improve documentation

Remove risk of command injection

.github/workflows/sdks-tests.yml

@@ -25,14 +25,18 @@ jobs:
       - uses: actions/checkout@v5
       - name: Define the Docker image we need to use
         id: define-image
+        env:
+          EVENT_NAME: ${{ github.event_name }}
+          DOCKER_IMAGE_INPUT: ${{ github.event.inputs.docker_image }}
         run: |
-          event=${{ github.event_name }}
           echo "docker-image=nightly" >> $GITHUB_OUTPUT
-          if [[ $event == 'workflow_dispatch' ]]; then
-            echo "docker-image=${{ github.event.inputs.docker_image }}" >> $GITHUB_OUTPUT
+          if [[ "$EVENT_NAME" == 'workflow_dispatch' ]]; then
+            echo "docker-image=$DOCKER_IMAGE_INPUT" >> $GITHUB_OUTPUT
           fi
       - name: Docker image is ${{ steps.define-image.outputs.docker-image }}
-        run: echo "Docker image is ${{ steps.define-image.outputs.docker-image }}"
+        env:
+          DOCKER_IMAGE: ${{ steps.define-image.outputs.docker-image }}
+        run: echo "Docker image is $DOCKER_IMAGE"
 
 ##########
 ## SDKs ##

crates/fuzzers/src/bin/fuzz-indexing.rs

@@ -113,9 +113,9 @@ fn main() {
 
                             for op in &operations {
                                 match op {
-                                    Either::Left(documents) => {
-                                        indexer.replace_documents(documents).unwrap()
-                                    }
+                                    Either::Left(documents) => indexer
+                                        .replace_documents(documents, Default::default())
+                                        .unwrap(),
                                     Either::Right(ids) => indexer.delete_documents(ids),
                                 }
                             }

crates/index-scheduler/src/dump.rs

@@ -164,6 +164,7 @@ impl<'a> Dump<'a> {
                     content_file: content_uuid.ok_or(Error::CorruptedDump)?,
                     documents_count,
                     allow_index_creation,
+                    on_missing_document: Default::default(),
                 },
                 KindDump::DocumentDeletion { documents_ids } => KindWithContent::DocumentDeletion {
                     documents_ids,

crates/index-scheduler/src/queue/batches.rs

@@ -502,11 +502,13 @@ impl Queue {
             *before_finished_at,
         )?;
 
-        batches = if query.reverse.unwrap_or_default() {
-            batches.into_iter().take(*limit).collect()
-        } else {
-            batches.into_iter().rev().take(*limit).collect()
-        };
+        if let Some(limit) = limit {
+            batches = if query.reverse.unwrap_or_default() {
+                batches.into_iter().take(*limit as usize).collect()
+            } else {
+                batches.into_iter().rev().take(*limit as usize).collect()
+            };
+        }
 
         Ok(batches)
     }
@@ -600,8 +602,11 @@ impl Queue {
             Box::new(batches.into_iter().rev()) as Box<dyn Iterator<Item = u32>>
         };
 
-        let batches =
-            self.batches.get_existing_batches(rtxn, batches.take(query.limit), processing)?;
+        let batches = self.batches.get_existing_batches(
+            rtxn,
+            batches.take(query.limit.unwrap_or(u32::MAX) as usize),
+            processing,
+        )?;
 
         Ok((batches, total))
     }

crates/index-scheduler/src/queue/batches_test.rs

@@ -28,21 +28,21 @@ fn query_batches_from_and_limit() {
 
     let proc = index_scheduler.processing_tasks.read().unwrap().clone();
     let rtxn = index_scheduler.env.read_txn().unwrap();
-    let query = Query { limit: 0, ..Default::default() };
+    let query = Query { limit: Some(0), ..Default::default() };
     let (batches, _) = index_scheduler
         .queue
         .get_batch_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &proc)
         .unwrap();
     snapshot!(snapshot_bitmap(&batches), @"[]");
 
-    let query = Query { limit: 1, ..Default::default() };
+    let query = Query { limit: Some(1), ..Default::default() };
     let (batches, _) = index_scheduler
         .queue
         .get_batch_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &proc)
         .unwrap();
     snapshot!(snapshot_bitmap(&batches), @"[2,]");
 
-    let query = Query { limit: 2, ..Default::default() };
+    let query = Query { limit: Some(2), ..Default::default() };
     let (batches, _) = index_scheduler
         .queue
         .get_batch_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &proc)
@@ -63,14 +63,14 @@ fn query_batches_from_and_limit() {
         .unwrap();
     snapshot!(snapshot_bitmap(&batches), @"[0,1,2,]");
 
-    let query = Query { from: Some(1), limit: 1, ..Default::default() };
+    let query = Query { from: Some(1), limit: Some(1), ..Default::default() };
     let (batches, _) = index_scheduler
         .queue
         .get_batch_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &proc)
         .unwrap();
     snapshot!(snapshot_bitmap(&batches), @"[1,]");
 
-    let query = Query { from: Some(1), limit: 2, ..Default::default() };
+    let query = Query { from: Some(1), limit: Some(2), ..Default::default() };
     let (batches, _) = index_scheduler
         .queue
         .get_batch_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &proc)

crates/index-scheduler/src/queue/mod.rs

@@ -31,9 +31,6 @@ use crate::{Error, IndexSchedulerOptions, Result, TaskId};
 
