {
+                                Setting::Set(new_search_api_key) => Some(new_search_api_key),
+                                Setting::Reset => None,
+                                Setting::NotSet => old_search_api_key,
+                            },
+                        };
+                        merged_remotes.insert(key, merged);
+                    }
+                    EitherOrBoth::Both((_, _), (_, None)) | EitherOrBoth::Right((_, None)) => {}
+                    EitherOrBoth::Left((key, node)) => {
+                        merged_remotes.insert(key, node);
+                    }
+                    EitherOrBoth::Right((key, Some(node))) => {
+                        let node = node.try_into_db_node(&key)?;
+                        merged_remotes.insert(key, node);
+                    }
+                }
+            }
+            merged_remotes
+        }
+        Setting::Reset => BTreeMap::new(),
+        Setting::NotSet => old_network.remotes,
+    };
+
+    analytics.publish(
+        PatchNetworkAnalytics {
+            network_size: merged_remotes.len(),
+            network_has_self: merged_self.is_some(),
+        },
+        &req,
+    );
+
+    let merged_network = DbNetwork { local: merged_self, remotes: merged_remotes };
+    index_scheduler.put_network(merged_network.clone())?;
+    debug!(returns = ?merged_network, "Patch network");
+    Ok(HttpResponse::Ok().json(merged_network))
+}

crates/meilisearch/src/routes/tasks.rs

@@ -1,3 +1,5 @@
+use std::io::ErrorKind;
+
 use actix_web::web::Data;
 use actix_web::{web, HttpRequest, HttpResponse};
 use deserr::actix_web::AwebQueryParameter;
@@ -16,6 +18,7 @@ use serde::Serialize;
 use time::format_description::well_known::Rfc3339;
 use time::macros::format_description;
 use time::{Date, Duration, OffsetDateTime, Time};
+use tokio::io::AsyncReadExt;
 use tokio::task;
 use utoipa::{IntoParams, OpenApi, ToSchema};
 
@@ -44,7 +47,11 @@ pub fn configure(cfg: &mut web::ServiceConfig) {
             .route(web::delete().to(SeqHandler(delete_tasks))),
     )
     .service(web::resource("/cancel").route(web::post().to(SeqHandler(cancel_tasks))))
-    .service(web::resource("/{task_id}").route(web::get().to(SeqHandler(get_task))));
+    .service(web::resource("/{task_id}").route(web::get().to(SeqHandler(get_task))))
+    .service(
+        web::resource("/{task_id}/documents")
+            .route(web::get().to(SeqHandler(get_task_documents_file))),
+    );
 }
 
 #[derive(Debug, Deserr, IntoParams)]
@@ -639,6 +646,76 @@ async fn get_task(
     }
 }
 
+/// Get a task's documents.
+///
+/// Get a [task's documents file](https://www.meilisearch.com/docs/learn/async/asynchronous_operations).
+#[utoipa::path(
+    get,
+    path = "/{taskUid}/documents",
+    tag = "Tasks",
+    security(("Bearer" = ["tasks.get", "tasks.*", "*"])),
+    params(("taskUid", format = UInt32, example = 0, description = "The task identifier", nullable = false)),
+    responses(
+        (status = 200, description = "The content of the task update", body = serde_json::Value, content_type = "application/x-ndjson"),
+        (status = 401, description = "The authorization header is missing", body = ResponseError, content_type = "application/json", example = json!(
+            {
+                "message": "The Authorization header is missing. It must use the bearer authorization method.",
+                "code": "missing_authorization_header",
+                "type": "auth",
+                "link": "https://docs.meilisearch.com/errors#missing_authorization_header"
+            }
+        )),
+        (status = 404, description = "The task uid does not exists", body = ResponseError, content_type = "application/json", example = json!(
+            {
+                "message": "Task :taskUid not found.",
+                "code": "task_not_found",
+                "type": "invalid_request",
+                "link": "https://docs.meilisearch.com/errors/#task_not_found"
+            }
+        ))
+    )
+)]
+async fn get_task_documents_file(
+    index_scheduler: GuardedData<ActionPolicy<{ actions::TASKS_GET }>, Data<IndexScheduler>>,
+    task_uid: web::Path<String>,
+) -> Result<HttpResponse, ResponseError> {
+    index_scheduler.features().check_get_task_documents_route()?;
+    let task_uid_string = task_uid.into_inner();
+
+    let task_uid: TaskId = match task_uid_string.parse() {
+        Ok(id) => id,
+        Err(_e) => {
+            return Err(index_scheduler::Error::InvalidTaskUid { task_uid: task_uid_string }.into())
+        }
+    };
+
+    let query = index_scheduler::Query { uids: Some(vec![task_uid]), ..Query::default() };
+    let filters = index_scheduler.filters();
+    let (tasks, _) = index_scheduler.get_tasks_from_authorized_indexes(&query, filters)?;
+
+    if let Some(task) = tasks.first() {
+        match task.content_uuid() {
+            Some(uuid) => {
+                let mut tfile = match index_scheduler.queue.update_file(uuid) {
+                    Ok(file) => tokio::fs::File::from_std(file),
+                    Err(file_store::Error::IoError(e)) if e.kind() == ErrorKind::NotFound => {
+                        return Err(index_scheduler::Error::TaskFileNotFound(task_uid).into())
+                    }
+                    Err(e) => return Err(e.into()),
+                };
+                // Yes, that's awful to put everything in memory when we could have streamed it from
+                // disk but it's really (really) complex to do with the current state of async Rust.
+                let mut content = String::new();
+                tfile.read_to_string(&mut content).await?;
+                Ok(HttpResponse::Ok().content_type("application/x-ndjson").body(content))
+            }
+            None => Err(index_scheduler::Error::TaskFileNotFound(task_uid).into()),
+        }
+    } else {
+        Err(index_scheduler::Error::TaskNotFound(task_uid).into())
+    }
+}
+
 pub enum DeserializeDateOption {
     Before,
     After,