 /// The number of database used by queue itself
 const NUMBER_OF_DATABASES: u32 = 1;
-/// The default limit for pagination
-const DEFAULT_LIMIT: usize = 20;
-
 /// Database const names for the `IndexScheduler`.
 mod db_name {
     pub const BATCH_TO_TASKS_MAPPING: &str = "batch-to-tasks-mapping";
@@ -43,11 +40,11 @@ mod db_name {
 ///
 /// An empty/default query (where each field is set to `None`) matches all tasks.
 /// Each non-null field restricts the set of tasks further.
-#[derive(Debug, Clone, PartialEq, Eq)]
+#[derive(Default, Debug, Clone, PartialEq, Eq)]
 pub struct Query {
-    /// The maximum number of tasks to be matched. Defaults to 20.
-    pub limit: usize,
-    /// The minimum [task id](`meilisearch_types::tasks::Task::uid`) to be matched. Defaults to 0.
+    /// The maximum number of tasks to be matched
+    pub limit: Option<u32>,
+    /// The minimum [task id](`meilisearch_types::tasks::Task::uid`) to be matched
     pub from: Option<u32>,
     /// The order used to return the tasks. By default the newest tasks are returned first and the boolean is `false`.
     pub reverse: Option<bool>,
@@ -86,29 +83,32 @@ pub struct Query {
     pub after_finished_at: Option<OffsetDateTime>,
 }
 
-impl Default for Query {
-    fn default() -> Self {
-        Self {
-            limit: DEFAULT_LIMIT,
-            from: Default::default(),
-            reverse: Default::default(),
-            uids: Default::default(),
-            batch_uids: Default::default(),
-            statuses: Default::default(),
-            types: Default::default(),
-            index_uids: Default::default(),
-            canceled_by: Default::default(),
-            before_enqueued_at: Default::default(),
-            after_enqueued_at: Default::default(),
-            before_started_at: Default::default(),
-            after_started_at: Default::default(),
-            before_finished_at: Default::default(),
-            after_finished_at: Default::default(),
-        }
-    }
-}
-
 impl Query {
+    /// Return `true` if every field of the query is set to `None`, such that the query
+    /// matches all tasks.
+    pub fn is_empty(&self) -> bool {
+        matches!(
+            self,
+            Query {
+                limit: None,
+                from: None,
+                reverse: None,
+                uids: None,
+                batch_uids: None,
+                statuses: None,
+                types: None,
+                index_uids: None,
+                canceled_by: None,
+                before_enqueued_at: None,
+                after_enqueued_at: None,
+                before_started_at: None,
+                after_started_at: None,
+                before_finished_at: None,
+                after_finished_at: None,
+            }
+        )
+    }
+
     /// Add an [index id](meilisearch_types::tasks::Task::index_uid) to the list of permitted indexes.
     pub fn with_index(self, index_uid: String) -> Self {
         let mut index_vec = self.index_uids.unwrap_or_default();
@@ -119,7 +119,7 @@ impl Query {
     // Removes the `from` and `limit` restrictions from the query.
     // Useful to get the total number of tasks matching a filter.
     pub fn without_limits(self) -> Self {
-        Query { limit: usize::MAX, from: None, ..self }
+        Query { limit: None, from: None, ..self }
     }
 }
 

crates/index-scheduler/src/queue/tasks.rs

@@ -465,11 +465,13 @@ impl Queue {
             *before_finished_at,
         )?;
 
-        tasks = if query.reverse.unwrap_or_default() {
-            tasks.into_iter().take(*limit).collect()
-        } else {
-            tasks.into_iter().rev().take(*limit).collect()
-        };
+        if let Some(limit) = limit {
+            tasks = if query.reverse.unwrap_or_default() {
+                tasks.into_iter().take(*limit as usize).collect()
+            } else {
+                tasks.into_iter().rev().take(*limit as usize).collect()
+            };
+        }
 
         Ok(tasks)
     }
@@ -527,7 +529,9 @@ impl Queue {
         } else {
             Box::new(tasks.into_iter().rev()) as Box<dyn Iterator<Item = u32>>
         };
-        let tasks = self.tasks.get_existing_tasks(rtxn, tasks.take(query.limit))?;
+        let tasks = self
+            .tasks
+            .get_existing_tasks(rtxn, tasks.take(query.limit.unwrap_or(u32::MAX) as usize))?;
 
         let ProcessingTasks { batch, processing, progress: _ } = processing_tasks;
 

crates/index-scheduler/src/queue/tasks_test.rs

@@ -28,21 +28,21 @@ fn query_tasks_from_and_limit() {
 
     let rtxn = index_scheduler.env.read_txn().unwrap();
     let processing = index_scheduler.processing_tasks.read().unwrap();
-    let query = Query { limit: 0, ..Default::default() };
+    let query = Query { limit: Some(0), ..Default::default() };
     let (tasks, _) = index_scheduler
         .queue
         .get_task_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &processing)
         .unwrap();
     snapshot!(snapshot_bitmap(&tasks), @"[]");
 
-    let query = Query { limit: 1, ..Default::default() };
+    let query = Query { limit: Some(1), ..Default::default() };
     let (tasks, _) = index_scheduler
         .queue
         .get_task_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &processing)
         .unwrap();
     snapshot!(snapshot_bitmap(&tasks), @"[2,]");
 
-    let query = Query { limit: 2, ..Default::default() };
+    let query = Query { limit: Some(2), ..Default::default() };
     let (tasks, _) = index_scheduler
         .queue
         .get_task_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &processing)
@@ -63,14 +63,14 @@ fn query_tasks_from_and_limit() {
         .unwrap();
     snapshot!(snapshot_bitmap(&tasks), @"[0,1,2,]");
 
-    let query = Query { from: Some(1), limit: 1, ..Default::default() };
+    let query = Query { from: Some(1), limit: Some(1), ..Default::default() };
     let (tasks, _) = index_scheduler
         .queue
         .get_task_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &processing)
         .unwrap();
     snapshot!(snapshot_bitmap(&tasks), @"[1,]");
 
-    let query = Query { from: Some(1), limit: 2, ..Default::default() };
+    let query = Query { from: Some(1), limit: Some(2), ..Default::default() };
     let (tasks, _) = index_scheduler
         .queue
         .get_task_ids_from_authorized_indexes(&rtxn, &query, &AuthFilter::default(), &processing)

crates/index-scheduler/src/scheduler/autobatcher_test.rs

@@ -40,6 +40,7 @@ fn doc_imp(
         content_file: Uuid::new_v4(),
         documents_count: 0,
         allow_index_creation,
+        on_missing_document: Default::default(),
     }
 }
 

crates/index-scheduler/src/scheduler/create_batch.rs

@@ -2,7 +2,7 @@ use std::fmt;
 use std::io::ErrorKind;
 
 use meilisearch_types::heed::RoTxn;
-use meilisearch_types::milli::update::IndexDocumentsMethod;
+use meilisearch_types::milli::update::{IndexDocumentsMethod, MissingDocumentPolicy};
 use meilisearch_types::settings::{Settings, Unchecked};
 use meilisearch_types::tasks::{BatchStopReason, Kind, KindWithContent, Status, Task};
 use roaring::RoaringBitmap;
@@ -63,8 +63,8 @@ pub(crate) enum Batch {
 
 #[derive(Debug)]
 pub(crate) enum DocumentOperation {
-    Replace(Uuid),
-    Update(Uuid),
+    Replace { content_file: Uuid, on_missing_document: MissingDocumentPolicy },
+    Update { content_file: Uuid, on_missing_document: MissingDocumentPolicy },
     Delete(Vec<String>),
 }
 
@@ -293,13 +293,22 @@ impl IndexScheduler {
                 for task in tasks.iter() {
                     match task.kind {
                         KindWithContent::DocumentAdditionOrUpdate {
-                            content_file, method, ..
+                            content_file,
+                            method,
+                            on_missing_document,
+                            ..
                         } => match method {
                             IndexDocumentsMethod::ReplaceDocuments => {
-                                operations.push(DocumentOperation::Replace(content_file))
+                                operations.push(DocumentOperation::Replace {
+                                    content_file,
+                                    on_missing_document,
+                                })
                             }
                             IndexDocumentsMethod::UpdateDocuments => {
-                                operations.push(DocumentOperation::Update(content_file))
+                                operations.push(DocumentOperation::Update {
+                                    content_file,
+                                    on_missing_document,
+                                })
                             }
                             _ => unreachable!("Unknown document merging method"),
                         },

crates/index-scheduler/src/scheduler/process_index_operation.rs

@@ -77,8 +77,8 @@ impl IndexScheduler {
                 let mut content_files = Vec::new();
                 for operation in &operations {
                     match operation {
-                        DocumentOperation::Replace(content_uuid)
-                        | DocumentOperation::Update(content_uuid) => {
+                        DocumentOperation::Replace { content_file: content_uuid, .. }
+                        | DocumentOperation::Update { content_file: content_uuid, .. } => {
                             let content_file = self.queue.file_store.get_update(*content_uuid)?;
                             let mmap = unsafe { memmap2::Mmap::map(&content_file)? };
                             content_files.push(mmap);
@@ -100,16 +100,16 @@ impl IndexScheduler {
                 let embedders = self.embedders(index_uid.clone(), embedders)?;
                 for operation in operations {
                     match operation {
-                        DocumentOperation::Replace(_content_uuid) => {
+                        DocumentOperation::Replace { content_file: _, on_missing_document } => {
                             let mmap = content_files_iter.next().unwrap();
                             indexer
-                                .replace_documents(mmap)
+                                .replace_documents(mmap, on_missing_document)
                                 .map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
                         }
-                        DocumentOperation::Update(_content_uuid) => {
+                        DocumentOperation::Update { content_file: _, on_missing_document } => {
                             let mmap = content_files_iter.next().unwrap();
                             indexer
-                                .update_documents(mmap)
+                                .update_documents(mmap, on_missing_document)
                                 .map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
                         }
                         DocumentOperation::Delete(document_ids) => {

crates/index-scheduler/src/scheduler/test.rs

@@ -294,6 +294,7 @@ fn document_addition_and_index_deletion() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -482,6 +483,7 @@ fn document_addition_and_index_deletion_on_unexisting_index() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,

crates/index-scheduler/src/scheduler/test_document_addition.rs

@@ -31,6 +31,7 @@ fn document_addition() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -67,6 +68,7 @@ fn document_addition_and_document_deletion() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -133,6 +135,7 @@ fn document_deletion_and_document_addition() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -185,6 +188,7 @@ fn test_document_replace() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -236,6 +240,7 @@ fn test_document_update() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -289,6 +294,7 @@ fn test_mixed_document_addition() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -340,6 +346,7 @@ fn test_document_replace_without_autobatching() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -395,6 +402,7 @@ fn test_document_update_without_autobatching() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -454,6 +462,7 @@ fn test_document_addition_cant_create_index_without_index() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: false,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -506,6 +515,7 @@ fn test_document_addition_cant_create_index_without_index_without_autobatching()
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: false,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -568,6 +578,7 @@ fn test_document_addition_cant_create_index_with_index() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: false,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -635,6 +646,7 @@ fn test_document_addition_cant_create_index_with_index_without_autobatching() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: false,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -707,6 +719,7 @@ fn test_document_addition_mixed_rights_with_index() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -764,6 +777,7 @@ fn test_document_addition_mixed_right_without_index_starts_with_cant_create() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -820,6 +834,7 @@ fn test_document_addition_with_multiple_primary_key() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -883,6 +898,7 @@ fn test_document_addition_with_multiple_primary_key_batch_wrong_key() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -943,6 +959,7 @@ fn test_document_addition_with_bad_primary_key() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -1029,6 +1046,7 @@ fn test_document_addition_with_set_and_null_primary_key() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,
@@ -1104,6 +1122,7 @@ fn test_document_addition_with_set_and_null_primary_key_inference_works() {
                     content_file: uuid,
                     documents_count,
                     allow_index_creation: true,
+                    on_missing_document: Default::default(),
                 },
                 None,
                 false,

crates/index-scheduler/src/scheduler/test_embedders.rs

@@ -173,6 +173,7 @@ fn import_vectors() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -263,6 +264,7 @@ fn import_vectors() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -399,6 +401,7 @@ fn import_vectors_first_and_embedder_later() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -539,6 +542,7 @@ fn import_vectors_first_and_embedder_later() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -640,6 +644,7 @@ fn delete_document_containing_vector() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: false,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -818,6 +823,7 @@ fn delete_embedder_with_user_provided_vectors() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: false,
+                on_missing_document: Default::default(),
             },
             None,
             false,

crates/index-scheduler/src/scheduler/test_failure.rs

@@ -52,6 +52,7 @@ fn fail_in_process_batch_for_document_addition() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -94,6 +95,7 @@ fn fail_in_update_task_after_process_batch_success_for_document_addition() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,
@@ -160,6 +162,7 @@ fn fail_in_process_batch_for_document_deletion() {
                 content_file: uuid,
                 documents_count,
                 allow_index_creation: true,
+                on_missing_document: Default::default(),
             },
             None,
             false,

crates/index-scheduler/src/test_utils.rs

@@ -197,6 +197,7 @@ pub(crate) fn replace_document_import_task(
         content_file: Uuid::from_u128(content_file_uuid),
         documents_count,
         allow_index_creation: true,
+        on_missing_document: Default::default(),
     }
 }
 

crates/meilisearch-types/src/error.rs

@@ -255,6 +255,7 @@ InvalidIndexLimit                              , InvalidRequest       , BAD_REQU
 InvalidIndexOffset                             , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexPrimaryKey                         , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexCustomMetadata                     , InvalidRequest       , BAD_REQUEST ;
+InvalidSkipCreation                            , InvalidRequest       , BAD_REQUEST ;
 InvalidIndexUid                                , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFacets                       , InvalidRequest       , BAD_REQUEST ;
 InvalidMultiSearchFacetsByIndex                , InvalidRequest       , BAD_REQUEST ;

crates/meilisearch-types/src/tasks.rs

@@ -5,7 +5,7 @@ use std::str::FromStr;
 
 use byte_unit::Byte;
 use enum_iterator::Sequence;
-use milli::update::IndexDocumentsMethod;
+use milli::update::{IndexDocumentsMethod, MissingDocumentPolicy};
 use milli::Object;
 use roaring::RoaringBitmap;
 use serde::{Deserialize, Serialize, Serializer};
@@ -114,6 +114,7 @@ pub enum KindWithContent {
         content_file: Uuid,
         documents_count: u64,
         allow_index_creation: bool,
+        on_missing_document: MissingDocumentPolicy,
     },
     DocumentDeletion {
         index_uid: String,

crates/meilisearch/src/lib.rs

@@ -629,7 +629,7 @@ fn import_dump(
 
             let mmap = unsafe { memmap2::Mmap::map(index_reader.documents_file())? };
 
-            indexer.replace_documents(&mmap)?;
+            indexer.replace_documents(&mmap, Default::default())?;
 
             let indexer_config = index_scheduler.indexer_config();
             let pool = &indexer_config.thread_pool;

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

@@ -20,7 +20,7 @@ use meilisearch_types::heed::RoTxn;
 use meilisearch_types::index_uid::IndexUid;
 use meilisearch_types::milli::documents::sort::recursive_sort;
 use meilisearch_types::milli::index::EmbeddingsWithMetadata;
-use meilisearch_types::milli::update::IndexDocumentsMethod;
+use meilisearch_types::milli::update::{IndexDocumentsMethod, MissingDocumentPolicy};
 use meilisearch_types::milli::vector::parsed_vectors::ExplicitVectors;
 use meilisearch_types::milli::{AscDesc, DocumentId};
 use meilisearch_types::serde_cs::vec::CS;
@@ -687,6 +687,11 @@ pub struct UpdateDocumentsQuery {
     #[param(example = "custom")]
     #[deserr(default, error = DeserrQueryParamError<InvalidIndexCustomMetadata>)]
     pub custom_metadata: Option<String>,
+
+    #[param(example = "true")]
+    #[deserr(default, try_from(&String) = from_string_skip_creation -> DeserrQueryParamError<InvalidSkipCreation>, error = DeserrQueryParamError<InvalidSkipCreation>)]
+    /// Only update documents if they already exist.
+    pub skip_creation: Option<bool>,
 }
 
 #[derive(Deserialize, Debug, Deserr, IntoParams)]
@@ -711,6 +716,23 @@ fn from_char_csv_delimiter(
     }
 }
 
+fn from_string_skip_creation(
+    s: &String,
+) -> Result<Option<bool>, DeserrQueryParamError<InvalidSkipCreation>> {
+    if s.eq_ignore_ascii_case("true") {
+        return Ok(Some(true));
+    }
+
+    if s.eq_ignore_ascii_case("false") {
+        return Ok(Some(false));
+    }
+
+    Err(DeserrQueryParamError::new(
+        format!("skipCreation must be either `true` or `false`. Found: `{}`", s),
+        Code::InvalidSkipCreation,
+    ))
+}
+
 aggregate_methods!(
     Replaced => "Documents Added",
     Updated => "Documents Updated",
@@ -840,6 +862,7 @@ pub async fn replace_documents(
         params.custom_metadata,
         dry_run,
         allow_index_creation,
+        params.skip_creation,
         &req,
     )
     .await?;
@@ -943,6 +966,7 @@ pub async fn update_documents(
         params.custom_metadata,
         dry_run,
         allow_index_creation,
+        params.skip_creation,
         &req,
     )
     .await?;
@@ -963,6 +987,7 @@ async fn document_addition(
     custom_metadata: Option<String>,
     dry_run: bool,
     allow_index_creation: bool,
+    skip_creation: Option<bool>,
     req: &HttpRequest,
 ) -> Result<SummarizedTaskView, MeilisearchHttpError> {
     let mime_type = extract_mime_type(req)?;
@@ -1083,6 +1108,11 @@ async fn document_addition(
         primary_key,
         allow_index_creation,
         index_uid: index_uid.to_string(),
+        on_missing_document: if matches!(skip_creation, Some(true)) {
+            MissingDocumentPolicy::Skip
+        } else {
+            MissingDocumentPolicy::Create
+        },
     };
 
     let scheduler = index_scheduler.clone();

crates/meilisearch/src/routes/metrics.rs

@@ -185,7 +185,7 @@ pub async fn get_metrics(
         // Fetch the finished batches...
         &Query {
             statuses: Some(vec![Status::Succeeded, Status::Failed]),
-            limit: 1,
+            limit: Some(1),
             ..Query::default()
         },
         auth_filters,
@@ -214,7 +214,7 @@ pub async fn get_metrics(
     let task_queue_latency_seconds = index_scheduler
         .get_tasks_from_authorized_indexes(
             &Query {
-                limit: 1,
+                limit: Some(1),
                 reverse: Some(true),
                 statuses: Some(vec![Status::Enqueued, Status::Processing]),
                 ..Query::default()

crates/meilisearch/src/routes/tasks.rs

@@ -126,7 +126,7 @@ pub struct TasksFilterQuery {
 impl TasksFilterQuery {
     pub(crate) fn into_query(self) -> Query {
         Query {
-            limit: self.limit.0 as usize,
+            limit: Some(self.limit.0),
             from: self.from.as_deref().copied(),
             reverse: self.reverse.as_deref().copied(),
             batch_uids: self.batch_uids.merge_star_and_none(),
@@ -225,8 +225,7 @@ pub struct TaskDeletionOrCancelationQuery {
 impl TaskDeletionOrCancelationQuery {
     fn into_query(self) -> Query {
         Query {
-            // We want to delete all tasks that match the given filters
-            limit: usize::MAX,
+            limit: None,
             from: None,
             reverse: None,
             batch_uids: self.batch_uids.merge_star_and_none(),

crates/milli/src/search/new/tests/integration.rs

@@ -64,7 +64,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     let payload = unsafe { memmap2::Mmap::map(&file).unwrap() };
 
     // index documents
-    indexer.replace_documents(&payload).unwrap();
+    indexer.replace_documents(&payload, Default::default()).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, operation_stats, primary_key) = indexer

crates/milli/src/test_index.rs

@@ -70,9 +70,11 @@ impl TempIndex {
         let mut indexer = indexer::DocumentOperation::new();
         match self.index_documents_config.update_method {
             IndexDocumentsMethod::ReplaceDocuments => {
-                indexer.replace_documents(&documents).unwrap()
+                indexer.replace_documents(&documents, Default::default()).unwrap()
+            }
+            IndexDocumentsMethod::UpdateDocuments => {
+                indexer.update_documents(&documents, Default::default()).unwrap()
             }
-            IndexDocumentsMethod::UpdateDocuments => indexer.update_documents(&documents).unwrap(),
         }
 
         let indexer_alloc = Bump::new();
@@ -232,7 +234,7 @@ fn aborting_indexation() {
         { "id": 2, "name": "bob", "age": 20 },
         { "id": 2, "name": "bob", "age": 20 },
     ]);
-    indexer.replace_documents(&payload).unwrap();
+    indexer.replace_documents(&payload, Default::default()).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, _operation_stats, primary_key) = indexer

crates/milli/src/update/index_documents/mod.rs

@@ -67,6 +67,21 @@ pub enum IndexDocumentsMethod {
     UpdateDocuments,
 }
 
+/// Controls whether new documents should be created when they don't already exist.
+///
+/// This policy is checked when processing a document whose ID is not found in the index.
+/// It applies to both update and replace operations.
+#[derive(Default, Debug, Clone, Copy, PartialEq, Eq, Hash, Serialize, Deserialize)]
+pub enum MissingDocumentPolicy {
+    /// Create the document if it doesn't exist. This is the default behavior.
+    #[default]
+    Create,
+
+    /// Skip the document silently if it doesn't exist. No error is returned, the document is simply
+    /// not indexed.
+    Skip,
+}
+
 pub struct IndexDocuments<'t, 'i, 'a, FP, FA> {
     wtxn: &'t mut heed::RwTxn<'i>,
     index: &'i Index,
@@ -1971,10 +1986,10 @@ mod tests {
         let mut new_fields_ids_map = db_fields_ids_map.clone();
 
         let mut indexer = indexer::DocumentOperation::new();
-        indexer.replace_documents(&doc1).unwrap();
-        indexer.replace_documents(&doc2).unwrap();
-        indexer.replace_documents(&doc3).unwrap();
-        indexer.replace_documents(&doc4).unwrap();
+        indexer.replace_documents(&doc1, Default::default()).unwrap();
+        indexer.replace_documents(&doc2, Default::default()).unwrap();
+        indexer.replace_documents(&doc3, Default::default()).unwrap();
+        indexer.replace_documents(&doc4, Default::default()).unwrap();
 
         let indexer_alloc = Bump::new();
         let (_document_changes, operation_stats, _primary_key) = indexer
@@ -2024,10 +2039,10 @@ mod tests {
         let mut new_fields_ids_map = db_fields_ids_map.clone();
 
         let mut indexer = indexer::DocumentOperation::new();
-        indexer.replace_documents(&doc1).unwrap();
-        indexer.update_documents(&doc2).unwrap();
-        indexer.update_documents(&doc3).unwrap();
-        indexer.update_documents(&doc4).unwrap();
+        indexer.replace_documents(&doc1, Default::default()).unwrap();
+        indexer.update_documents(&doc2, Default::default()).unwrap();
+        indexer.update_documents(&doc3, Default::default()).unwrap();
+        indexer.update_documents(&doc4, Default::default()).unwrap();
 
         let indexer_alloc = Bump::new();
         let (document_changes, operation_stats, primary_key) = indexer
@@ -2112,11 +2127,11 @@ mod tests {
         let mut new_fields_ids_map = db_fields_ids_map.clone();
 
         let mut indexer = indexer::DocumentOperation::new();
-        indexer.replace_documents(&doc1).unwrap();
-        indexer.update_documents(&doc2).unwrap();
-        indexer.update_documents(&doc3).unwrap();
-        indexer.replace_documents(&doc4).unwrap();
-        indexer.update_documents(&doc5).unwrap();
+        indexer.replace_documents(&doc1, Default::default()).unwrap();
+        indexer.update_documents(&doc2, Default::default()).unwrap();
+        indexer.update_documents(&doc3, Default::default()).unwrap();
+        indexer.replace_documents(&doc4, Default::default()).unwrap();
+        indexer.update_documents(&doc5, Default::default()).unwrap();
 
         let indexer_alloc = Bump::new();
         let (document_changes, operation_stats, primary_key) = indexer
@@ -2307,7 +2322,7 @@ mod tests {
         let indexer_alloc = Bump::new();
         let embedders = RuntimeEmbedders::default();
         let mut indexer = indexer::DocumentOperation::new();
-        indexer.replace_documents(&documents).unwrap();
+        indexer.replace_documents(&documents, Default::default()).unwrap();
         indexer.delete_documents(&["2"]);
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2362,13 +2377,13 @@ mod tests {
             { "id": 3, "name": "jean", "age": 25 },
         ]);
         let mut indexer = indexer::DocumentOperation::new();
-        indexer.update_documents(&documents).unwrap();
+        indexer.update_documents(&documents, Default::default()).unwrap();
 
         let documents = documents!([
             { "id": 2, "catto": "jorts" },
             { "id": 3, "legs": 4 },
         ]);
-        indexer.update_documents(&documents).unwrap();
+        indexer.update_documents(&documents, Default::default()).unwrap();
         indexer.delete_documents(&["1", "2"]);
 
         let indexer_alloc = Bump::new();
@@ -2426,7 +2441,7 @@ mod tests {
         let indexer_alloc = Bump::new();
         let embedders = RuntimeEmbedders::default();
         let mut indexer = indexer::DocumentOperation::new();
-        indexer.update_documents(&documents).unwrap();
+        indexer.update_documents(&documents, Default::default()).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2479,7 +2494,7 @@ mod tests {
         let indexer_alloc = Bump::new();
         let embedders = RuntimeEmbedders::default();
         let mut indexer = indexer::DocumentOperation::new();
-        indexer.update_documents(&documents).unwrap();
+        indexer.update_documents(&documents, Default::default()).unwrap();
         indexer.delete_documents(&["1", "2"]);
 
         let (document_changes, _operation_stats, primary_key) = indexer
@@ -2536,7 +2551,7 @@ mod tests {
             { "id": 2, "doggo": { "name": "jean", "age": 20 } },
             { "id": 3, "name": "bob", "age": 25 },
         ]);
-        indexer.update_documents(&documents).unwrap();
+        indexer.update_documents(&documents, Default::default()).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2595,7 +2610,7 @@ mod tests {
             { "id": 2, "doggo": { "name": "jean", "age": 20 } },
             { "id": 3, "name": "bob", "age": 25 },
         ]);
-        indexer.update_documents(&documents).unwrap();
+        indexer.update_documents(&documents, Default::default()).unwrap();
 
         indexer.delete_documents(&["1", "2", "1", "2"]);
 
@@ -2651,7 +2666,7 @@ mod tests {
         let documents = documents!([
             { "id": 1, "doggo": "kevin" },
         ]);
-        indexer.update_documents(&documents).unwrap();
+        indexer.update_documents(&documents, Default::default()).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2705,7 +2720,7 @@ mod tests {
             { "id": 1, "catto": "jorts" },
         ]);
 
-        indexer.replace_documents(&documents).unwrap();
+        indexer.replace_documents(&documents, Default::default()).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2916,7 +2931,7 @@ mod tests {
         let documents = documents!([
             { "id": 1, "doggo": "bernese" },
         ]);
-        indexer.replace_documents(&documents).unwrap();
+        indexer.replace_documents(&documents, Default::default()).unwrap();
 
         // FINISHING
         let (document_changes, _operation_stats, primary_key) = indexer
@@ -2978,7 +2993,7 @@ mod tests {
         let documents = documents!([
             { "id": 0, "catto": "jorts" },
         ]);
-        indexer.replace_documents(&documents).unwrap();
+        indexer.replace_documents(&documents, Default::default()).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -3036,7 +3051,7 @@ mod tests {
         let documents = documents!([
             { "id": 1, "catto": "jorts" },
         ]);
-        indexer.replace_documents(&documents).unwrap();
+        indexer.replace_documents(&documents, Default::default()).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(

crates/milli/src/update/new/indexer/document_operation.rs

@@ -21,7 +21,7 @@ use crate::update::new::indexer::current_edition::sharding::Shards;
 use crate::update::new::steps::IndexingStep;
 use crate::update::new::thread_local::MostlySend;
 use crate::update::new::{DocumentIdentifiers, Insertion, Update};
-use crate::update::{AvailableIds, IndexDocumentsMethod};
+use crate::update::{AvailableIds, IndexDocumentsMethod, MissingDocumentPolicy};
 use crate::{DocumentId, Error, FieldsIdsMap, Index, InternalError, Result, UserError};
 
 #[derive(Default)]
@@ -37,20 +37,28 @@ impl<'pl> DocumentOperation<'pl> {
     /// Append a replacement of documents.
     ///
     /// The payload is expected to be in the NDJSON format
-    pub fn replace_documents(&mut self, payload: &'pl Mmap) -> Result<()> {
+    pub fn replace_documents(
+        &mut self,
+        payload: &'pl Mmap,
+        on_missing_document: MissingDocumentPolicy,
+    ) -> Result<()> {
         #[cfg(unix)]
         payload.advise(memmap2::Advice::Sequential)?;
-        self.operations.push(Payload::Replace(&payload[..]));
+        self.operations.push(Payload::Replace { payload: &payload[..], on_missing_document });
         Ok(())
     }
 
     /// Append an update of documents.
     ///
     /// The payload is expected to be in the NDJSON format
-    pub fn update_documents(&mut self, payload: &'pl Mmap) -> Result<()> {
+    pub fn update_documents(
+        &mut self,
+        payload: &'pl Mmap,
+        on_missing_document: MissingDocumentPolicy,
+    ) -> Result<()> {
         #[cfg(unix)]
         payload.advise(memmap2::Advice::Sequential)?;
-        self.operations.push(Payload::Update(&payload[..]));
+        self.operations.push(Payload::Update { payload: &payload[..], on_missing_document });
         Ok(())
     }
 
@@ -98,34 +106,40 @@ impl<'pl> DocumentOperation<'pl> {
 
             let mut bytes = 0;
             let result = match operation {
-                Payload::Replace(payload) => extract_addition_payload_changes(
-                    indexer,
-                    index,
-                    rtxn,
-                    primary_key_from_op,
-                    &mut primary_key,
-                    new_fields_ids_map,
-                    &mut available_docids,
-                    &mut bytes,
-                    &docids_version_offsets,
-                    IndexDocumentsMethod::ReplaceDocuments,
-                    shards,
-                    payload,
-                ),
-                Payload::Update(payload) => extract_addition_payload_changes(
-                    indexer,
-                    index,
-                    rtxn,
-                    primary_key_from_op,
-                    &mut primary_key,
-                    new_fields_ids_map,
-                    &mut available_docids,
-                    &mut bytes,
-                    &docids_version_offsets,
-                    IndexDocumentsMethod::UpdateDocuments,
-                    shards,
-                    payload,
-                ),
+                Payload::Replace { payload, on_missing_document } => {
+                    extract_addition_payload_changes(
+                        indexer,
+                        index,
+                        rtxn,
+                        primary_key_from_op,
+                        &mut primary_key,
+                        new_fields_ids_map,
+                        &mut available_docids,
+                        &mut bytes,
+                        &docids_version_offsets,
+                        IndexDocumentsMethod::ReplaceDocuments,
+                        shards,
+                        payload,
+                        on_missing_document,
+                    )
+                }
+                Payload::Update { payload, on_missing_document } => {
+                    extract_addition_payload_changes(
+                        indexer,
+                        index,
+                        rtxn,
+                        primary_key_from_op,
+                        &mut primary_key,
+                        new_fields_ids_map,
+                        &mut available_docids,
+                        &mut bytes,
+                        &docids_version_offsets,
+                        IndexDocumentsMethod::UpdateDocuments,
+                        shards,
+                        payload,
+                        on_missing_document,
+                    )
+                }
                 Payload::Deletion(to_delete) => extract_deletion_payload_changes(
                     index,
                     rtxn,
@@ -180,6 +194,7 @@ fn extract_addition_payload_changes<'r, 'pl: 'r>(
     method: IndexDocumentsMethod,
     shards: Option<&Shards>,
     payload: &'pl [u8],
+    on_missing_document: MissingDocumentPolicy,
 ) -> Result<hashbrown::HashMap<&'pl str, PayloadOperations<'pl>>> {
     use IndexDocumentsMethod::{ReplaceDocuments, UpdateDocuments};
 
@@ -271,6 +286,10 @@ fn extract_addition_payload_changes<'r, 'pl: 'r>(
 
                             match method {
                                 ReplaceDocuments => {
+                                    if matches!(on_missing_document, MissingDocumentPolicy::Skip) {
+                                        continue;
+                                    }
+
                                     entry.insert(PayloadOperations::new_replacement(
                                         docid,
                                         true, // is new
@@ -278,6 +297,10 @@ fn extract_addition_payload_changes<'r, 'pl: 'r>(
                                     ));
                                 }
                                 UpdateDocuments => {
+                                    if matches!(on_missing_document, MissingDocumentPolicy::Skip) {
+                                        continue;
+                                    }
+
                                     entry.insert(PayloadOperations::new_update(
                                         docid,
                                         true, // is new
@@ -297,6 +320,12 @@ fn extract_addition_payload_changes<'r, 'pl: 'r>(
                 },
                 Entry::Vacant(entry) => match method {
                     ReplaceDocuments => {
+                        if payload_operations.is_new
+                            && matches!(on_missing_document, MissingDocumentPolicy::Skip)
+                        {
+                            continue;
+                        }
+
                         entry.insert(PayloadOperations::new_replacement(
                             payload_operations.docid,
                             payload_operations.is_new,
@@ -304,6 +333,12 @@ fn extract_addition_payload_changes<'r, 'pl: 'r>(
                         ));
                     }
                     UpdateDocuments => {
+                        if payload_operations.is_new
+                            && matches!(on_missing_document, MissingDocumentPolicy::Skip)
+                        {
+                            continue;
+                        }
+
                         entry.insert(PayloadOperations::new_update(
                             payload_operations.docid,
                             payload_operations.is_new,
@@ -448,8 +483,8 @@ pub struct DocumentOperationChanges<'pl> {
 }
 
 pub enum Payload<'pl> {
-    Replace(&'pl [u8]),
-    Update(&'pl [u8]),
+    Replace { payload: &'pl [u8], on_missing_document: MissingDocumentPolicy },
+    Update { payload: &'pl [u8], on_missing_document: MissingDocumentPolicy },
     Deletion(&'pl [&'pl str]),
 }
 

crates/milli/tests/search/facet_distribution.rs

@@ -47,7 +47,7 @@ fn test_facet_distribution_with_no_facet_values() {
     let documents = mmap_from_objects(vec![doc1, doc2]);
 
     // index documents
-    indexer.replace_documents(&documents).unwrap();
+    indexer.replace_documents(&documents, Default::default()).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, _operation_stats, primary_key) = indexer

crates/milli/tests/search/mod.rs

@@ -85,7 +85,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     let payload = unsafe { memmap2::Mmap::map(&file).unwrap() };
 
     // index documents
-    indexer.replace_documents(&payload).unwrap();
+    indexer.replace_documents(&payload, Default::default()).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, operation_stats, primary_key) = indexer

crates/milli/tests/search/query_criteria.rs

@@ -319,7 +319,7 @@ fn criteria_ascdesc() {
     file.sync_all().unwrap();
 
     let payload = unsafe { memmap2::Mmap::map(&file).unwrap() };
-    indexer.replace_documents(&payload).unwrap();
+    indexer.replace_documents(&payload, Default::default()).unwrap();
     let (document_changes, _operation_stats, primary_key) = indexer
         .into_changes(
             &indexer_alloc,

crates/milli/tests/search/typo_tolerance.rs

@@ -126,7 +126,7 @@ fn test_typo_disabled_on_word() {
     let embedders = RuntimeEmbedders::default();
     let mut indexer = indexer::DocumentOperation::new();
 
-    indexer.replace_documents(&documents).unwrap();
+    indexer.replace_documents(&documents, Default::default()).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, _operation_stats, primary_key) = indexer